research paper about memory

Journal of Experimental Psychology: Learning, Memory, and Cognition

• Read this journal
• Read free articles
• Journal snapshot
• Advertising information

Journal scope statement

The Journal of Experimental Psychology: Learning, Memory, and Cognition ® publishes original experimental and theoretical research on human cognition, with a special emphasis on learning, memory, language, and higher cognition.

The journal publishes impactful articles of any length, including literature reviews, meta-analyses, replications, theoretical notes, and commentaries on previously published works in the Journal of Experimental Psychology: Learning, Memory, and Cognition .

The journal places a high emphasis on methodological soundness and analytic rigor, as well as on the reproducibility and replicability of research.

Disclaimer: APA and the editors of the Journal of Experimental Psychology: Learning, Memory, and Cognition assume no responsibility for statements and opinions advanced by the authors of its articles.

Equity, diversity, and inclusion

Journal of Experimental Psychology: Learning, Memory, and Cognition supports equity, diversity, and inclusion (EDI) in its practices. More information on these initiatives is available under EDI Efforts .

Open Science

The APA Journals Program is committed to publishing transparent, rigorous research; improving reproducibility in science; and aiding research discovery. Open science practices vary per editor discretion. View the initiatives implemented by this journal .

Editor's Choice

Each issue of the Journal of Experimental Psychology: Learning, Memory, and Cognition will honor one accepted manuscript per issue by selecting it as an “ Editor’s Choice ” paper. Selection is based on the discretion of the editor if the paper offers an unusually large potential impact to the field and/or elevates an important future direction for science.

Author and editor spotlights

Explore journal highlights : free article summaries, editor interviews and editorials, journal awards, mentorship opportunities, and more.

Prior to submission, please carefully read and follow the submission guidelines detailed below. Manuscripts that do not conform to the submission guidelines may be returned without review.

To submit to the editorial office of Aaron S. Benjamin, please submit manuscripts electronically through the Manuscript Submission Portal in Microsoft Word or Open Office format.

Prepare manuscripts according to the Publication Manual of the American Psychological Association using the 7 th edition. Manuscripts may be copyedited for bias-free language (see Chapter 5 of the Publication Manual ).

Submit Manuscript

Aaron S. Benjamin University of Illinois at Urbana–Champaign 827 Psychology Bldg. 603 E. Daniel Street M/C 716 Champaign, IL 61820

Acceptable formats include: Microsoft Word Document (.doc or .docx) and LaTeX (.tex) with Portable Document Format (.pdf) of the manuscript file.

When submitting a manuscript electronically, authors should type their abstract directly into the Abstract text box in the submission form or save their abstract as a text file and copy and paste it from that file. Any special formatting (from a Word document, for example) will be lost when the form is submitted.

The editorial policy of the journal encompasses integrative articles containing multiple experiments as well as articles reporting single experiments. The journal also publishes commentaries and theoretical notes. Please note that theoretical notes are limited to a maximum of 25 pages of text. Commentaries on articles should be at maximum half the length of the target article.

JEP:LMC  is now using a software system to screen submitted content for similarity with other published content. The system compares the initial version of each submitted manuscript against a database of 40+ million scholarly documents, as well as content appearing on the open web. This allows APA to check submissions for potential overlap with material previously published in scholarly journals (e.g., lifted or republished material).

Direct replications

The journal publishes direct replications. Submissions should include “A Replication of XX Study” in the subtitle of the manuscript as well as in the abstract.

• Registered Reports

Journal of Experimental Psychology: Learning, Memory, and Cognition  ( JEP:LMC ) accepts Registered Reports (RR) as an article type. To learn more about RR, please consult the general guidelines and information at the Center for Open Science and at Peer Community In . We are a PCI-RR-interested journal, which means that articles that pass review at PCI can enter the review process at JEP:LMC .

Registered Reports at JEP:LMC must meet the usual stringent standards for scientific rigor and validity. In addition, as with any article at JEP:LMC , articles are expected to advance theory about basic cognitive functions.

Access the flowchart on editorial decision-making about Registered Reports at JEP:LMC . Please download it and consider whether your proposal meets requirements prior to submission.

Registered Reports submissions can propose novel research or replications of prior work that was published in JEP:LMC or elsewhere.

Proposals can include preliminary data. Multi-experiment proposals in which some experiments are completed are others are proposed are often an appealing approach to a complicated question.

Publications may be allowed to include exploratory analyses not included in the proposal. These analyses should be clearly labeled as exploratory and their inclusion will be informed in part by the outcome of Stage 2 peer review.

Power should be high and should allow fair comparison of competing hypotheses. Sample size can take into account the costs of sampling (for example, with hard-to-recruit populations).

There is no deadline for Stage 2 submission following Stage 1 acceptance, but the authors must accept some risk that scientific findings that arise in the interim may influence Stage 2 review. No Stage 2 paper will be rejected simply because the core result has been published by others in that interim period.

Cover letter

Your cover letter must include the following information:

• Contact information for each author, including valid email address, affiliation, mailing address, as well as phone and fax numbers
• A statement that your manuscript is original, not previously published, and not under concurrent consideration elsewhere
• If your cover letter does not contain this information, you will receive a written request from JEP: LMC asking for a revised cover letter.

When submitting your manuscript, please include your cover letter in its entirety by copying it to the text box provided in the Manuscript Submission Portal . Unfortunately, the portal cannot currently accept formatted cover letters. Even if you send a formatted version of your letter to the Journal, we must also have a text version of your cover letter.

If you are submitting a revision, feel free to email a formatted version of your cover letter to the editorial office , though you must still also send a text version of your cover letter through the portal as you're uploading your revision.

Manuscript submission acknowledgment

Once authors have submitted their manuscript through the submission portal, an email acknowledging receipt of the manuscript will be sent to the corresponding author. If this acknowledgment email does not reach the corresponding author within a few days of submitting the manuscript, please contact the editorial office .

Anonymous review

Anonymous review is optional. If an author wants anonymous review, a “Masked” article type must be selected upon submission, and the request should be included in the author's cover letter.

The manuscript receiving anonymous review should be formatted as follows:

Make sure that the manuscript itself contains no clues to the authors' identity, including grant numbers, names of institutions providing IRB approval, self-citations, and links to online repositories for data, materials, code, or preregistrations (e.g., Create a View-only Link for a Project). All author-identifying information should be removed from the manuscript, including authors' names and affiliations on the title page, footnotes, and author notes. The properties of the file should also not reveal the authors' names.

The authors' contact information should instead be included in the cover letter, which is not seen by the reviewers.

If your manuscript was mask reviewed, please ensure that the final version for production includes a byline and full author note for typesetting.

Related Journals of Experimental Psychology

For the other JEP journals, authors should submit manuscripts according to the manuscript submission guidelines for each individual JEP journal:

• Journal of Experimental Psychology: Animal Learning and Cognition
• Journal of Experimental Psychology: Applied
• Journal of Experimental Psychology: General
• Journal of Experimental Psychology: Human Perception and Performance

When one of the editors believes a manuscript is clearly more appropriate for an alternative APA journal, the editor may redirect the manuscript with the approval of the author.

Manuscript preparation

Prepare manuscripts according to the Publication Manual of the American Psychological Association  using the 7th edition. Manuscripts may be copyedited for bias-free language (see Chapter 5 of the Publication Manual ).

Review APA's Journal Manuscript Preparation Guidelines before submitting your article.

Double-space all copy. Other formatting instructions, as well as instructions on preparing tables, figures, references, metrics, and abstracts, appear in the Manual . Additional guidance on APA Style is available on the APA Style website .

Below are additional instructions regarding the preparation of display equations, computer code, and tables.

Display equations

We strongly encourage you to use MathType (third-party software) or Equation Editor 3.0 (built into pre-2007 versions of Word) to construct your equations, rather than the equation support that is built into Word 2007 and Word 2010. Equations composed with the built-in Word 2007/Word 2010 equation support are converted to low-resolution graphics when they enter the production process and must be rekeyed by the typesetter, which may introduce errors.

To construct your equations with MathType or Equation Editor 3.0:

• Go to the Text section of the Insert tab and select Object.
• Select MathType or Equation Editor 3.0 in the drop-down menu.

If you have an equation that has already been produced using Microsoft Word 2007 or 2010 and you have access to the full version of MathType 6.5 or later, you can convert this equation to MathType by clicking on MathType Insert Equation. Copy the equation from Microsoft Word and paste it into the MathType box. Verify that your equation is correct, click File, and then click Update. Your equation has now been inserted into your Word file as a MathType Equation.

Use Equation Editor 3.0 or MathType only for equations or for formulas that cannot be produced as Word text using the Times or Symbol font.

Computer code

Because altering computer code in any way (e.g., indents, line spacing, line breaks, page breaks) during the typesetting process could alter its meaning, we treat computer code differently from the rest of your article in our production process. To that end, we request separate files for computer code.

In online supplemental material

We request that runnable source code be included as supplemental material to the article. For more information, visit Supplementing Your Article With Online Material .

In the text of the article

If you would like to include code in the text of your published manuscript, please submit a separate file with your code exactly as you want it to appear, using Courier New font with a type size of 8 points. We will make an image of each segment of code in your article that exceeds 40 characters in length. (Shorter snippets of code that appear in text will be typeset in Courier New and run in with the rest of the text.) If an appendix contains a mix of code and explanatory text, please submit a file that contains the entire appendix, with the code keyed in 8-point Courier New.

Use Word's insert table function when you create tables. Using spaces or tabs in your table will create problems when the table is typeset and may result in errors.

LaTex files

LaTex files (.tex) should be uploaded with all other files such as BibTeX Generated Bibliography File (.bbl) or Bibliography Document (.bib) together in a compressed ZIP file folder for the manuscript submission process. In addition, a Portable Document Format (.pdf) of the manuscript file must be uploaded for the peer-review process.

Academic writing and English language editing services

Authors who feel that their manuscript may benefit from additional academic writing or language editing support prior to submission are encouraged to seek out such services at their host institutions, engage with colleagues and subject matter experts, and/or consider several vendors that offer discounts to APA authors .

Please note that APA does not endorse or take responsibility for the service providers listed. It is strictly a referral service.

Use of such service is not mandatory for publication in an APA journal. Use of one or more of these services does not guarantee selection for peer review, manuscript acceptance, or preference for publication in any APA journal.

Submitting supplemental materials

APA can place supplemental materials online, available via the published article in the PsycArticles ® database. Please see Supplementing Your Article With Online Material for more details.

Author contribution statements using CRediT

The APA Publication Manual (7th ed.) stipulates that “authorship encompasses…not only persons who do the writing but also those who have made substantial scientific contributions to a study.” In the spirit of transparency and openness, Behavioral Neuroscience has adopted the Contributor Roles Taxonomy (CRediT) to describe each author's individual contributions to the work. CRediT offers authors the opportunity to share an accurate and detailed description of their diverse contributions to a manuscript.

Submitting authors will be asked to identify the contributions of all authors at initial submission according to this taxonomy. If the manuscript is accepted for publication, the CRediT designations will be published as an Author Contributions Statement in the author note of the final article. All authors should have reviewed and agreed to their individual contribution(s) before submission.

CRediT includes 14 contributor roles, as described below:

• Conceptualization: Ideas; formulation or evolution of overarching research goals and aims.
• Data curation: Management activities to annotate (produce metadata), scrub data and maintain research data (including software code, where it is necessary for interpreting the data itself) for initial use and later reuse.
• Formal analysis: Application of statistical, mathematical, computational, or other formal techniques to analyze or synthesize study data.
• Funding acquisition: Acquisition of the financial support for the project leading to this publication.
• Investigation: Conducting a research and investigation process, specifically performing the experiments, or data/evidence collection.
• Methodology: Development or design of methodology; creation of models.
• Project administration: Management and coordination responsibility for the research activity planning and execution.
• Resources: Provision of study materials, reagents, materials, patients, laboratory samples, animals, instrumentation, computing resources, or other analysis tools.
• Software: Programming, software development; designing computer programs; implementation of the computer code and supporting algorithms; testing of existing code components.
• Supervision: Oversight and leadership responsibility for the research activity planning and execution, including mentorship external to the core team.
• Validation: Verification, whether as a part of the activity or separate, of the overall replication/reproducibility of results/experiments and other research outputs.
• Visualization: Preparation, creation and/or presentation of the published work, specifically visualization/data presentation.
• Writing—original draft: Preparation, creation and/or presentation of the published work, specifically writing the initial draft (including substantive translation).
• Writing—review and editing: Preparation, creation and/or presentation of the published work by those from the original research group, specifically critical review, commentary or revision—including pre- or post-publication stages.

Authors can claim credit for more than one contributor role, and the same role can be attributed to more than one author.

Abstract and keywords

All manuscripts must include an abstract containing a maximum of 250 words typed on a separate page. After the abstract, please supply up to five keywords or brief phrases.

List references in alphabetical order. Each listed reference should be cited in text, and each text citation should be listed in the references section.

Examples of basic reference formats:

Journal article

McCauley, S. M., & Christiansen, M. H. (2019). Language learning as language use: A cross-linguistic model of child language development. Psychological Review , 126 (1), 1–51. https://doi.org/10.1037/rev0000126

Authored book

Brown, L. S. (2018). Feminist therapy (2nd ed.). American Psychological Association. https://doi.org/10.1037/0000092-000

Chapter in an edited book

Balsam, K. F., Martell, C. R., Jones. K. P., & Safren, S. A. (2019). Affirmative cognitive behavior therapy with sexual and gender minority people. In G. Y. Iwamasa & P. A. Hays (Eds.), Culturally responsive cognitive behavior therapy: Practice and supervision (2nd ed., pp. 287–314). American Psychological Association. https://doi.org/10.1037/0000119-012

Data set citation

Alegria, M., Jackson, J. S., Kessler, R. C., & Takeuchi, D. (2016). Collaborative Psychiatric Epidemiology Surveys (CPES), 2001–2003 [Data set]. Inter-university Consortium for Political and Social Research. https://doi.org/10.3886/ICPSR20240.v8

Software/Code citation

Viechtbauer, W. (2010). Conducting meta-analyses in R with the metafor package.  Journal of Statistical Software , 36(3), 1–48. https://www.jstatsoft.org/v36/i03/

Wickham, H. et al., (2019). Welcome to the tidyverse. Journal of Open Source Software, 4 (43), 1686, https://doi.org/10.21105/joss.01686

All data, program code, and other methods must be cited in the text and listed in the references section.

Preferred formats for graphics files are TIFF and JPG, and preferred format for vector-based files is EPS. Graphics downloaded or saved from web pages are not acceptable for publication. Multipanel figures (i.e., figures with parts labeled a, b, c, d, etc.) should be assembled into one file. When possible, please place symbol legends below the figure instead of to the side.

• All color line art and halftones: 300 DPI
• Black and white line tone and gray halftone images: 600 DPI

Line weights

• Color (RGB, CMYK) images: 2 pixels
• Grayscale images: 4 pixels
• Stroke weight: 0.5 points

APA offers authors the option to publish their figures online in color without the costs associated with print publication of color figures.

The same caption will appear on both the online (color) and print (black and white) versions. To ensure that the figure can be understood in both formats, authors should add alternative wording (e.g., “the red (dark gray) bars represent”) as needed.

For authors who prefer their figures to be published in color both in print and online, original color figures can be printed in color at the editor's and publisher's discretion provided the author agrees to pay:

• $900 for one figure
• An additional $600 for the second figure
• An additional $450 for each subsequent figure

Permissions

Authors of accepted papers must obtain and provide to the editor on final acceptance all necessary permissions to reproduce in print and electronic form any copyrighted work, including test materials (or portions thereof), photographs, and other graphic images (including those used as stimuli in experiments).

On advice of counsel, APA may decline to publish any image whose copyright status is unknown.

• Download Permissions Alert Form (PDF, 13KB)

Reporting standards

Journal article reporting standards.

Authors should consider the APA Style Journal Article Reporting Standards (JARS) for a helpful resource for reporting data and the outcomes of inferential statistical tests. The standards offer ways to improve transparency in reporting to ensure that readers have the information necessary to evaluate the quality of the research and to facilitate collaboration and replication.

• recommend the division of hypotheses, analyses, and conclusions into primary, secondary, and exploratory groupings to allow for a full understanding of quantitative analyses presented in a manuscript and to enhance reproducibility;
• offer modules for authors reporting on replications, clinical trials, longitudinal studies, and observational studies, as well as the analytic methods of structural equation modeling and Bayesian analysis; and
• include guidelines on reporting of study preregistration (including making protocols public); participant characteristics (including demographic characteristics); inclusion and exclusion criteria; psychometric characteristics of outcome measures and other variables; and planned data diagnostics and analytic strategy.

The guidelines focus on transparency in methods reporting, recommending descriptions of how the researcher's own perspective affected the study, as well as the contexts in which the research and analysis took place.

Transparency and openness

Empirical research, including meta-analyses, submitted to the Journal of Experimental Psychology: Learning, Memory, and Cognition must meet Level 1 (Disclosure) for all eight aspects of research planning and reporting as well as Level 2 (Requirement) for citation; data, materials, and code transparency; and study and analysis plan preregistration. Authors should include a subsection in the method section titled “Transparency and openness.” This subsection should detail the efforts the authors have made to comply with the TOP guidelines.

For example:

• We report how we determined our sample size, all data exclusions (if any), all manipulations, and all measures in the study, and the study follows JARS (Appelbaum, et al., 2018). All data, analysis code, and research materials are available at [stable link to repository]. Data were analyzed using R, version 4.0.0 (R Core Team, 2020) and the package ggplot , version 3.2.1 (Wickham, 2016). This study’s design and its analysis were not pre-registered.

Data, materials, and code

Authors must state whether data, code, and study materials are posted to a trusted repository and, if so, how to access them, including their location and any limitations on use. If they cannot be made available, authors must state the legal or ethical reasons why they are not available. Trusted repositories adhere to policies that make data discoverable, accessible, usable, and preserved for the long term. Trusted repositories also assign unique and persistent identifiers.. Recommended repositories include APA’s repository on the Open Science Framework (OSF), or authors can access a full list of other recommended repositories .

In a subsection titled "Transparency and Openness" at the end of the method section, specify whether and where the data and material will be available or note the legal or ethical reasons for not doing so. For submissions with quantitative or simulation analytic methods, state whether the study analysis code (e.g., scripts for generating stimuli, conducting simulations, or performing data analyses) is posted to a trusted repository, and, if so, where to access it or the legal or ethical reason why it is not available.

• All data have been made publicly available at the [trusted repository name] and can be accessed at [persistent URL or DOI].
• Materials and analysis code for this study are not available.
• The code behind this analysis/simulation has been made publicly available at the [trusted repository name] and can be accessed at [persistent URL or DOI].

Preregistration of studies and analysis plans

Preregistration of studies and specific hypotheses can be a useful tool for making strong theoretical claims. Likewise, preregistration of analysis plans can be useful for distinguishing confirmatory and exploratory analyses. Investigators are encouraged to preregister their studies and analysis plans prior to conducting the research via a publicly accessible registry system (e.g., OSF , ClinicalTrials.gov, or other trial registries in the WHO Registry Network).

There are many available templates; for example, APA, the British Psychological Society, and the German Psychological Society partnered with the Leibniz Institute for Psychology and Center for Open Science to create Preregistration Standards for Quantitative Research in Psychology (Bosnjak et al., 2022).

Articles must state whether or not any work was preregistered and, if so, how to access the preregistration. Preregistrations must be available to reviewers; authors may submit a masked copy via stable link or supplemental material. Links in the method section should be replaced with an identifiable copy on acceptance.

• This study’s design was preregistered; see [STABLE LINK OR DOI].
• This study’s design and hypotheses were preregistered; see [STABLE LINK OR DOI].
• This study’s analysis plan was preregistered; see [STABLE LINK OR DOI].
• This study was not preregistered.

Open science badges

Articles are eligible for open science badges recognizing publicly available data, materials, and/or preregistration plans and analyses. These badges are awarded on a self-disclosure basis.

At submission, authors must confirm that criteria have been fulfilled in a signed badge disclosure form (PDF, 33KB) that must be submitted as supplemental material. If all criteria are met as confirmed by the editor, the form will then be published with the article as supplemental material.

Authors should also note their eligibility for the badge(s) in the cover letter.

For all badges, items must be made available on an open-access repository with a persistent identifier in a format that is time-stamped, immutable, and permanent. For the preregistered badge, this is an institutional registration system.

Data and materials must be made available under an open license allowing others to copy, share, and use the data, with attribution and copyright as applicable.

Available badges are:

Note that it may not be possible to preregister a study or to share data and materials. Applying for open science badges is optional.

Publication policies

For full details on publication policies, including use of Artificial Intelligence tools, please see APA Publishing Policies .

APA policy prohibits an author from submitting the same manuscript for concurrent consideration by two or more publications.

See also APA Journals ® Internet Posting Guidelines .

APA requires authors to reveal any possible conflict of interest in the conduct and reporting of research (e.g., financial interests in a test or procedure, funding by pharmaceutical companies for drug research).

• Download Full Disclosure of Interests Form (PDF, 41KB)

In light of changing patterns of scientific knowledge dissemination, APA requires authors to provide information on prior dissemination of the data and narrative interpretations of the data/research appearing in the manuscript (e.g., if some or all were presented at a conference or meeting, posted on a listserv, shared on a website, including academic social networks like ResearchGate, etc.). This information (2–4 sentences) must be provided as part of the Author Note.

Ethical Principles

It is a violation of APA Ethical Principles to publish "as original data, data that have been previously published" (Standard 8.13).

In addition, APA Ethical Principles specify that "after research results are published, psychologists do not withhold the data on which their conclusions are based from other competent professionals who seek to verify the substantive claims through reanalysis and who intend to use such data only for that purpose, provided that the confidentiality of the participants can be protected and unless legal rights concerning proprietary data preclude their release" (Standard 8.14).

APA expects authors to adhere to these standards. Specifically, APA expects authors to have their data available throughout the editorial review process and for at least 5 years after the date of publication.

Authors are required to state in writing that they have complied with APA ethical standards in the treatment of their sample, human or animal, or to describe the details of treatment.

• Download Certification of Compliance With APA Ethical Principles Form (PDF, 26KB)

The APA Ethics Office provides the full Ethical Principles of Psychologists and Code of Conduct electronically on its website in HTML, PDF, and Word format. You may also request a copy by emailing or calling the APA Ethics Office (202-336-5930). You may also read "Ethical Principles," December 1992, American Psychologist , Vol. 47, pp. 1597–1611.

Other information

See APA’s Publishing Policies page for more information on publication policies, including information on author contributorship and responsibilities of authors, author name changes after publication, the use of generative artificial intelligence, funder information and conflict-of-interest disclosures, duplicate publication, data publication and reuse, and preprints.

Visit the Journals Publishing Resource Center for more resources for writing, reviewing, and editing articles for publishing in APA journals.

Incoming editor

Jeffrey Starns, PhD University of Massachusetts Amherst

Outgoing editor

Aaron S. Benjamin, PhD University of Illinois at Urbana–Champaign, United States

Associate editors

Erik M. Altmann, PhD Michigan State University, United States

Tanjeem Azad, PhD Kwantlen Polytechnic University, Canada

Julie M. Bugg, PhD Washington University in St. Louis, United States

Heather J. Ferguson, PhD University of Kent, United Kingdom

Andrew Hollingworth, PhD University of Iowa, United States

Keith Hutchison, PhD Montana State University, United States

Klaus Oberauer, PhD University of Zürich, Switzerland

Matthew G. Rhodes, PhD Colorado State University, United States

Jörg Rieskamp, PhD University of Basel, Switzerland

Evan Risko, PhD University of Waterloo, Canada

Adam Sanborn, PhD University of Warwick, United Kingdom

L. Robert Slevc, PhD University of Maryland, College Park, United States

Jonathan Garrett Tullis, PhD University of Arizona, United States

Nash Unsworth, PhD University of Oregon, United States

Ronaldo Vigo, PhD Ohio University, United States

Tessa C. Warren, PhD University of Pittsburgh, United States

Duane G. Watson, PhD Vanderbilt University, United States

Jennifer Wiley, PhD University of Illinois at Chicago, United States

Incoming consulting editors

Laurel Brehm, PhD University of California, Santa Barbara, United States

Robert Davies, PhD Lancaster University, United Kingdom

Lena Jäger, PhD University of Zurich, Switzerland

Lynn Lohnas, PhD Syracuse University, United States

Adam Osth, PhD University of Melbourne, Australia

Andrea Patalano, PhD Wesleyan University, United States

Kimele Persaud, PhD Rutgers University Newark, United States

Pooja Sydney, PhD University of Kentucky, United States

Aaron Veldre, PhD University of Technology Sydney, Australia

Outgoing consulting editors

Sally Andrews, PhD University of Sydney, Australia

Julie E. Boland, PhD University of Michigan, United States

Laura Carlson, PhD University of Notre Dame, United States

Sven Mattys, PhD The University of York, United Kingdom

Antje S. Meyer, PhD Max Planck Institute for Psycholinguistics & Radboud University Nijmegen, Netherlands

Adrian Staub, PhD University of Massachusetts Amherst, United States

Anna M. Woollams, PhD University of Manchester, United Kingdom

Consulting editors

Jeanette Altarriba, PhD University at Albany & State University of New York, United States

Blair C. Armstrong, PhD University of Toronto, Canada

Jason Arndt, PhD Middlebury College, United States

Kate Arrington, PhD Lehigh University, United States

Hunter Ball, PhD University of Texas at Arlington, United States

Karl-Heinz Thomas Bäuml, PhD Regensburg University, Germany

C. J. Brainerd, PhD Cornell University, United States

David W. Braithwaite, PhD Florida State University, United States

Gene A. Brewer, PhD Arizona State University, United States

Thomas Busey, PhD Indiana University, Bloomington, United States

Andrew C. Butler, PhD Washington University in St. Louis, United States

Valérie Camos, PhD Université de Fribourg, Switzerland

Shana Carpenter, PhD Iowa State University, United States

Anne M. Cleary, PhD Colorado State University, United States

Gabriel I. Cook, PhD Claremont McKenna College, United States

Sarah Creel, PhD University of California, San Diego, United States

Matthew J. C. Crump, PhD Brooklyn College of CUNY, United States

Peter F. Delaney, PhD University of North Carolina at Greensboro, United States

Gary S. Dell, PhD University of Illinois at Urbana–Champaign, United States

Gesine Dreisbach, PhD Universität Regensburg, Germany

Nicolas Dumay, PhD University of Exeter, United Kingdom

John Dunlosky, PhD Kent State University, United States

Lisa K. Fazio, PhD Vanderbilt University, United States

Laurie Beth Feldman, PhD State University of New York at Albany, United States

Myra A. Fernandes, PhD University of Waterloo, Canada

Ruth Filik, PhD University of Nottingham, United Kingdom

Bridgid Finn, PhD Educational Testing Service, United States

Rico Fischer, PhD University of Greifswald, Germany

Wendy S. Francis, PhD University of Texas at El Paso, United States

David A. Gallo, PhD University of Chicago, United States

Tamar H. Gollan, PhD University of California, San Diego, United States

Corentin Gonthier, PhD University of Rennes 2, France

James A. Hampton, PhD City University of London, United Kingdom

Deborah Hannula, PhD University of Wisconsin-Milwaukee

Alice F. Healy, PhD University of Colorado, Boulder, United States

William S. Horton, PhD Northwestern University, United States

Kathleen L. Hourihan, PhD Memorial University of Newfoundland, Canada

Yi Ting Huang, PhD University of Maryland College Park, United States

R. Reed Hunt, PhD University of Mississippi, United States

Edward Matthew Husband, PhD University of Oxford, United Kingdom

Helene Intraub, PhD University of Delaware, United States

Andrew F. Jarosz, PhD Mississippi State University, United States

Luis Jimenez, PhD Universidad de Santiago de Compostela, Spain

Michael J. Kane, PhD University of North Carolina at Greensboro, United States

Jeffrey D. Karpicke, PhD Purdue University, United States

David Kellen, PhD Syracuse University, United States

Jonathan W. Kelly, PhD Iowa State University, United States

Annette Kinder, PhD Freie Universitaet Berlin, Germany

Sachiko Kinoshita, PhD Macquarie University, Australia

Karl Christoph Klauer, PhD Albert-Ludwigs-Universität Freiburg, Germany

Iring Koch, PhD RWTH Aachen University, Germany

Agnieszka E. Konopka, PhD University of Aberdeen, United Kingdom

Mike E. Le Pelley, PhD UNSW Sydney, Australia

Robyn A. LeBoeuf, PhD Washington University in St. Louis, United States

Vanessa M. Loaiza, PhD University of Essex, United Kingdom

Stephen J. Lupker, PhD University of Western Ontario, Canada

Randi Martin, PhD Rice University, United States

John Paul Minda, PhD University of Western Ontario, Canada

Laura Morett University of Alabama, United States

Weimin Mou, PhD University of Alberta, Canada

Moshe Naveh-Benjamin, PhD University of Missouri, United States

James H. Neely, PhD University at Albany & State University of New York, United States

David E. Over, PhD Durham University, United Kingdom

Thorsten Pachur, PhD Max Planck Institute for Human Development, Germany

Manuel Perea, PhD Universitat de València, Spain

John Philbeck, PhD George Washington University, United States

Valerie F. Reyna, PhD Cornell University, United States

Henry L. Roediger III, PhD Washington University in St. Louis, United States

Nathan S. Rose, PhD University of Notre Dame, United States

Caren M. Rotello, PhD University of Massachusetts, United States

Jan Rummel, PhD Heidelberg University, Germany

Lili Sahakyan, PhD University of Illinois at Urbana–Champaign, United States

Elizabeth Roye Schotter, PhD University of South Florida, United States

Colleen M. Seifert, PhD University of Michigan, Ann Arbor, United States

Rebekah Smith, PhD University of Mississippi, United States

Benjamin C. Storm, PhD University of California, Santa Cruz, United States

Aimee M. Surprenant, PhD Memorial University of Newfoundland, Canada

Holly A. Taylor, PhD Tufts University, United States

Anjali Thapar, PhD Bryn Mawr College, United States

Catherine Thevenot, PhD University of Lausanne, Switzerland

Ayanna K. Thomas, PhD Tufts University, United States

Monika Undorf, PhD University of Mannheim, Germany

Mieke Verfaellie, PhD Boston University & VA Boston Healthcare System, United States

Timothy J. Vickery, PhD University of Delaware, United States

Michael S. Vitevitch, PhD University of Kansas, United States

Chris Wahlheim, PhD University of North Carolina at Greensboro, United States

Deanne Westerman, PhD Binghamton University & State University of New York, United States

Holger Wiese, PhD Durham University, United Kingdom

Naohide Yamamoto, PhD Queensland University of Technology, Australia

Andrew Yonelinas, PhD University of California Davis, United States

Editorial fellows

Abhilasha Kumar, PhD Bowdoin College, United States

Kimele Persaud, PhD Rutgers University, United States

Peer review coordinator

Kris Guin American Psychological Association

Abstracting and indexing services providing coverage of Journal of Experimental Psychology: Learning, Memory, and Cognition ®

• Academic OneFile
• Academic Search Alumni Edition
• Academic Search Complete
• Academic Search Elite
• Academic Search Index
• Academic Search Premier
• Advanced Placement Psychology Collection
• Biological Abstracts (BIOSIS)
• BIOSIS Previews
• Business Source Alumni Edition
• Business Source Complete
• Business Source Corporate Plus
• Business Source Elite
• Business Source Index
• Business Source Premier
• Cabell's Directory of Publishing Opportunities in Psychology
• Chartered Association of Business Schools (CABS) Academic Journal Guide
• Communication & Mass Media Complete
• Communication Source
• Current Abstracts
• Current Contents: Social & Behavioral Sciences
• EBSCO MegaFILE
• Education Abstracts
• Education Full Text
• Education Research Complete
• Education Source
• Educational Research Abstracts Online
• Educator's Reference Complete
• Ergonomics Abstracts
• ERIH (European Reference Index for the Humanities and Social Sciences)
• Expanded Academic ASAP
• General OneFile
• General Science Collection
• Health & Wellness Resource Center and Alternative Health Module
• Health Reference Center Academic
• Humanities and Social Sciences Index Retrospective
• IBZ / IBR (Internationale Bibliographie der Rezensionen Geistes- und Sozialwissenschaftlicher Literatur)
• InfoTrac Custom
• Journal Citations Report: Social Sciences Edition
• MLA International Bibliography
• NSA Collection
• OmniFile Full Text Mega
• Professional ProQuest Central
• ProQuest Central
• ProQuest Discovery
• ProQuest Platinum Periodicals
• ProQuest Psychology Journals
• ProQuest Research Library
• ProQuest Social Science Journals
• Psychology Collection
• RILM Abstracts of Music Literature
• Science Citation Index Expanded
• Social Sciences Abstracts
• Social Sciences Citation Index
• Social Sciences Full Text
• Social Sciences Index Retrospective
• Studies on Women and Gender Abstracts
• TOC Premier

Transparency and Openness Promotion

APA endorses the Transparency and Openness Promotion (TOP) Guidelines by a community working group in conjunction with the Center for Open Science ( Nosek et al. 2015 ). The TOP Guidelines cover eight fundamental aspects of research planning and reporting that can be followed by journals and authors at three levels of compliance.

• Level 1: Disclosure—The article must disclose whether or not the materials are available.
• Level 2: Requirement—The article must share materials when legally and ethically permitted (or disclose the legal and/or ethical restriction when not permitted).
• Level 3: Verification—A third party must verify that the standard is met.

Empirical research, including meta-analyses, submitted to the Journal of Experimental Psychology: Learning, Memory, and Cognition must, at a minimum, meet Level 1 (Disclosure) for all eight aspects of research planning and reporting as well as Level 2 (Requirement) for citation; data, code, and materials transparency; and study and analysis plan preregistration. Authors should include a subsection in their methods description titled “Transparency and openness.” This subsection should detail the efforts the authors have made to comply with the TOP guidelines.

The list below summarizes the minimal TOP requirements of the journal. Please refer to the Center for Open Science TOP guidelines for details, and contact the editor (Aaron S. Benjamin, PhD) with any further questions. APA recommends sharing data, materials, and code via  trusted repositories (e.g.,  APA’s repository  on the Open Science Framework (OSF)). Trusted repositories adhere to policies that make data discoverable, accessible, usable, and preserved for the long term. Trusted repositories also assign unique and persistent identifiers.

We encourage investigators to preregister their studies and to share protocols and analysis plans prior to conducting the research. There are many available preregistration forms (e.g., the APA Preregistration for Quantitative Research in Psychology template, ClinicalTrials.gov , or other preregistration templates available via OSF ). Completed preregistration forms should be posted on a publicly accessible registry system (e.g., OSF , ClinicalTrials.gov, or other trial registries in the WHO Registry Network).

A list of participating journals is also available from APA.

The following list presents the eight fundamental aspects of research planning and reporting, the TOP level required by the Journal of Experimental Psychology: Learning, Memory, and Cognition , and a brief description of the journal's policy.

• Citation: Level 2, Requirement—All data, program code, and other methods developed by others must be appropriately cited in the text and listed in the references section.
• Data Transparency: Level 2, Requirement—Article states whether the raw and/or processed data on which study conclusions are based are posted to a trusted repository and either how to access them or the legal or ethical reasons why they are not available.
• Analytic Methods (Code) Transparency: Level 2, Requirement—Article states whether computer code or syntax needed to reproduce analyses in an article is posted to a trusted repository and either how to access it or the legal or ethical reasons why it is not available.
• Research Materials Transparency: Level 2, Requirement—Article states whether materials described in the method section are posted to a trusted repository and either how to access them or the legal or ethical reasons why they are not available.
• Design and Analysis Transparency (Reporting Standards): Level 1, Disclosure—Authors should consider APA Style Journal Article Reporting Standards (JARS-Quant, JARS-Qual, and/or MARS) as a helpful resource for reporting data and the outcomes of inferential statistical tests.
• Study Preregistration: Level 2, Requirement—Article states whether the study design and (if applicable) hypotheses of any of the work reported was preregistered and, if so, how to access it. Access to the preregistration must be available at submission. Authors requesting masked review should submit a masked copy via stable link or supplemental material.
• Analysis Plan Preregistration: Level 2, Requirement—Article states whether any of the work reported preregistered an analysis plan and, if so, how to access it. Access to the preregistration must be available at submission. Authors requesting masked review should submit a masked copy via stable link or supplemental material.
• Replication: Level 3, Verification—The journal publishes replications and Registered Reports.

Other open science initiatives

• Open Science badges: Offered
• Public significance statements: Not offered
• Author contribution statements using CRediT: Required
• Registered Reports: Published
• Replications: Published

Explore open science at APA .

Inclusive study designs

Definitions and further details on inclusive study designs are available on the Journals EDI homepage .

Inclusive reporting standards

• Bias-free language and community-driven language guidelines (required)
• Data sharing and data availability statements (required)

More information on this journal’s reporting standards is listed under the submission guidelines tab .

Pathways to authorship and editorship

Editorial fellowships.

Editorial fellowships help early-career psychologists gain firsthand experience in scholarly publishing and editorial leadership roles. This journal offers an editorial fellowship program for early-career psychologists from historically excluded communities.

Other EDI offerings

Orcid reviewer recognition.

Open Research and Contributor ID (ORCID) Reviewer Recognition provides a visible and verifiable way for journals to publicly credit reviewers without compromising the confidentiality of the peer-review process. This journal has implemented the ORCID Reviewer Recognition feature in Editorial Manager, meaning that reviewers can be recognized for their contributions to the peer-review process.

Masked peer review

This journal offers masked peer review (where both the authors’ and reviewers’ identities are not known to the other). Research has shown that masked peer review can help reduce implicit bias against traditionally female names or early-career scientists with smaller publication records (Budden et al., 2008; Darling, 2015).

Announcement

• Call for editorial fellowships nominations

Editor Spotlight

• Read an interview with Editor Aaron S. Benjamin, PhD
• Editorial by Aaron S. Benjamin, PhD February 2019

From APA Journals Article Spotlight ®

• Qualitative changes in the processes supporting math performance across learning
• General valence asymmetry in similarity

Journal Alert

Sign up to receive email alerts on the latest content published.

Welcome! Thank you for subscribing.

Subscriptions and access

• Pricing and individual access
• APA PsycArticles database

Calls for Papers

Access options

• APA publishing resources
• Educators and students
• Editor resource center

APA Publishing Insider

APA Publishing Insider is a free monthly newsletter with tips on APA Style, open science initiatives, active calls for papers, research summaries, and more.

Social media

Contact Journals

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

Long-term memory effects on working memory updating development

Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Software, Validation, Visualization, Writing – original draft

* E-mail: [email protected]

Affiliation University of Urbino, Urbino, Italy

Roles Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Supervision, Validation, Writing – review & editing

Affiliation University of Pavia, Pavia, Italy

• Caterina Artuso, 
• Paola Palladino

• Published: May 31, 2019
• https://doi.org/10.1371/journal.pone.0217697
• Reader Comments

Long-term memory (LTM) associations appear as important to cognition as single memory contents. Previous studies on updating development have focused on cognitive processes and components, whereas our investigation examines how contents, associated with different LTM strength (strong or weak), might be differentially updated at different ages. To this end, we manipulated association strength of information given at encoding, in order to focus on updating pre-existing LTM associations; specifically, associations for letters. In particular, we controlled for letters usage frequency at the sub-lexical level. We used a task where we dissociated inhibition online (i.e., RTs for updating and controlling inhibition from the same set) and offline (i.e., RTs for controlling inhibition from previously updated sets). Mixed-effect analyses were conducted and showed a substantial behavioural cost when strong associations had to be dismantled online (i.e., longer RTs), compared to weak ones; here, in primary school age children. Interestingly, this effect was independent of age; in fact, children from 7–8 to 9–10 years were comparably sensitive to the strength of LTM associations in updating. However, older children were more effective in offline inhibitory control.

Citation: Artuso C, Palladino P (2019) Long-term memory effects on working memory updating development. PLoS ONE 14(5): e0217697. https://doi.org/10.1371/journal.pone.0217697

Editor: Burcu Arslan, Educational Testing Service, UNITED STATES

Received: November 26, 2018; Accepted: May 16, 2019; Published: May 31, 2019

Copyright: © 2019 Artuso, Palladino. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the paper and its Supporting Information files.

Funding: This work was supported by Blue Sky Research (BRS) 2017 Established Investigator awarded to PP. The funder played no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Working memory (WM) is a capacity limited system, able to maintain actively sets of representations useful in complex cognitive skills such as reading [ 1 , 2 ] or text comprehension [ 3 , 4 ]. WM performance improves substantially over childhood with linear increases [ 5 , 6 ]. These developmental improvements may be driven by increases in storage capacity [ 7 ], rehearsal strategies [ 8 ], or also updating processes [ 9 ].

In fact, given capacity limits and the continuous flow of information to be processed, it is important to explore a mechanism that potentially allows WM content to be updated constantly via maintenance of relevant information and inhibition of irrelevant information. Updating investigation is usually applied to memory contents [ 10 ]. However, usually, updating tasks are based on binding and unbinding processes between memory contents (e.g., [ 11 ]). Binding updating (but not content updating) is a more sensitive measure in accounting for performance in accuracy-based updating tasks [ 12 ]. In addition, the role of associative contextual bindings in episodic memory retrieval was also supported [ 13 ]. Overall, it appears that the monitoring of associative bindings between contents is a specific challenge for the updating process (see also [ 14 , 15 ]).

In the current paper, we aimed to study how updating of long-term memory (LTM) bindings (or LTM associations) develops in primary school children (in particular from third to fifth grade). First, we briefly review development of updating components and the role of LTM representations in WM tasks through childhood; in particular, lexical-semantic and sub-lexical representations. Next, we will focus on sub-lexical LTM representations and how these are updated specifically, introducing the aims of the current study.

Updating processes, components and development

Development of the WM updating function is a recent research topic that has arisen from adult studies and modelling research. In a recent developmental study, an accuracy-based updating task modelled after the one developed by [ 4 ] was administered to children [ 9 ]; here, they were able to differentiate between inhibition (i.e., ability to suppress same-lists intrusions) and proactive interference (PI) control (i.e., ability to suppress previous-lists intrusions). They showed that memory performance improves with age, together with development of inhibitory process efficiency. However, the most interesting finding here, is that these two components are relatively dissociable. The inhibition of information explained a considerable amount of variance, but of a similar percentage magnitude at ages 7, 11 and 15 years (42%, 49% and 46%, respectively); thus, its developmental contribution is less pronounced. On the other hand, the PI control component explained smaller amounts of variance across all ages, but especially at 7 years (25%), at 11 years (17%) and at 15 years (13%; [ 9 ]); thus its developmental role appeared more pronounced.

This two-component model of updating development is consistent with other models that emphasize additional features of updating and/or investigate alternative mechanisms [ 16 ]; here, the authors decomposed the updating process, individuating at least three components: retrieval (i.e., searching for a specific representation among many competing elements maintained in the region of direct access; see also [ 17 ]); transformation (i.e., modifying a representation maintained in WM); and the most distinctive component, item-removal (i.e., replacement of previously relevant content -now irrelevant- with new relevant information; [ 16 , 18 ]).

Within this theoretical framework, age-related differences through development, from 8 years to adulthood were found [ 19 ]. They found that only the retrieval component has age-related effects, with clear development from 8 years; no differences were observed for transformation or item-removal, despite their crucial role in updating.

LTM associations and WM development

The role of LTM associations in WM performance has been previously explored in order to understand how enduring properties of verbal material affects ongoing performance, mainly through simple WM tasks involving recall (e.g., [ 20 , 21 ]). The impact of informational organization in LTM on WM performance can be observed at different processing levels, e.g., lexical, sub-lexical and semantic.

In general, it has been shown that LTM associations interact with recall, facilitating the process; the more strongly items are associated in LTM, the more WM performance will benefit. That said, few studies have investigated the influence of lexical/semantic LTM representations on verbal WM performance in children, although previous research seems to suggest that effects are similar in children and adults (e.g., [ 22 , 23 , 24 ]).

Semantically-related information enhanced WM performance more than descriptive or unrelated information [ 22 ]. Similar lexico-semantic effects to adults across development were reported [ 23 ]. In an immediate serial recall task with words, they found replication of effects observed in adults, (e.g., lexicality, word frequency and imageability) from 6 to 22 years. These were accounted for by similar redintegration processes that would operate effectively on high frequency words because their phonological representations are more easily accessed by partial information. Accordingly, item frequency effects on recall are observed with the relevant item only, and occur at the time the individual item is retrieved/recalled (see also [ 20 , 21 , 25 ]).

How LTM lexical/semantic knowledge (such as lexicality and language familiarity effects) impacts on WM performance was examined by [ 24 ]. They compared children aged 5 and 9 years in tasks of immediate serial recall, finding evidence of the semantic-similarity effect in 5 year-olds. In fact, the specific organization of semantic LTM was found to enhance recall performance.

Overall, these studies have focused on WM recall tasks (i.e., entailing temporary maintenance of information in WM; [ 2 ]) and suggest that the more associated the information is, the better memory performance will be. In addition, studies suggest that developmental changes of the LTM system happens between the age of 5 and 11 years [ 24 ]; thus, interactions between LMT and WM recall are linked to developmental changes in WM capacity and efficiency [ 6 ]. In contrast, here, we focused on the interaction between LTM and updating; here, a complex WM function comprising not only temporary maintenance of information, but also inhibition and item-removal [ 9 , 16 , 18 ].

How LTM associations are updated

To the best of our knowledge, few studies have investigated the updating of LTM associations between verbal materials [ 14 , 26 ]. Indeed, updating can be distinguished from recall, as it allows memory focus to remain attuned to the most relevant information in any specific moment.

In an initial study, the strength of association between LTM stimuli was manipulated [ 26 ]; and how strength might modulate the updating process itself. Following the literature on the beneficial effects of highly-associated LTM information (e.g., [ 20 , 25 ]), Artuso and Palladino [ 26 ] investigated whether strong or weak associations were updated differently. Strength was represented by the frequency of sub-lexical associations between consonants. Association strength was manipulated at encoding, in order to observe how strong and weak associations were updated subsequently. Overall, it was shown that the stronger the LTM association, the longer latencies (i.e., to substitute information and to control for irrelevant information) were required. Therefore, a processing cost was found for updating; this is in direct opposition to recall, which is boosted by association strength [ 14 ].

In a further study, the association strength was manipulated at both encoding and updating, and added two conditions (i.e., strong associations that were updated to strong, and weak associations updated to strong), in order to gain a more complete view of accumulation and disruption of specific associations [ 14 ]. Here, the data supported the view that as pre-existing associations became stronger, they became harder to dismantle (i.e., longer RTs). In addition, when a strong association had to be recreated, this was usually enhanced (i.e., with shorter RTs from weak to strong association). The result was observed for both processing speed (inhibition process) and interference control (i.e., of a previously activated strong association). In particular, it was shown that inhibitory control requested was greater for items strongly associated, indicating, in turn, the long lasting of the pre-existing LTM association. Those experiments demonstrated clearly that associations from LTM modulate the updating process. In fact, these results suggested that, on the one hand, strong associations are dismantled and updated with greater difficulty (i.e., they require longer RTs), and on the other, that strong associations are activated more easily (i.e., requiring shorter RTs). This evidence supported the idea that the nature of updating rests in the interplay between dismantling and reconstructing bindings via different memory systems such as WM [ 11 , 24 ] and episodic LTM [ 13 ].

In the numerical domain, it was found that numerical similarity produces facilitation during updating of information. When the numbers involved in updating were near as far as concern numerical distance, or similar through sharing a digit, substitution occurred more quickly [ 27 , 28 ]. There, it was proposed that updating is a function of the overlapping features [ 29 ] between numbers to update and those stored in LTM; the greater the amount of overlap, the quicker the update will be, as both numbers share many (already activated) features. In summary, if, as well as inhibition [ 9 ], item-removal in LTM association is a distinctive updating component [ 16 ], it is important to investigate how the strength of this inter- item association retained in LTM affects WM processing (e.g., updating, [ 14 ]).

The current study

As previously described, studies on updating development have focused on processes and components [ 9 , 19 ], whereas our aim is to examine the associative effects of updating through development. In particular, given that LTM inter-item associations seem as important as single contents [ 14 ], we aimed to investigate whether associated information modulates updating performance in development.

Hence, we manipulated LTM associations for letters as they represent initial elements of learning and therefore, should be highly familiar to children, in addition to their established use in many studies on their role across cognition. In particular, we controlled for their frequency of use at the sub-lexical level. Broad evidence has shown recall accuracy is greater for words containing high frequency phoneme combinations in English (“phonotactic effect”, see [ 25 ]). Performance would likely benefit from use of stored phonotactic representations for familiar words to fill in incomplete traces prior to output. In contrast, for unfamiliar words, no stored representations are available to reconstruct partial traces, and this will lead to diminished accuracy at recall. In addition, recall is better for high phonotactic frequency of the language in WM. As fully described in [ 25 ] the “phonotactic effect” elicits better recall for ‘consonant-vowel-consonant’ non-words containing ‘consonant-vowel’ and ‘vowel-consonant’ combinations, common in the language’s native phonology, than for non-words containing low probability ‘consonant-vowel’ combinations. Such effect would reflect the influence of phonotactic knowledge about properties of that language [ 25 ].

With this aim, we administered an updating task previously used with both children [ 30 ] and adults [ 12 , 31 ], focused on active item-removal of information shown to be the most distinctive component of updating [ 14 , 16 ]; but see also [ 19 ]. In particular, this task allows collection of both online response times (RTs) during updating (i.e., dismantling of an item-set) and offline accuracy/RTs after updating of a memory set, in order to ensure updating effectiveness and inhibition of irrelevant information [ 31 ].

Therefore, we believe this task could include at least two different types of inhibition, that is online (i.e., inhibition within the same set) as well as offline (i.e., inhibition of the previously updated set of information). Thus, the specific object of our investigation is how information, associated with different strength in LTM, i.e., strongly or weakly, might be differently updated at various ages. To this end, we manipulated association strength of the information at encoding (but not updating), in order to focus on the specific function of dismantling pre-existing LTM associations rather than reconstruction of new associations. We hypothesize that, in line with adult studies (e.g., [ 23 ]), we should observe similar effects with children, as soon as LTM representations are strengthened and consolidated (i.e., with a behavioural cost for updating strongly associated information). In particular, we should observe an increase in online updating RTs when inhibiting and dismantling a strong pre-existing association (once encoded), and a decrease when dismantling a weak pre-existing association (once encoded). Accordingly, offline, we predict greater difficulty in inhibiting items from strong LTM associations, relative to weak ones).

Participants

The initial sample was of 90 children. At the end of the testing phase, we were informed from teachers that one child had received a diagnosis of learning disorder. We therefore decided to not include his data in the final sample. Thus, a sample of 89 children took part in the study. They did not present any specific learning, neurological or psychiatric disorder. Children were divided into two groups: 44 younger children (aged 7–8 years) and 45 older children (aged 9–10 years). These specific ages were chosen as they represent the most crucial steps for children to become more and more skilled in reading and writing, and access to meaning of written texts is more automatized. In addition, and in line with previous studies suggesting the relevance of the specific age range 5–11 years (e.g., [ 6 , 24 ]), we chose two central and crucial steps that are coherent with previous research and allow comparison. All children were Italian native speaker. See breakdown of participants’ characteristics in Table 1 .

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

Descriptive statistics (mean, standard deviations for accuracy rate and score range) for the Italian vocabulary and nonverbal reasoning test. SDs are in brackets.

https://doi.org/10.1371/journal.pone.0217697.t001

Children came from a public school located in Northern Italy, within an urban environment and mixed socio-economic background. All children had normal or corrected-to-normal vision. The study was conducted in accordance with the Ethical Standards laid down in the 1964 Declaration of Helsinki and the standard ethical procedures recommended by the Italian Psychological Association (AIP). The study was reviewed and approved by the IRB (ethical committee) of the University of Pavia/IUSS before the study began. Written informed parental consent (as well as oral informed child assent) was obtained prior to participating, according to the ethical norms in our university.

Children were administered two tasks to assess general cognitive abilities (see following method sections for full description). Descriptive statistics for the two general cognitive abilities administered to the two age groups are displayed in Table 1 . Analyses on the accuracy scores (independent sample t-tests) showed age-related differences in the vocabulary test, t( 87) = 2.09, p = .04, with older children better scoring than younger children, but no differences in the visuospatial reasoning test, t( 88) = 1.02, p = .31.

Materials and procedures

In order to verify that children’s general cognitive performance adhered to the average for their age, they were presented with two measures: a standardized Italian vocabulary test and a nonverbal reasoning test. In particular, the vocabulary can be taken as an index of crystallized intelligence, whereas the nonverbal reasoning test is held to measure fluid intelligence.

In addition, a computerized letter updating task was administered. The vocabulary test and the nonverbal reasoning test were administered in a classroom-based group session. The updating task was administered individually at school, in a quiet room. The group session lasted on average 15 minutes, and the updating task lasted about 20–25 minutes. The two sessions were non-consecutive, in order to avoid possible fatigue effects.

Italian vocabulary and nonverbal reasoning

The vocabulary and nonverbal reasoning subtests, taken from the Primary Mental Aptitude Battery [ 32 ] were presented to the whole class group during a school day; both have a four alternative multiple-choice structure. The vocabulary subtest has 30 items and the nonverbal reasoning subtest, 25 items. Participants had time constraints for both subtests; specifically, 5 minutes for the vocabulary and 6 minutes for the nonverbal reasoning.

Letter updating task

The task we used in the current paper was described in detail previously, in [ 14 ]. As in [ 14 ] the stimuli were sub-lexical units between two consonants of the Latin alphabet. The association was based on LTM consonant representation; that is, on the basis of their combined frequency in the spoken Italian language. We evaluated this from the lexicon of frequency of Italian spoken language [ 33 ], a corpus of about 490,000 lemmas collected in four main Italian cities, emerging from different subgroups of discourse. High and low frequency lemmas were selected. Low frequency ranged from 0 to 2 (i.e., lemmas with less than 3 occurrences in the corpus). High frequency lemmas had at least 3 occurrences in the corpus.

Next, we inferred strong and weak sub-lexical associations between consonants, based on the lemmas’ frequency. That said, we should note there is no specific frequency information for consonant couples, only for lemmas of the corpus. So, for example, from the lemma “ ardere” which is low frequency, we inferred the low frequency sub-lexical association “ rd ”. In addition, low frequency associations, typically, were in the middle of the lemma, whereas high frequency lemmas were at the beginning of the lemma. In addition, we checked the corpus to find occurrences of low frequency sub-lexical associations in different lemmas, in order to preclude their presence in high frequency lemmas. We included in the low frequency sub-lexical associations those one occurring in low frequency lemmas only.

We employed the following set of consonants: B C D F G H L N P R S T. Strong associations were: T-R, S-P, P-R, N-T, B-R, C-H, G-R, F-R. Weak ones were: F-L, S-N, G-H, P-S, G-L, R-D, N-D, L-T. Strong and weak associations between consonants were controlled in order to avoid obviously familiar or meaningful couplets. Each association was legal, and thus possible at the sub-lexical level of the Italian language [ 14 ].

As described in [ 14 ] and in order to avoid ceiling (i.e., with two items) or floor effects (i.e., with four items), we used memory sets composed of three letters (i.e., triplets), which have been established as being within average memory span [ 34 ]. Some letters were overrepresented relative to others, but we controlled for this bias by randomizing these across association strengths. Further, the position of the sub-lexical unit within the triplet (i.e., in positions 1/2 or 2/3 ) was randomized between trials. We did not control for potential position effects, as it was shown in a previous experiment that position did not interact with either updating or strength (see [ 14 ], Experiment 2)

The third letter of each triplet was another consonant, which was always unrelated to the other two. Specifically, the link between the sub-lexical unit and the third letter was always linguistically impossible in Italian (e.g., see example from Fig 1 where C-H is a strong association, and the link between H and B (H-B) is impossible in Italian). This was done in order to avoid any LTM (strong or weak) or some other meaningful way association between these letters.

After encoding the first triplet ( CHB ), participants had to maintain it actively in memory (pre-updating maintenance process: + + +). Next, they were instructed to update part of the association, that is, to remove the item C and substitute the G . Thus, the triplet they were now maintaining was GHB . Lastly, they had to maintain the recently updated triplet (post-updating maintenance process). At recognition, a single red probe was displayed: here, participants had to recognize if the probed item belonged to the most recent studied/updated item or not. In the example, a target probe was presented ( B ), to which they had to give a positive answer.

https://doi.org/10.1371/journal.pone.0217697.g001

Design and analyses

A three factor mixed design was implemented: Strength and Phase were within-participants factors, and Age group between-participants. The variable Strength had two levels: strong-to-weak and weak-to-weak. ‘Strong-to-weak’ represented associations between letters where the association was strong at encoding, but modified with a weak one upon updating (e.g., from C-H to G-H). Weak-to-weak represented associations between letters occurred where the association was weak at encoding and updated with another weak association (e.g., from P-S to P-R). For each trial, we considered two main phases of encoding (i.e., studying/encoding the initial triplet), and updating (i.e., partial into the triplet). Although the trial was constituted of four phases, only encoding and updating phases (i.e., phases that produce effects on RTs, see [ 31 ]) were entered into the analysis.

In addition, to make the task less predictable and ensure participants were engaged, we included several control trials (approximately 20% of the total number). Here, no updating occurred, and maintenance alone was required throughout the trials. These data were not included in further analyses, but were checked to ensure that all updating trials had longer RTs than controls ( p < .05 for each comparison; control vs. strong-to-weak, and weak-to-weak; [ 14 ]).

Procedure was described in detail previously [ 14 , 26 ]. The task was administered on a standard PC and consisted of four phase subject-paced trials, where participants pressed the spacebar to start each trial, and after each phase, in order to proceed with the task.

In each phase, triplets were always displayed in the centre of the screen. Each trial started with an encoding phase (Phase 1; see an example with letters in Fig 1 , where a strong-to-weak association is represented), where participants had to memorize the first triplet of consonants (e.g., C-H-B). A pre-updating maintenance phase followed (Phase 2), where three pluses were displayed; this indicated that the previously encoded triplet had to be actively maintained. Then, at updating (Phase 3), participants had to substitute the no-longer-relevant information (here, C) with newly relevant information (here, G). Concurrently, they needed to maintain previously relevant detail (here, H-B), thus, updating the triplet (i.e., from C-H-B to G-H-B). Finally, a post-updating maintenance (Phase 4) ended the sequence, to control for recency biases. See also Fig 1 .

Only one letter of the triplet had to be updated; this letter could be presented in any position of the triplet (i.e., left letter, right letter, or center). Position was balanced across trials, and only new consonants were presented across each phase. When a consonant did not change, a plus symbol was presented, in order to encourage active maintenance of previously encoded/memorized information.

At the end of each trial (Phase 5), participants were presented with a probe recognition task: a single red consonant was displayed in the centre of the screen. Here, they had to indicate whether this belonged to the most-recently studied triplet or not. They responded by pressing one of two keys on the keyboard; one (M for Yes ) for target probes requiring a positive answer (i.e., belonging to the final triplet of the trial); another one (Z for No ) for probes requiring a negative answer (i.e., not previously presented in the trial. For these, we included both lures i.e., (probes encoded in the trial, then substituted at updating step) and negative probes (i.e., probes not presented in that trial), mixed within the trial. Half the probes were targets (50%); the other half was equally shared between lures (25%) and negative probes (25%).

Afterwards, each participant was presented with a practice block of eight trials to familiarize themselves with the task. One hundred and twenty trials were then presented shared equally in four blocks. We recorded subject-paced RT at each of the four phases, in addition to probe recognition accuracy at Phase 5.

Results and discussion

Updating task: accuracy and data treatment.

Participants performed accurately on an average of 92.80% of trials. As expected, participants were very good in completing the task and very few errors were produced. Accuracy was analysed to verify adequate performance, but the main focus of the analysis was on RT. We ran a mixed 2 x 3 ANOVA, with Strength (weak-to-weak, strong-to-weak) as within-participants factor and Age Group (younger children, older children) as between-participants factor on mean accuracy rates of target, lures and negative responses. A main effect of Age Group reached significance, F (1, 87) = 8.38, p = .005. Accuracy rate was significantly lower in younger children (116/120 correct trials) than in older children (118/120 correct trials). Only subject-paced RTs for trials that ended with correct probe recognition were analysed. Trials with RTs of less than 150 ms, or exceeding a participant’s mean RT for each condition by more than three intra-individual standard deviations, were considered outliers, and therefore excluded from further analyses (3.92%).

In addition, updating measures (in particular, indexes of RT at the updating step), were highly inter-correlated, suggesting good reliability of the task. In particular, RTs for weak-to-weak associations were strongly correlated, r (89) = .84, p < .001, to RTs for strong-to-weak.

Overview of the statistical analyses

We used a mixed-effects model approach to test our hypotheses; the most important advantage of such models is that they allow simultaneous consideration of all factors that may contribute to understanding the structure of the data [ 35 ]. Raw RTS were logarithmically transformed to normalize them. These factors comprise not only the standard fixed-effects factors controlled by the experimenter (in our case, age group and strength) but also random-effects factors; that is, factors whose levels are drawn at random from a population (in our case, children). To test the effect of age group (younger children, older children) and strength (strong-to-weak, weak-to-weak) on the variables of online RT, and offline RT three mixed-models were used: one for online RT (with encoding and updating phases as additional factors), another one for RT of correctly detected target probes, and a third for RT of correctly rejected lures. See specific details in the subsections below.

All analyses were performed using the R software [ 36 ]; for generalized mixed-effect models, the R package lme4 was used [ 37 ]; and the lmer test package was used to obtain Type III ANOVA Tables. Results for each dependent variable are presented below. For planned comparisons, Tukey correction was used to control the Type I error rate.

Online RT analyses

A linear mixed-effects model was constructed with 3-way interactions between Age Group (younger children, older children), Strength (strong-to-weak, weak-to-weak), and Phase (encode, update). The model revealed a significant effect of Age Group, F (1, 87) = 8.11, p = .006. Overall, older children ( M = 2709.26 ms, SD = 67 ms) were faster than younger children ( M = 2960.22 ms, SD = 66 ms). Moreover, Strength affected the online processing, F (1, 261) = 5.71, p = .01; strong-to-weak associations ( M = 2898.36 ms, SD = 65 ms) were hardly updated than weak-to-weak ones ( M = 2768.30 ms, SD = 68 ms).

In addition, the Phase by Strength interaction reached significance, F (1, 261) = 7.18, p = .008. Post-hoc comparisons showed no differences at encode across associations, t( 261) = -0.21, p = .83; in contrast, at updating, strong-to-weak associations showed longer RTs compared to weak-to-weak associations, t( 261) = 3.59, p = .004, as shown in Fig 2 . No other interaction reached significance.

Plot dots represent mean predicted RTs (ms) and bars represent 95% CIs.

https://doi.org/10.1371/journal.pone.0217697.g002

Offline RT analyses: Target probes

A linear mixed-effects model was constructed with 2-way interactions between Age Group (younger children, older children) and Strength (strong-to-weak, weak-to-weak). The model revealed a significant effect of Strength, F (1, 87.353) = 11.13, p = .001. Indeed, we found significantly longer RTs for correct recognition of a target probe from strong-to-weak associations ( M = 2058.33 ms, SD = 58 ms), compared to weak-to-weak associations ( M = 1867.85 ms, SD = 43 ms). No other effect reached significance.

Offline RT analyses: Lures

First, we conducted a control analysis with Strength (weak-to-weak, strong-to-weak), and Probe (lure, negative) as within-participant factors and Age Group (younger children, older children) as between-participant factor, for lures vs. negative probe RTs. Importantly here, we found a main effect of Probe, F (1, 87) = 8.61, p = .004, showing longer RTs to recognize and respond to lures ( M = 2395.68 ms, SD = 52 ms) than to negative probes ( M = 2208.06 ms, SD = 44 ms).

In addition, to test our hypotheses more specifically, a linear mixed-effects model was constructed with 2-way interactions between Age Group (younger children, older children) and Strength (strong-to-weak, weak-to-weak) and was run on lures only, as these represent a measure of the ability to inhibit irrelevant information once completed the updating task. The model revealed a significant effect of Age Group, F (1, 85.250) = 16.92, p < .001. In addition, we found a main effect of Strength, F (1, 87.394) = 45.75, p < .001.

The two-way Strength by Age Group interaction reached significance, F (1, 87.394) = 25.57, p < .001. Subsequently, post-hoc comparisons showed that rejection of a lure from a strong-to-weak association needed longer RTs (compared to weak-to-weak association), but only for older children, t (87.39) = 8.41, p < .001. Rejection of a lure from a strong-to-weak association did not differ from a weak-to-weak condition in younger children, t (87.39) = 1.20, p = .23, as shown in Fig 3 .

Plot dots represent mean predicted RTs (ms) at lure rejection and bars represent 95% CIs.

https://doi.org/10.1371/journal.pone.0217697.g003

We believe our task is mainly based on phonological/orthographic knowledge and less on lexico-semantic knowledge (see also [ 30 ]). In fact, in order to engage with the task rapidly and effectively, the child should have developed an automatic access to orthographic/letter form representation. Therefore, we do not predict any specific vocabulary-related effect. However, in order to control for the role of vocabulary in the process examined, we ran the same mixed-effect models, covarying for vocabulary. Overall, the results did not change, showing the same effects and significance levels for both target probes (main effect of Strength, p = .002) and lures (Age group, Strength, and two-ways interaction, all ps < .001).

Conclusions

In this study, our aim was to investigate how LTM associations affect updating development. Updating is a complex activity that involves inhibition at different levels such as from the same lists set, or from previous lists [ 9 ], with the distinguishing component of the item-removal process [ 16 , 18 ]. More specifically here, we analysed how the strength of LTM association between items affects updating from a developmental perspective.

Typically, the literature on adults shows enhanced recall for strongly associated items; the stronger the pre-existing association in LTM, the better the performance in WM. For updating, a somewhat different process is indicated (i.e. not only maintenance of information in the short term, but also removal of irrelevant information). In this case, the opposite was shown: the stronger the pre-existing association, the harder it is to dismantle it [ 26 ].

In addition, the first notable difference between updating and recall (i.e., slowing of RTs in the former) could be related to the number of cognitive operations required in the task. Indeed, recall involves maintenance of information only; whereas updating entails a further item-removal component. Therefore, it is reasonable to assume that an additional operation (i.e., item-removal) will add a cost in terms of longer processing latencies. However, results comparing updating performance compared to recall have demonstrated the reverse effect; that is a cost rather than a benefit. This difference is likely to be due to the nature of updating, an essential process in adaptation of WM content to new elements. In other words, updating involves integration of new elements, as well as new bindings between elements (after disrupting previous ones), thus inhibiting and removing/substituting irrelevant information [ 11 , 16 ].

A recent model of updating [ 9 ] showed that updating develops via two main components of inhibition, one more related to control of inhibition from same lists; another one of inhibition from previous lists. The former, shows fewer developmental differences, the latter (also called PI control in [ 9 ]) shows greater age-related differences. In our view, the task used here with children is suitable for consideration of both components in terms of processing speed (an index useful in studying development via more subtle and fine-grained measurement). In fact, in the current task, each participant needs to maintain information and inhibit it, when no longer relevant, by substituting with new information during the tasks (same list inhibition component). Further, to ensure effective updating, s/he has to control for interference from previously studied items which are no longer relevant (i.e., inhibition from previously studied items set).

In particular, in accordance with [ 9 ] model, we found different outcomes consistent with the measures considered. Accordingly, the online RT showed a global age-related effect (older children faster than younger children), but not specific for strength with which letter were associated (in fact, no interaction). This finding could be accounted for, if we consider the development of self-monitoring (i.e., the ability to control one own’s behaviour) in children. That is, monitoring skills develop between 7 to 10 years, and subtle but important improvements are found over the primary school years [ 38 ]. Our self-paced task, where the child had to press the spacebar when s/he thinks to have memorized/updated a given mental set, requires a self-judgment of performance from the child him/herself. In particular, it has been shown that children (from 8 years of age) are more accurate in judgment of learning when given after a delay of about 2 minutes, than immediately after study [ 39 ]. Thus our task (which requires self-monitoring of learning during the study/updating phases, and immediately after, in order to press the spacebar) might not enhance an appropriate child self-regulation. For this reason, we believe we did not find age-related effects relative to strength for self-paced RTs and thus, failed to replicate the effects found with adults [ 14 , 26 ].

Conversely, for offline inhibitory control (i.e., lure recognition), we found more pronounced developmental effects, with significant differences; older children took more time to reject strong lures than weak, whereas no difference was observed for younger children ( Fig 3 ). Therefore, we found that online inhibition component was less affected by developmental change: younger children are able to perform updating tasks successfully. The real challenge in updating (i.e., due to control for previously relevant information) elicits significantly better performance from 10 years onwards. Here, in fact there is no need for self-regulation (i.e., as in the probe recognition task) as the task is not self-paced. The modulation of association strength development in older children (but not in younger) could be well accounted by the development of both lexico-orthographic knowledge and executive mechanisms that can work simultaneously [ 5 , 6 ].

This finding supports claims that the ability to inhibit irrelevant information is a fundamental mechanism that underlines many other developmental changes [ 40 , 41 , 42 ]. In particular, decreased susceptibility to interference is observable as age increases; 7/8 years olds children were shown to be more susceptible to interference than 9/10 years old [ 40 ], as we found in our study. However, we believe the novelty of the current study lies in the specificity of the experimental manipulation. Notably, these results indicated that, from 10 years onward, children found highly familiar stimuli (such as letters) more intrusive and difficult to control when strongly associated. Therefore, although we find that older children are less susceptible to interference, it seems that they are more sensitive to strong and weakly associated stimuli, similarly to performance in adults [ 14 , 26 ].

Future studies should further investigate any additional benefits/costs in updating strong and weak LTM associations, by also manipulating the strength of the item-association at updating [ 14 ]. Through this further manipulation, a more fine-grained examination of the dismantling and recreation of associations during updating would be enabled, including analysis of the relative ease/difficulty of the process. In addition, it could be useful to administer the task to children with specific learning disorders in order to show possible modulation of WM performance by LTM knowledge. Specifically, the task could then be useful to implement ad hoc measures to train children to remediate identified weaknesses, both in educational and clinical settings. We might also speculate that, as we found that strong LTM associations are more difficult to modify, this could in turn indicate the importance of correct support for the child, so that s/he will not act to strengthen incorrect sub-lexical/phonotactic associations. Indeed, it is likely that the more those incorrect associations are reinforced, the harder will be to modify/update them.

In conclusion, the present study demonstrated how WM updating is affected by LTM strength of association in a developmental sample. A significant cost of dismantling and updating strong associations was shown, and this effect was independent from age; all children from 7 to 10 years were comparably sensitive to association strength. In addition, results allowed us to differentiate age-related effects for interference control in updating of strong LTM associations; older children (but not younger) were more susceptible to interference from strongly-associated information.

Supporting information

S1 data. dataset online rts..

https://doi.org/10.1371/journal.pone.0217697.s001

S2 Data. Dataset recognition probe RTs.

https://doi.org/10.1371/journal.pone.0217697.s002

Acknowledgments

We wish to thank all children and schools participating in the study. We also thank Beatrice Colombani for her help with data collection.

• View Article
• Google Scholar
• PubMed/NCBI
• 32. Thurstone L. L. & Thurstone T. G. (1963). Italian edition 1981. PMA : Batteria delle attitudini mentali primarie , 7–11 anni [PMA : Primary mental abilities , 7–11 years ]. Edizioni Giunti OS: Firenze.
• 33. De Mauro T., Mancini F., Vedovelli M. & Voghera M. (1993). Lessico di frequenza dell'italiano parlato. [Frequency Lexicon of Spoken Italian Language] Milano: Etas Libri.
• 36. R Development Core Team. (2010). R: A language and environment for statistical computing. Vienna, Austria: R foundation for statistical computing.

REVIEW article

Working memory from the psychological and neurosciences perspectives: a review.

• 1 Department of Neurosciences, School of Medical Sciences, Universiti Sains Malaysia, Kubang Kerian, Malaysia
• 2 Center for Neuroscience Services and Research, Universiti Sains Malaysia, Kubang Kerian, Malaysia

Since the concept of working memory was introduced over 50 years ago, different schools of thought have offered different definitions for working memory based on the various cognitive domains that it encompasses. The general consensus regarding working memory supports the idea that working memory is extensively involved in goal-directed behaviors in which information must be retained and manipulated to ensure successful task execution. Before the emergence of other competing models, the concept of working memory was described by the multicomponent working memory model proposed by Baddeley and Hitch. In the present article, the authors provide an overview of several working memory-relevant studies in order to harmonize the findings of working memory from the neurosciences and psychological standpoints, especially after citing evidence from past studies of healthy, aging, diseased, and/or lesioned brains. In particular, the theoretical framework behind working memory, in which the related domains that are considered to play a part in different frameworks (such as memory’s capacity limit and temporary storage) are presented and discussed. From the neuroscience perspective, it has been established that working memory activates the fronto-parietal brain regions, including the prefrontal, cingulate, and parietal cortices. Recent studies have subsequently implicated the roles of subcortical regions (such as the midbrain and cerebellum) in working memory. Aging also appears to have modulatory effects on working memory; age interactions with emotion, caffeine and hormones appear to affect working memory performances at the neurobiological level. Moreover, working memory deficits are apparent in older individuals, who are susceptible to cognitive deterioration. Another younger population with working memory impairment consists of those with mental, developmental, and/or neurological disorders such as major depressive disorder and others. A less coherent and organized neural pattern has been consistently reported in these disadvantaged groups. Working memory of patients with traumatic brain injury was similarly affected and shown to have unusual neural activity (hyper- or hypoactivation) as a general observation. Decoding the underlying neural mechanisms of working memory helps support the current theoretical understandings concerning working memory, and at the same time provides insights into rehabilitation programs that target working memory impairments from neurophysiological or psychological aspects.

Introduction

Working memory has fascinated scholars since its inception in the 1960’s ( Baddeley, 2010 ; D’Esposito and Postle, 2015 ). Indeed, more than a century of scientific studies revolving around memory in the fields of psychology, biology, or neuroscience have not completely agreed upon a unified categorization of memory, especially in terms of its functions and mechanisms ( Cowan, 2005 , 2008 ; Baddeley, 2010 ). From the coining of the term “memory” in the 1880’s by Hermann Ebbinghaus, to the distinction made between primary and secondary memory by William James in 1890, and to the now widely accepted and used categorizations of memory that include: short-term, long-term, and working memories, studies that have tried to decode and understand this abstract concept called memory have been extensive ( Cowan, 2005 , 2008 ). Short and long-term memory suggest that the difference between the two lies in the period that the encoded information is retained. Other than that, long-term memory has been unanimously understood as a huge reserve of knowledge about past events, and its existence in a functioning human being is without dispute ( Cowan, 2008 ). Further categorizations of long-term memory include several categories: (1) episodic; (2) semantic; (3) Pavlovian; and (4) procedural memory ( Humphreys et al., 1989 ). For example, understanding and using language in reading and writing demonstrates long-term storage of semantics. Meanwhile, short-term memory was defined as temporarily accessible information that has a limited storage time ( Cowan, 2008 ). Holding a string of meaningless numbers in the mind for brief delays reflects this short-term component of memory. Thus, the concept of working memory that shares similarities with short-term memory but attempts to address the oversimplification of short-term memory by introducing the role of information manipulation has emerged ( Baddeley, 2012 ). This article seeks to present an up-to-date introductory overview of the realm of working memory by outlining several working memory studies from the psychological and neurosciences perspectives in an effort to refine and unite the scientific knowledge concerning working memory.

The Multicomponent Working Memory Model

When one describes working memory, the multicomponent working memory model is undeniably one of the most prominent working memory models that is widely cited in literatures ( Baars and Franklin, 2003 ; Cowan, 2005 ; Chein et al., 2011 ; Ashkenazi et al., 2013 ; D’Esposito and Postle, 2015 ; Kim et al., 2015 ). Baddeley and Hitch (1974) proposed a working memory model that revolutionized the rigid and dichotomous view of memory as being short or long-term, although the term “working memory” was first introduced by Miller et al. (1960) . The working memory model posited that as opposed to the simplistic functions of short-term memory in providing short-term storage of information, working memory is a multicomponent system that manipulates information storage for greater and more complex cognitive utility ( Baddeley and Hitch, 1974 ; Baddeley, 1996 , 2000b ). The three subcomponents involved are phonological loop (or the verbal working memory), visuospatial sketchpad (the visual-spatial working memory), and the central executive which involves the attentional control system ( Baddeley and Hitch, 1974 ; Baddeley, 2000b ). It was not until 2000 that another component termed “episodic buffer” was introduced into this working memory model ( Baddeley, 2000a ). Episodic buffer was regarded as a temporary storage system that modulates and integrates different sensory information ( Baddeley, 2000a ). In short, the central executive functions as the “control center” that oversees manipulation, recall, and processing of information (non-verbal or verbal) for meaningful functions such as decision-making, problem-solving or even manuscript writing. In Baddeley and Hitch (1974) ’s well-cited paper, information received during the engagement of working memory can also be transferred to long-term storage. Instead of seeing working memory as merely an extension and a useful version of short-term memory, it appears to be more closely related to activated long-term memory, as suggested by Cowan (2005 , 2008 ), who emphasized the role of attention in working memory; his conjectures were later supported by Baddeley (2010) . Following this, the current development of the multicomponent working memory model could be retrieved from Baddeley’s article titled “Working Memory” published in Current Biology , in Figure 2 ( Baddeley, 2010 ).

An Embedded-Processes Model of Working Memory

Notwithstanding the widespread use of the multicomponent working memory model, Cowan (1999 , 2005 ) proposed the embedded-processes model that highlights the roles of long-term memory and attention in facilitating working memory functioning. Arguing that the Baddeley and Hitch (1974) model simplified perceptual processing of information presentation to the working memory store without considering the focus of attention to the stimuli presented, Cowan (2005 , 2010 ) stressed the pivotal and central roles of working memory capacity for understanding the working memory concept. According to Cowan (2008) , working memory can be conceptualized as a short-term storage component with a capacity limit that is heavily dependent on attention and other central executive processes that make use of stored information or that interact with long-term memory. The relationships between short-term, long-term, and working memory could be presented in a hierarchical manner whereby in the domain of long-term memory, there exists an intermediate subset of activated long-term memory (also the short-term storage component) and working memory belongs to the subset of activated long-term memory that is being attended to ( Cowan, 1999 , 2008 ). An illustration of Cowan’s theoretical framework on working memory can be traced back to Figure 1 in his paper titled “What are the differences between long-term, short-term, and working memory?” published in Progress in Brain Research ( Cowan, 2008 ).

Alternative Models

Cowan’s theoretical framework toward working memory is consistent with Engle (2002) ’s view, in which it was posited that working memory capacity is comparable to directed or held attention information inhibition. Indeed, in their classic study on reading span and reading comprehension, Daneman and Carpenter (1980) demonstrated that working memory capacity, which was believed to be reflected by the reading span task, strongly correlated with various comprehension tests. Surely, recent and continual growth in the memory field has also demonstrated the development of other models such as the time-based resource-sharing model proposed by several researchers ( Barrouillet et al., 2004 , 2009 ; Barrouillet and Camos, 2007 ). This model similarly demonstrated that cognitive load and working memory capacity that were so often discussed by working memory researchers were mainly a product of attention that one receives to allocate to tasks at hand ( Barrouillet et al., 2004 , 2009 ; Barrouillet and Camos, 2007 ). In fact, the allocated cognitive resources for a task (such as provided attention) and the duration of such allocation dictated the likelihood of success in performing the tasks ( Barrouillet et al., 2004 , 2009 ; Barrouillet and Camos, 2007 ). This further highlighted the significance of working memory in comparison with short-term memory in that, although information retained during working memory is not as long-lasting as long-term memory, it is not the same and deviates from short-term memory for it involves higher-order processing and executive cognitive controls that are not observed in short-term memory. A more detailed presentation of other relevant working memory models that shared similar foundations with Cowan’s and emphasized the roles of long-term memory can be found in the review article by ( D’Esposito and Postle, 2015 ).

In addition, in order to understand and compare similarities and disparities in different proposed models, about 20 years ago, Miyake and Shah (1999) suggested theoretical questions to authors of different models in their book on working memory models. The answers to these questions and presentations of models by these authors gave rise to a comprehensive definition of working memory proposed by Miyake and Shah (1999 , p. 450), “working memory is those mechanisms or processes that are involved in the control, regulation, and active maintenance of task-relevant information in the service of complex cognition, including novel as well as familiar, skilled tasks. It consists of a set of processes and mechanisms and is not a fixed ‘place’ or ‘box’ in the cognitive architecture. It is not a completely unitary system in the sense that it involves multiple representational codes and/or different subsystems. Its capacity limits reflect multiple factors and may even be an emergent property of the multiple processes and mechanisms involved. Working memory is closely linked to LTM, and its contents consist primarily of currently activated LTM representations, but can also extend to LTM representations that are closely linked to activated retrieval cues and, hence, can be quickly activated.” That said, in spite of the variability and differences that have been observed following the rapid expansion of working memory understanding and its range of models since the inception of the multicomponent working memory model, it is worth highlighting that the roles of executive processes involved in working memory are indisputable, irrespective of whether different components exist. Such notion is well-supported as Miyake and Shah, at the time of documenting the volume back in the 1990’s, similarly noted that the mechanisms of executive control were being heavily investigated and emphasized ( Miyake and Shah, 1999 ). In particular, several domains of working memory such as the focus of attention ( Cowan, 1999 , 2008 ), inhibitory controls ( Engle and Kane, 2004 ), maintenance, manipulation, and updating of information ( Baddeley, 2000a , 2010 ), capacity limits ( Cowan, 2005 ), and episodic buffer ( Baddeley, 2000a ) were executive processes that relied on executive control efficacy (see also Miyake and Shah, 1999 ; Barrouillet et al., 2004 ; D’Esposito and Postle, 2015 ).

The Neuroscience Perspective

Following such cognitive conceptualization of working memory developed more than four decades ago, numerous studies have intended to tackle this fascinating working memory using various means such as decoding its existence at the neuronal level and/or proposing different theoretical models in terms of neuronal activity or brain activation patterns. Table 1 offers the summarized findings of these literatures. From the cognitive neuroscientific standpoint, for example, the verbal and visual-spatial working memories were examined separately, and the distinction between the two forms was documented through studies of patients with overt impairment in short-term storage for different verbal or visual tasks ( Baddeley, 2000b ). Based on these findings, associations or dissociations with the different systems of working memory (such as phonological loops and visuospatial sketchpad) were then made ( Baddeley, 2000b ). It has been established that verbal and acoustic information activates Broca’s and Wernicke’s areas while visuospatial information is represented in the right hemisphere ( Baddeley, 2000b ). Not surprisingly, many supporting research studies have pointed to the fronto-parietal network involving the dorsolateral prefrontal cortex (DLPFC), the anterior cingulate cortex (ACC), and the parietal cortex (PAR) as the working memory neural network ( Osaka et al., 2003 ; Owen et al., 2005 ; Chein et al., 2011 ; Kim et al., 2015 ). More precisely, the DLPFC has been largely implicated in tasks demanding executive control such as those requiring integration of information for decision-making ( Kim et al., 2015 ; Jimura et al., 2017 ), maintenance and manipulation/retrieval of stored information or relating to taxing loads (such as capacity limit) ( Osaka et al., 2003 ; Moore et al., 2013 ; Vartanian et al., 2013 ; Rodriguez Merzagora et al., 2014 ), and information updating ( Murty et al., 2011 ). Meanwhile, the ACC has been shown to act as an “attention controller” that evaluates the needs for adjustment and adaptation of received information based on task demands ( Osaka et al., 2003 ), and the PAR has been regarded as the “workspace” for sensory or perceptual processing ( Owen et al., 2005 ; Andersen and Cui, 2009 ). Figure 1 attempted to translate the theoretical formulation of the multicomponent working memory model ( Baddeley, 2010 ) to specific regions in the human brain. It is, however, to be acknowledged that the current neuroscientific understanding on working memory adopted that working memory, like other cognitive systems, involves the functional integration of the brain as a whole; and to clearly delineate its roles into multiple components with only a few regions serving as specific buffers was deemed impractical ( D’Esposito and Postle, 2015 ). Nonetheless, depicting the multicomponent working memory model in the brain offers a glimpse into the functional segregation of working memory.

TABLE 1. Working memory (WM) studies in the healthy brain.

FIGURE 1. A simplified depiction (adapted from the multicomponent working memory model by Baddeley, 2010 ) as implicated in the brain, in which the central executive assumes the role to exert control and oversee the manipulation of incoming information for intended execution. ACC, Anterior cingulate cortex.

Further investigation has recently revealed that other than the generally informed cortical structures involved in verbal working memory, basal ganglia, which lies in the subcortical layer, plays a role too ( Moore et al., 2013 ). Particularly, the caudate and thalamus were activated during task encoding, and the medial thalamus during the maintenance phase, while recorded activity in the fronto-parietal network, which includes the DLPFC and the parietal lobules, was observed only during retrieval ( Moore et al., 2013 ). These findings support the notion that the basal ganglia functions to enhance focusing on a target while at the same time suppressing irrelevant distractors during verbal working memory tasks, which is especially crucial at the encoding phase ( Moore et al., 2013 ). Besides, a study conducted on mice yielded a similar conclusion in which the mediodorsal thalamus aided the medial prefrontal cortex in the maintenance of working memory ( Bolkan et al., 2017 ). In another study by Murty et al. (2011) in which information updating, which is one of the important aspects of working memory, was investigated, the midbrain including the substantia nigra/ventral tegmental area and caudate was activated together with DLPFC and other parietal regions. Taken together, these studies indicated that brain activation of working memory are not only limited to the cortical layer ( Murty et al., 2011 ; Moore et al., 2013 ). In fact, studies on cerebellar lesions subsequently discovered that patients suffered from impairments in attention-related working memory or executive functions, suggesting that in spite of the motor functions widely attributed to the cerebellum, the cerebellum is also involved in higher-order cognitive functions including working memory ( Gottwald et al., 2004 ; Ziemus et al., 2007 ).

Shifting the attention to the neuronal network involved in working memory, effective connectivity analysis during engagement of a working memory task reinforced the idea that the DLPFC, PAR and ACC belong to the working memory circuitry, and bidirectional endogenous connections between all these regions were observed in which the left and right PAR were the modeled input regions ( Dima et al., 2014 ) (refer to Supplementary Figure 1 in Dima et al., 2014 ). Effective connectivity describes the attempt to model causal influence of neuronal connections in order to better understand the hidden neuronal states underlying detected neuronal responses ( Friston et al., 2013 ). Another similar study of working memory using an effective connectivity analysis that involved more brain regions, including the bilateral middle frontal gyrus (MFG), ACC, inferior frontal cortex (IFC), and posterior parietal cortex (PPC) established the modulatory effect of working memory load in this fronto-parietal network with memory delay as the driving input to the bilateral PPC ( Ma et al., 2012 ) (refer to Figure 1 in Ma et al., 2012 ).

Moving away from brain regions activated but toward the in-depth neurobiological side of working memory, it has long been understood that the limited capacity of working memory and its transient nature, which are considered two of the defining characteristics of working memory, indicate the role of persistent neuronal firing (see Review Article by D’Esposito and Postle, 2015 ; Zylberberg and Strowbridge, 2017 ; see also Silvanto, 2017 ), that is, continuous action potentials are generated in neurons along the neural network. However, this view was challenged when activity-silent synaptic mechanisms were found to also be involved ( Mongillo et al., 2008 ; Rose et al., 2016 ; see also Silvanto, 2017 ). Instead of holding relevant information through heightened and persistent neuronal firing, residual calcium at the presynaptic terminals was suggested to have mediated the working memory process ( Mongillo et al., 2008 ). This synaptic theory was further supported when TMS application produced a reactivation effect of past information that was not needed or attended at the conscious level, hence the TMS application facilitated working memory efficacy ( Rose et al., 2016 ). As it happens, this provided evidence from the neurobiological viewpoint to support Cowan’s theorized idea of “activated long-term memory” being a feature of working memory as non-cued past items in working memory that were assumed to be no longer accessible were actually stored in a latent state and could be brought back into consciousness. However, the researchers cautioned the use of the term “activated long-term memory” and opted for “prioritized long-term memory” because these unattended items maintained in working memory seemed to employ a different mechanism than items that were dropped from working memory ( Rose et al., 2016 ). Other than the synaptic theory, the spiking working memory model proposed by Fiebig and Lansner (2017) that borrowed the concept from fast Hebbian plasticity similarly disagreed with persistent neuronal activity and demonstrated that working memory processes were instead manifested in discrete oscillatory bursts.

Age and Working Memory

Nevertheless, having established a clear working memory circuitry in the brain, differences in brain activations, neural patterns or working memory performances are still apparent in different study groups, especially in those with diseased or aging brains. For a start, it is well understood that working memory declines with age ( Hedden and Gabrieli, 2004 ; Ziaei et al., 2017 ). Hence, older participants are expected to perform poorer on a working memory task when making comparison with relatively younger task takers. In fact, it was reported that decreases in cortical surface area in the frontal lobe of the right hemisphere was associated with poorer performers ( Nissim et al., 2017 ). In their study, healthy (those without mild cognitive impairments [MCI] or neurodegenerative diseases such as dementia or Alzheimer’s) elderly people with an average age of 70 took the n-back working memory task while magnetic resonance imaging (MRI) scans were obtained from them ( Nissim et al., 2017 ). The outcomes exhibited that a decrease in cortical surface areas in the superior frontal gyrus, pars opercularis of the inferior frontal gyrus, and medial orbital frontal gyrus that was lateralized to the right hemisphere, was significantly detected among low performers, implying an association between loss of brain structural integrity and working memory performance ( Nissim et al., 2017 ). There was no observed significant decline in cortical thickness of the studied brains, which is assumed to implicate neurodegenerative tissue loss ( Nissim et al., 2017 ).

Moreover, another extensive study that examined cognitive functions of participants across the lifespan using functional magnetic resonance imaging (fMRI) reported that the right lateralized fronto-parietal regions in addition to the ventromedial prefrontal cortex (VMPFC), posterior cingulate cortex, and left angular and middle frontal gyri (the default mode regions) in older adults showed reduced modulation of task difficulty, which was reflective of poorer task performance ( Rieck et al., 2017 ). In particular, older-age adults (55–69 years) exhibited diminished brain activations (positive modulation) as compared to middle-age adults (35–54 years) with increasing task difficulty, whereas lesser deactivation (negative modulation) was observed between the transition from younger adults (20–34 years) to middle-age adults ( Rieck et al., 2017 ). This provided insights on cognitive function differences during an individual’s lifespan at the neurobiological level, which hinted at the reduced ability or efficacy of the brain to modulate functional regions to increased difficulty as one grows old ( Rieck et al., 2017 ). As a matter of fact, such an opinion was in line with the Compensation-Related Utilization of Neural Circuits Hypothesis (CRUNCH) proposed by Reuter-Lorenz and Cappell (2008) . The CRUNCH likewise agreed upon reduced neural efficiency in older adults and contended that age-associated cognitive decline brought over-activation as a compensatory mechanism; yet, a shift would occur as task loads increase and under-activation would then be expected because older adults with relatively lesser cognitive resources would max out their ‘cognitive reserve’ sooner than younger adults ( Reuter-Lorenz and Park, 2010 ; Schneider-Garces et al., 2010 ).

In addition to those findings, emotional distractors presented during a working memory task were shown to alter or affect task performance in older adults ( Oren et al., 2017 ; Ziaei et al., 2017 ). Based on the study by Oren et al. (2017) who utilized the n-back task paired with emotional distractors with neutral or negative valence in the background, negative distractors with low load (such as 1-back) resulted in shorter response time (RT) in the older participants ( M age = 71.8), although their responses were not significantly more accurate when neutral distractors were shown. Also, lesser activations in the bilateral MFG, VMPFC, and left PAR were reported in the old-age group during negative low load condition. This finding subsequently demonstrated the results of emotional effects on working memory performance in older adults ( Oren et al., 2017 ). Further functional connectivity analyses revealed that the amygdala, the region well-known to be involved in emotional processing, was deactivated and displayed similar strength in functional connectivity regardless of emotional or load conditions in the old-age group ( Oren et al., 2017 ). This finding went in the opposite direction of that observed in the younger group in which the amygdala was strongly activated with less functional connections to the bilateral MFG and left PAR ( Oren et al., 2017 ). This might explain the shorter reported RT, which was an indication of improved working memory performance, during the emotional working memory task in the older adults as their amygdala activation was suppressed as compared to the younger adults ( Oren et al., 2017 ).

Interestingly, a contrasting neural connection outcome was reported in the study by Ziaei et al. (2017) in which differential functional networks relating to emotional working memory task were employed by the two studied groups: (1) younger ( M age = 22.6) and (2) older ( M age = 68.2) adults. In the study, emotional distractors with positive, neutral, and negative valence were presented during a visual working memory task and older adults were reported to adopt two distinct networks involving the VMPFC to encode and process positive and negative distractors while younger adults engaged only one neural pathway ( Ziaei et al., 2017 ). The role of amygdala engagement in processing only negative items in the younger adults, but both negative and positive distractors in the older adults, could be reflective of the older adults’ better ability at regulating negative emotions which might subsequently provide a better platform for monitoring working memory performance and efficacy as compared to their younger counterparts ( Ziaei et al., 2017 ). This study’s findings contradict those by Oren et al. (2017) in which the amygdala was found to play a bigger role in emotional working memory tasks among older participants as opposed to being suppressed as reported by Oren et al. (2017) . Nonetheless, after overlooking the underlying neural mechanism relating to emotional distractors, it was still agreed that effective emotional processing sustained working memory performance among older/elderly people ( Oren et al., 2017 ; Ziaei et al., 2017 ).

Aside from the interaction effect between emotion and aging on working memory, the impact of caffeine was also investigated among elders susceptible to age-related cognitive decline; and those reporting subtle cognitive deterioration 18-months after baseline measurement showed less marked effects of caffeine in the right hemisphere, unlike those with either intact cognitive ability or MCI ( Haller et al., 2017 ). It was concluded that while caffeine’s effects were more pronounced in MCI participants, elders in the early stages of cognitive decline displayed diminished sensitivity to caffeine after being tested with the n-back task during fMRI acquisition ( Haller et al., 2017 ). It is, however, to be noted that the working memory performance of those displaying minimal cognitive deterioration was maintained even though their brain imaging uncovered weaker brain activation in a more restricted area ( Haller et al., 2017 ). Of great interest, such results might present a useful brain-based marker that can be used to identify possible age-related cognitive decline.

Similar findings that demonstrated more pronounced effects of caffeine on elderly participants were reported in an older study, whereas older participants in the age range of 50–65 years old exhibited better working memory performance that offset the cognitive decline observed in those with no caffeine consumption, in addition to displaying shorter reaction times and better motor speeds than observed in those without caffeine ( Rees et al., 1999 ). Animal studies using mice showed replication of these results in mutated mice models of Alzheimer’s disease or older albino mice, both possibly due to the reported results of reduced amyloid production or brain-derived neurotrophic factor and tyrosine-kinase receptor. These mice performed significantly better after caffeine treatment in tasks that supposedly tapped into working memory or cognitive functions ( Arendash et al., 2006 ). Such direct effects of caffeine on working memory in relation to age was further supported by neuroimaging studies ( Haller et al., 2013 ; Klaassen et al., 2013 ). fMRI uncovered increased brain activation in regions or networks of working memory, including the fronto-parietal network or the prefrontal cortex in old-aged ( Haller et al., 2013 ) or middle-aged adults ( Klaassen et al., 2013 ), even though the behavioral measures of working memory did not differ. Taken together, these outcomes offered insight at the neurobiological level in which caffeine acts as a psychoactive agent that introduces changes and alters the aging brain’s biological environment that explicit behavioral testing might fail to capture due to performance maintenance ( Haller et al., 2013 , 2017 ; Klaassen et al., 2013 ).

With respect to physiological effects on cognitive functions (such as effects of caffeine on brain physiology), estradiol, the primary female sex hormone that regulates menstrual cycles, was found to also modulate working memory by engaging different brain activity patterns during different phases of the menstrual cycle ( Joseph et al., 2012 ). The late follicular (LF) phase of the menstrual cycle, characterized by high estradiol levels, was shown to recruit more of the right hemisphere that was associated with improved working memory performance than did the early follicular (EF) phase, which has lower estradiol levels although overall, the direct association between estradiol levels and working memory was inconclusive ( Joseph et al., 2012 ). The finding that estradiol levels modified brain recruitment patterns at the neurobiological level, which could indirectly affect working memory performance, presents implications that working memory impairment reported in post-menopausal women (older aged women) could indicate a link with estradiol loss ( Joseph et al., 2012 ). In 2000, post-menopausal women undergoing hormone replacement therapy, specifically estrogen, were found to have better working memory performance in comparison with women who took estrogen and progestin or women who did not receive the therapy ( Duff and Hampson, 2000 ). Yet, interestingly, a study by Janowsky et al. (2000) showed that testosterone supplementation counteracted age-related working memory decline in older males, but a similar effect was not detected in older females who were supplemented with estrogen. A relatively recent paper might have provided the explanation to such contradicting outcomes ( Schöning et al., 2007 ). As demonstrated in the study using fMRI, the nature of the task (such as verbal or visual-spatial) might have played a role as a higher level of testosterone (in males) correlated with activations of the left inferior parietal cortex, which was deemed a key region in spatial processing that subsequently brought on better performance in a mental-rotation task. In contrast, significant correlation between estradiol and other cortical activations in females in the midluteal phase, who had higher estradiol levels, did not result in better performance of the task compared to women in the EF phase or men ( Schöning et al., 2007 ). Nonetheless, it remains premature to conclude that age-related cognitive decline was a result of hormonal (estradiol or testosterone) fluctuations although hormones might have modulated the effect of aging on working memory.

Other than the presented interaction effects of age and emotions, caffeine, and hormones, other studies looked at working memory training in the older population in order to investigate working memory malleability in the aging brain. Findings of improved performance for the same working memory task after training were consistent across studies ( Dahlin et al., 2008 ; Borella et al., 2017 ; Guye and von Bastian, 2017 ; Heinzel et al., 2017 ). Such positive results demonstrated effective training gains regardless of age difference that could even be maintained until 18 months later ( Dahlin et al., 2008 ) even though the transfer effects of such training to other working memory tasks need to be further elucidated as strong evidence of transfer with medium to large effect size is lacking ( Dahlin et al., 2008 ; Guye and von Bastian, 2017 ; Heinzel et al., 2017 ; see also Karbach and Verhaeghen, 2014 ). The studies showcasing the effectiveness of working memory training presented a useful cognitive intervention that could partially stall or delay cognitive decline. Table 2 presents an overview of the age-related working memory studies.

TABLE 2. Working memory (WM) studies in relation to age.

The Diseased Brain and Working Memory

Age is not the only factor influencing working memory. In recent studies, working memory deficits in populations with mental or neurological disorders were also being investigated (see Table 3 ). Having identified that the working memory circuitry involves the fronto-parietal region, especially the prefrontal and parietal cortices, in a healthy functioning brain, targeting these areas in order to understand how working memory is affected in a diseased brain might provide an explanation for the underlying deficits observed at the behavioral level. For example, it was found that individuals with generalized or social anxiety disorder exhibited reduced DLPFC activation that translated to poorer n-back task performance in terms of accuracy and RT when compared with the controls ( Balderston et al., 2017 ). Also, VMPFC and ACC, representing the default mode network (DMN), were less inhibited in these individuals, indicating that cognitive resources might have been divided and resulted in working memory deficits due to the failure to disengage attention from persistent anxiety-related thoughts ( Balderston et al., 2017 ). Similar speculation can be made about individuals with schizophrenia. Observed working memory deficits might be traced back to impairments in the neural networks that govern attentional-control and information manipulation and maintenance ( Grot et al., 2017 ). The participants performed a working memory binding task, whereby they had to make sure that the word-ellipse pairs presented during the retrieval phase were identical to those in the encoding phase in terms of location and verbal information; results concluded that participants with schizophrenia had an overall poorer performance compared to healthy controls when they were asked to actively bind verbal and spatial information ( Grot et al., 2017 ). This was reflected in the diminished activation in the schizophrenia group’s ventrolateral prefrontal cortex and the PPC that were said to play a role in manipulation and reorganization of information during encoding and maintenance of information after encoding ( Grot et al., 2017 ).

TABLE 3. Working memory (WM) studies in the diseased brain.

In addition, patients with major depressive disorder (MDD) displayed weaker performance in the working memory updating domain in which information manipulation was needed when completing a visual working memory task ( Le et al., 2017 ). The working memory task employed in the study was a delayed recognition task that required participants to remember and recognize the faces or scenes as informed after stimuli presentation while undergoing fMRI scan ( Le et al., 2017 ). Subsequent functional connectivity analyses revealed that the fusiform face area (FFA), parahippocampal place area (PPA), and left MFG showed aberrant activity in the MDD group as compared to the control group ( Le et al., 2017 ). These brain regions are known to be the visual association area and the control center of working memory and have been implicated in visual working memory updating in healthy adults ( Le et al., 2017 ). Therefore, altered visual cortical functions and load-related activation in the prefrontal cortex in the MDD group implied that the cognitive control for visual information processing and updating might be impaired at the input or control level, which could have ultimately played a part in the depressive symptoms ( Le et al., 2017 ).

Similarly, during a verbal delayed match to sample task that asked participants to sub-articulatorly rehearse presented target letters for subsequent letter-matching, individuals with bipolar affective disorder displayed aberrant neural interactions between the right amygdala, which is part of the limbic system implicated in emotional processing as previously described, and ipsilateral cortical regions often concerned with verbal working memory, pointing out that the cortico-amygdalar connectivity was disrupted, which led to verbal working memory deficits ( Stegmayer et al., 2015 ). As an attempt to gather insights into previously reported hyperactivation in the amygdala in bipolar affective disorder during an articulatory working memory task, functional connectivity analyses revealed that negative functional interactions seen in healthy controls were not replicated in patients with bipolar affective disorder ( Stegmayer et al., 2015 ). Consistent with the previously described study about emotional processing effects on working memory in older adults, this reported outcome was suggestive of the brain’s failed attempts to suppress pathological amygdalar activation during a verbal working memory task ( Stegmayer et al., 2015 ).

Another affected group with working memory deficits that has been the subject of research interest was children with developmental disorders such as attention deficit/hyperactivity disorder (ADHD), developmental dyscalculia, and reading difficulties ( Rotzer et al., 2009 ; Ashkenazi et al., 2013 ; Wang and Gathercole, 2013 ; Maehler and Schuchardt, 2016 ). For instance, looking into the different working memory subsystems based on Baddeley’s multicomponent working memory model in children with dyslexia and/or ADHD and children with dyscalculia and/or ADHD through a series of tests, it was reported that distinctive working memory deficits by groups could be detected such that phonological loop (e.g., digit span) impairment was observed in the dyslexia group, visuospatial sketchpad (e.g., Corsi block tasks) deficits in the dyscalculia group, while central executive (e.g., complex counting span) deficits in children with ADHD ( Maehler and Schuchardt, 2016 ). Meanwhile, examination of working memory impairment in a delayed match-to-sample visual task that put emphasis on the maintenance phase of working memory by examining the brainwaves of adults with ADHD using electroencephalography (EEG) also revealed a marginally significantly lower alpha band power in the posterior regions as compared to healthy individuals, and such an observation was not significantly improved after working memory training (Cogmed working memory training, CWMT Program) ( Liu et al., 2016 ). The alpha power was considered important in the maintenance of working memory items; and lower working memory accuracy paired with lower alpha band power was indeed observed in the ADHD group ( Liu et al., 2016 ).

Not dismissing the above compiled results, children encountering disabilities in mathematical operations likewise indicated deficits in the working memory domain that were traceable to unusual brain activities at the neurobiological level ( Rotzer et al., 2009 ; Ashkenazi et al., 2013 ). It was speculated that visuospatial working memory plays a vital role when arithmetic problem-solving is involved in order to ensure intact mental representations of the numerical information ( Rotzer et al., 2009 ). Indeed, Ashkenazi et al. (2013) revealed that Block Recall, a variant of the Corsi Block Tapping test and a subtest of the Working Memory Test Battery for Children (WMTB-C) that explored visuospatial sketchpad ability, was significantly predictive of math abilities. In relation to this, studies investigating brain activation patterns and performance of visuospatial working memory task in children with mathematical disabilities identified the intraparietal sulcus (IPS), in conjunction with other regions in the prefrontal and parietal cortices, to have less activation when visuospatial working memory was deemed involved (during an adapted form of Corsi Block Tapping test made suitable for fMRI [ Rotzer et al., 2009 ]); in contrast the control group demonstrated correlations of the IPS in addition to the fronto-parietal cortical activation with the task ( Rotzer et al., 2009 ; Ashkenazi et al., 2013 ). These brain activity variations that translated to differences in overt performances between healthily developing individuals and those with atypical development highlighted the need for intervention and attention for the disadvantaged groups.

Traumatic Brain Injury and Working Memory

Physical injuries impacting the frontal or parietal lobes would reasonably be damaging to one’s working memory. This is supported in studies employing neuropsychological testing to assess cognitive impairments in patients with traumatic brain injury; and poorer cognitive performances especially involving the working memory domains were reported (see Review Articles by Dikmen et al., 2009 ; Dunning et al., 2016 ; Phillips et al., 2017 ). Research on cognitive deficits in traumatic brain injury has been extensive due to the debilitating conditions brought upon an individual daily life after the injury. Traumatic brain injuries (TBI) refer to accidental damage to the brain after being hit by an object or following rapid acceleration or deceleration ( Farrer, 2017 ). These accidents include falls, assaults, or automobile accidents and patients with TBI can be then categorized into three groups; (1) mild TBI with GCS – Glasgow Coma Scale – score of 13–15; (2) moderate TBI with GCS score of 9–12; and (3) severe TBI with GCS score of 3–8 ( Farrer, 2017 ). In a recently published meta-analysis that specifically looked at working memory impairments in patients with moderate to severe TBI, patients displayed reduced cognitive functions in verbal short-term memory in addition to verbal and visuospatial working memory in comparison to control groups ( Dunning et al., 2016 ). It was also understood from the analysis that the time lapse since injury and age of injury were deciding factors that influenced these cognitive deficits in which longer time post-injury or older age during injury were associated with greater cognitive decline ( Dunning et al., 2016 ).

Nonetheless, it is to be noted that such findings relating to age of injury could not be generalized to the child population since results from the pediatric TBI cases showed that damage could negatively impact developmental skills that could indicate a greater lag in cognitive competency as the child’s frontal lobe had yet to mature ( Anderson and Catroppa, 2007 ; Mandalis et al., 2007 ; Nadebaum et al., 2007 ; Gorman et al., 2012 ). These studies all reported working memory impairment of different domains such as attentional control, executive functions, or verbal and visuospatial working memory in the TBI group, especially for children with severe TBI ( Mandalis et al., 2007 ; Nadebaum et al., 2007 ; Gorman et al., 2012 ). Investigation of whether working memory deficits are domain-specific or -general or involve one or more mechanisms, has yielded inconsistent results. For example, Perlstein et al. (2004) found that working memory was impaired in the TBI group only when complex manipulation such as sequential coding of information is required and not accounted for by processing speed or maintenance of information, but two teams of researchers ( Perbal et al., 2003 ; Gorman et al., 2012 ) suggested otherwise. From their study on timing judgments, Perbal et al. (2003) concluded that deficits were not related to time estimation but more on generalized attentional control, working memory and processing speed problems; while Gorman et al. (2012) also attributed the lack of attentional focus to impairments observed during the working memory task. In fact, in a later study by Gorman et al. (2016) , it was shown that processing speed mediated TBI effects on working memory even though the mediation was partial. On the other hand, Vallat-Azouvi et al. (2007) reported impairments in the working memory updating domain that came with high executive demands for TBI patients. Also, Mandalis et al. (2007) similarly highlighted potential problems with attention and taxing cognitive demands in the TBI group.

From the neuroscientific perspective, hyper-activation or -connectivity in the working memory circuitry was reported in TBI patients in comparison with healthy controls when both groups engaged in working memory tasks, suggesting that the brain attempted to compensate for or re-establish lost connections upon the injury ( Dobryakova et al., 2015 ; Hsu et al., 2015 ; Wylie et al., 2015 ). For a start, it was observed that participants with mild TBI displayed increased activation in the right prefrontal cortex during a working memory task when comparing to controls ( Wylie et al., 2015 ). Interestingly, this activation pattern only occurred in patients who did not experience a complete recovery 1 week after the injury ( Wylie et al., 2015 ). Besides, low activation in the DMN was observed in mild TBI patients without cognitive recovery, and such results seemed to be useful in predicting recovery in patients in which the patients did not recover when hypoactivation (low activation) was reported, and vice versa ( Wylie et al., 2015 ). This might be suggestive of the potential of cognitive recovery simply by looking at the intensity of brain activation of the DMN, for an increase in activation of the DMN seemed to be superseded before cognitive recovery was present ( Wylie et al., 2015 ).

In fact, several studies lent support to the speculation mentioned above as hyperactivation or hypoactivation in comparison with healthy participants was similarly identified. When sex differences were being examined in working memory functional activity in mild TBI patients, hyperactivation was reported in male patients when comparing to the male control group, suggesting that the hyperactivation pattern might be the brain’s attempt at recovering impaired functions; even though hypoactivation was shown in female patients as compared to the female control group ( Hsu et al., 2015 ). The researchers from the study further explained that such hyperactivation after the trauma acted as a neural compensatory mechanism so that task performance could be maintained while hypoactivation with a poorer performance could have been the result of a more severe injury ( Hsu et al., 2015 ). Therefore, the decrease in activation in female patients, in addition to the observed worse performance, was speculated to be due to a more serious injury sustained by the female patients group ( Hsu et al., 2015 ).

In addition, investigation of the effective connectivity of moderate and severe TBI participants during a working memory task revealed that the VMPFC influenced the ACC in these TBI participants when the opposite was observed in healthy subjects ( Dobryakova et al., 2015 ). Moreover, increased inter-hemispheric transfer due to an increased number of connections between the left and right hemispheres (hyper-connectivity) without clear directionality of information flow (redundant connectivity) was also reported in the TBI participants ( Dobryakova et al., 2015 ). This study was suggestive of location-specific changes in the neural network connectivity following TBI depending on the cognitive functions at work, other than providing another support to the neural compensatory hypothesis due to the observed hyper-connectivity ( Dobryakova et al., 2015 ).

Nevertheless, inconsistent findings should not be neglected. In a study that also focused on brain connectivity analysis among patients with mild TBI by Hillary et al. (2011) , elevated task-related connectivity in the right hemisphere, in particular the prefrontal cortex, was consistently demonstrated during a working memory task while the control group showed greater left hemispheric activation. This further supported the right lateralization of the brain to reallocate cognitive resources of TBI patients post-injury. Meanwhile, the study did not manage to obtain the expected outcome in terms of greater clustering of whole-brain connections in TBI participants as hypothesized ( Hillary et al., 2011 ). That said, no significant loss or gain of connections due to the injury could be concluded from the study, as opposed to the hyper- or hypoactivation or hyper-connectivity frequently highlighted in other similar researches ( Hillary et al., 2011 ). Furthermore, a study by Chen et al. (2012) also failed to establish the same results of increased brain activation. Instead, with every increase of the working memory load, increase in brain activation, as expected to occur and as demonstrated in the control group, was unable to be detected in the TBI group ( Chen et al., 2012 ).

Taken all the insightful studies together, another aspect not to be neglected is the neuroimaging techniques employed in contributing to the literature on TBI. Modalities other than fMRI, which focuses on localization of brain activities, show other sides of the story of working memory impairments in TBI to offer a more holistic understanding. Studies adopting electroencephalography (EEG) or diffusor tensor imaging (DTI) reported atypical brainwaves coherence or white matter integrity in patients with TBI ( Treble et al., 2013 ; Ellis et al., 2016 ; Bailey et al., 2017 ; Owens et al., 2017 ). Investigating the supero-lateral medial forebrain bundle (MFB) that innervates and consequently terminates at the prefrontal cortex, microstructural white matter damage at the said area was indicated in participants with moderate to severe TBI by comparing its integrity with the control group ( Owens et al., 2017 ). Such observation was backed up by evidence showing that the patients performed more poorly on attention-loaded cognitive tasks of factors relating to slow processing speed than the healthy participants, although a direct association between MFB and impaired attentional system was not found ( Owens et al., 2017 ).

Correspondingly, DTI study of the corpus callosum (CC), which described to hold a vital role in connecting and coordinating both hemispheres to ensure competent cognitive functions, also found compromised microstructure of the CC with low fractional anisotropy and high mean diffusivity, both of which are indications of reduced white matter integrity ( Treble et al., 2013 ). This reported observation was also found to be predictive of poorer verbal or visuospatial working memory performance in callosal subregions connecting the parietal and temporal cortices ( Treble et al., 2013 ). Adding on to these results, using EEG to examine the functional consequences of CC damage revealed that interhemispheric transfer time (IHTT) of the CC was slower in the TBI group than the control group, suggesting an inefficient communication between the two hemispheres ( Ellis et al., 2016 ). In addition, the TBI group with slow IHTT as well exhibited poorer neurocognitive functioning including working memory than the healthy controls ( Ellis et al., 2016 ).

Furthermore, comparing the working memory between TBI, MDD, TBI-MDD, and healthy participants discovered that groups with MDD and TBI-MDD performed poorer on the Sternberg working memory task but functional connectivity on the other hand, showed that increased inter-hemispheric working memory gamma connectivity was observed in the TBI and TBI-MDD groups ( Bailey et al., 2017 ). Speculation provided for the findings of such neuronal state that was not reflected in the explicit working memory performance was that the deficits might not be detected or tested by the utilized Sternberg task ( Bailey et al., 2017 ). Another explanation attempting to answer the increase in gamma connectivity in these groups was the involvement of the neural compensatory mechanism after TBI to improve performance ( Bailey et al., 2017 ). Nevertheless, such outcome implies that behavioral performances or neuropsychological outcomes might not always be reflective of the functional changes happening in the brain.

Yet, bearing in mind that TBI consequences can be vast and crippling, cognitive improvement or recovery, though complicated due to the injury severity-dependent nature, is not impossible (see Review Article by Anderson and Catroppa, 2007 ; Nadebaum et al., 2007 ; Dikmen et al., 2009 ; Chen et al., 2012 ). As reported by Wylie et al. (2015) , cognitive improvement together with functional changes in the brain could be detected in individuals with mild TBI. Increased activation in the brain during 6-week follow-up was also observed in the mild TBI participants, implicating the regaining of connections in the brain ( Chen et al., 2012 ). Administration of certain cognitively enhancing drugs such as methylphenidate was reported to be helpful in improving working memory performance too ( Manktelow et al., 2017 ). Methylphenidate as a dopamine reuptake inhibitor was found to have modulated the neural activity in the left cerebellum which subsequently correlated with improved working memory performance ( Manktelow et al., 2017 ). A simplified summary of recent studies on working memory and TBI is tabulated in Table 4 .

TABLE 4. Working memory (WM) studies in the TBI group.

General Discussion and Future Direction

In practice, all of the aforementioned studies contribute to the working memory puzzle by addressing the topic from different perspectives and employing various methodologies to study it. Several theoretical models of working memory that conceptualized different working memory mechanisms or domains (such as focus of attention, inhibitory controls, maintenance and manipulation of information, updating and integration of information, capacity limits, evaluative and executive controls, and episodic buffer) have been proposed. Coupled with the working memory tasks of various means that cover a broad range (such as Sternberg task, n-back task, Corsi block-tapping test, Wechsler’s Memory Scale [WMS], and working memory subtests in the Wechsler Adult Intelligence Scale [WAIS] – Digit Span, Letter Number Sequencing), it has been difficult, if not highly improbable, for working memory studies to reach an agreement upon a consistent study protocol that is acceptable for generalization of results due to the constraints bound by the nature of the study. Various data acquisition and neuroimaging techniques that come with inconsistent validity such as paper-and-pen neuropsychological measures, fMRI, EEG, DTI, and functional near-infrared spectroscopy (fNIRS), or even animal studies can also be added to the list. This poses further challenges to quantitatively measure working memory as only a single entity. For example, when studying the neural patterns of working memory based on Cowan’s processes-embedded model using fMRI, one has to ensure that the working memory task selected is fMRI-compatible, and demands executive control of attention directed at activated long-term memory (domain-specific). That said, on the one hand, there are tasks that rely heavily on the information maintenance such as the Sternberg task; on the other hand, there are also tasks that look into the information manipulation updating such as the n-back or arithmetic task. Meanwhile, the digit span task in WAIS investigates working memory capacity, although it can be argued that it also encompasses the domain on information maintenance and updating-. Another consideration involves the different natures (verbal/phonological and visuospatial) of the working memory tasks as verbal or visuospatial information is believed to engage differing sensory mechanisms that might influence comparison of working memory performance between tasks of different nature ( Baddeley and Hitch, 1974 ; Cowan, 1999 ). For instance, though both are n-back tasks that includes the same working memory domains, the auditory n-back differs than the visual n-back as the information is presented in different forms. This feature is especially crucial with regards to the study populations as it differentiates between verbal and visuospatial working memory competence within individuals, which are assumed to be domain-specific as demonstrated by vast studies (such as Nadler and Archibald, 2014 ; Pham and Hasson, 2014 ; Nakagawa et al., 2016 ). These test variations undeniably present further difficulties in selecting an appropriate task. Nevertheless, the adoption of different modalities yielded diverging outcomes and knowledge such as behavioral performances, functional segregation and integration in the brain, white matter integrity, brainwave coherence, and oxy- and deoxyhaemoglobin concentrations that are undeniably useful in application to different fields of study.

In theory, the neural efficiency hypothesis explains that increased efficiency of the neural processes recruit fewer cerebral resources in addition to displaying lower activation in the involved neural network ( Vartanian et al., 2013 ; Rodriguez Merzagora et al., 2014 ). This is in contrast with the neural compensatory hypothesis in which it attempted to understand diminished activation that is generally reported in participants with TBI ( Hillary et al., 2011 ; Dobryakova et al., 2015 ; Hsu et al., 2015 ; Wylie et al., 2015 ; Bailey et al., 2017 ). In the diseased brain, low activation has often been associated with impaired cognitive function ( Chen et al., 2012 ; Dobryakova et al., 2015 ; Wylie et al., 2015 ). Opportunely, the CRUNCH model proposed within the field of aging might be translated and integrated the two hypotheses here as it suitably resolved the disparity of cerebral hypo- and hyper-activation observed in weaker, less efficient brains as compared to healthy, adept brains ( Reuter-Lorenz and Park, 2010 ; Schneider-Garces et al., 2010 ). Moreover, other factors such as the relationship between fluid intelligence and working memory might complicate the current understanding of working memory as a single, isolated construct since working memory is often implied in measurements of the intelligence quotient ( Cowan, 2008 ; Vartanian et al., 2013 ). Indeed, the process overlap theory of intelligence proposed by Kovacs and Conway (2016) in which the constructs of intelligence were heavily scrutinized (such as general intelligence factors, g and its smaller counterparts, fluid intelligence or reasoning, crystallized intelligence, perceptual speed, and visual-spatial ability), and fittingly connected working memory capacity with fluid reasoning. Cognitive tests such as Raven’s Progressive Matrices or other similar intelligence tests that demand complex cognition and were reported in the paper had been found to correlate strongly with tests of working memory ( Kovacs and Conway, 2016 ). Furthermore, in accordance with such views, in the same paper, neuroimaging studies found intelligence tests also activated the same fronto-parietal network observed in working memory ( Kovacs and Conway, 2016 ).

On the other hand, even though the roles of the prefrontal cortex in working memory have been widely established, region specificity and localization in the prefrontal cortex in relation to the different working memory domains such as manipulation or delayed retention of information remain at the premature stage (see Review Article by D’Esposito and Postle, 2015 ). It has been postulated that the neural mechanisms involved in working memory are of high-dimensionality and could not always be directly captured and investigated using neurophysiological techniques such as fMRI, EEG, or patch clamp recordings even when comparing with lesion data ( D’Esposito and Postle, 2015 ). According to D’Esposito and Postle (2015) , human fMRI studies have demonstrated that a rostral-caudal functional gradient related to level of abstraction required of working memory along the frontal cortex (in which different regions in the prefrontal cortex [from rostral to caudal] might be associated with different abstraction levels) might exist. Other functional gradients relating to different aspects of working memory were similarly unraveled ( D’Esposito and Postle, 2015 ). These proposed mechanisms with different empirical evidence point to the fact that conclusive understanding regarding working memory could not yet be achieved before the inconsistent views are reconciled.

Not surprisingly, with so many aspects of working memory yet to be understood and its growing complexity, the cognitive neuroscience basis of working memory requires constant research before an exhaustive account can be gathered. From the psychological conceptualization of working memory as attempted in the multicomponent working memory model ( Baddeley and Hitch, 1974 ), to the neural representations of working memory in the brain, especially in the frontal regions ( D’Esposito and Postle, 2015 ), one important implication derives from the present review of the literatures is that working memory as a psychological construct or a neuroscientific mechanism cannot be investigated as an isolated event. The need for psychology and neuroscience to interact with each other in an active feedback cycle exists in which this cognitive system called working memory can be dissected at the biological level and refined both empirically, and theoretically.

In summary, the present article offers an account of working memory from the psychological and neuroscientific perspectives, in which theoretical models of working memory are presented, and neural patterns and brain regions engaging in working memory are discussed among healthy and diseased brains. It is believed that working memory lays the foundation for many other cognitive controls in humans, and decoding the working memory mechanisms would be the first step in facilitating understanding toward other aspects of human cognition such as perceptual or emotional processing. Subsequently, the interactions between working memory and other cognitive systems could reasonably be examined.

Author Contributions

WC wrote the manuscript with critical feedback and consultation from AAH. WC and AAH contributed to the final version of the manuscript. JA supervised the process and proofread the manuscript.

This work was supported by the Transdisciplinary Research Grant Scheme (TRGS) 203/CNEURO/6768003 and the USAINS Research Grant 2016.

Conflict of Interest Statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The reviewer EB and handling Editor declared their shared affiliation.

Andersen, R. A., and Cui, H. (2009). Review intention, action planning, and decision making in parietal-frontal circuits. Neuron 63, 568–583. doi: 10.1016/j.neuron.2009.08.028

PubMed Abstract | CrossRef Full Text | Google Scholar

Anderson, V., and Catroppa, C. (2007). Memory outcome at 5 years post-childhood traumatic brain injury. Brain Inj. 21, 1399–1409. doi: 10.1080/02699050701785070

Arendash, G. W., Schleif, W., Rezai-Zadeh, K., Jackson, E. K., Zacharia, L. C., Cracchiolo, J. R., et al. (2006). Caffeine protects Alzheimer’s mice against cognitive impairment and reduces brain β-amyloid production. Neuroscience 142, 941–952. doi: 10.1016/j.neuroscience.2006.07.021

Ashkenazi, S., Rosenberg-lee, M., Metcalfe, A. W. S., Swigart, A. G., and Menon, V. (2013). Neuropsychologia visuo – spatial working memory is an important source of domain-general vulnerability in the development of arithmetic cognition. Neuropsychologia 51, 2305–2317. doi: 10.1016/j.neuropsychologia.2013.06.031

Baars, B. J., and Franklin, S. (2003). How conscious experience and working memory interact. Trends Cogn. Sci. 7, 166–172. doi: 10.1016/S1364-6613(03)00056-1

CrossRef Full Text | Google Scholar

Baddeley, A. (1996). Exploring the central executive. Q. J. Exp. Psychol. A 49, 5–28. doi: 10.1080/713755608

Baddeley, A. (2010). Working memory. Curr. Biol. 20, R136–R140. doi: 10.1016/j.cub.2009.12.014

Baddeley, A. (2012). Working memory: theories, models, and controversies. Annu. Rev. Psychol. 63, 1–29. doi: 10.1146/annurev-psych-120710-100422

Baddeley, A., and Hitch, G. (1974). Working memory. Psychol. Learn. Motiv. 8, 47–89. doi: 10.1016/j.cub.2009.12.014

Baddeley, A. D. (2000a). The episodic buffer : a new component of working memory? Trends Cogn. Sci. 4, 417–423. doi: 10.1016/S1364-6613(00)01538-2

Baddeley, A. D. (2000b). Short-Term and Working Memory. The Oxford Handbook of Memory. Oxford: Oxford University Press.

Google Scholar

Bailey, N. W., Rogasch, N. C., Hoy, K. E., Maller, J. J., Segrave, R. A., Sullivan, C. M., et al. (2017). Increased gamma connectivity during working memory retention following traumatic brain injury. Brain Inj. 31, 379–389. doi: 10.1080/02699052.2016.1239273

Balderston, N. L., Vytal, K. E., O’Connell, K., Torrisi, S., Letkiewicz, A., Ernst, M., et al. (2017). Anxiety patients show reduced working memory related dlPFC activation during safety and threat. Depress. Anxiety 34, 25–36. doi: 10.1002/da.22518

Barrouillet, P., Bernardin, S., and Camos, V. (2004). Time constraints and resource sharing in adults’ working memory spans. J. Exp. Psychol. Gen. 133, 83–100. doi: 10.1037/0096-3445.133.1.83

Barrouillet, P., and Camos, V. (2007). “The time-based resource-sharing model of working memory,” in The Cognitive Neuroscience of Working Memory , ed. N. Osaka (Oxford: Oxford University Press), 59–80. doi: 10.1093/acprof:oso/9780198570394.003.0004

Barrouillet, P., Gavens, N., Vergauwe, E., Gaillard, V., and Camos, V. (2009). Working memory span development: a time-based resource-sharing model account. Dev. Psychol. 45, 477–490. doi: 10.1037/a0014615

Bolkan, S. S., Stujenske, J. M., Parnaudeau, S., Spellman, T. J., Rauffenbart, C., Abbas, A. I., et al. (2017). Thalamic projections sustain prefrontal activity during working memory maintenance. Nat. Neurosci. 20, 987–996. doi: 10.1038/nn.4568

Borella, E., Carretti, B., Sciore, R., Capotosto, E., Taconnat, L., Cornoldi, C., et al. (2017). Training working memory in older adults: is there an advantage of using strategies? Psychol. Aging 32, 178–191. doi: 10.1037/pag0000155

Chein, J. M., Moore, A. B., and Conway, A. R. A. (2011). NeuroImage domain-general mechanisms of complex working memory span. Neuroimage 54, 550–559. doi: 10.1016/j.neuroimage.2010.07.067

Chen, C. J., Wu, C. H., Liao, Y. P., Hsu, H. L., Tseng, Y. C., Liu, H. L., et al. (2012). Working memory in patients with mild traumatic brain injury: functional MR imaging analysis. Radiology 264, 844–851. doi: 10.1148/radiol.12112154

Cowan, N. (1999). “An embedded-processes model of working memory,” in Models of Working Memory: Mechanisms of Active Maintenance and Executive Control , eds A. Miyake and P. Shah (Cambridge: Cambridge University Press). doi: 10.1017/S0140525X01003922

Cowan, N. (2005). Working memory capacity. Exp. Psychol. 54, 245–246. doi: 10.1027/1618-3169.54.3.245

Cowan, N. (2008). What are the differences between long-term, short-term, and working memory? Prog. Brain Res. 169, 323–338. doi: 10.1016/S0079-6123(07)00020-9

Cowan, N. (2010). The magical mystery four. Curr. Dir. Psychol. Sci. 19, 51–57. doi: 10.1177/0963721409359277

Dahlin, E., Nyberg, L., Bäckman, L., and Neely, A. S. (2008). Plasticity of executive functioning in young and older adults: immediate training gains, transfer, and long-term maintenance. Psychol. Aging 23, 720–730. doi: 10.1037/a0014296

Daneman, M., and Carpenter, P. A. (1980). Individual differences in working memory and reading. J. Verbal Learn. Verbal Behav. 19, 450–466. doi: 10.1016/S0022-5371(80)90312-6

D’Esposito, M., and Postle, B. R. (2015). The cognitive neuroscience of working memory. Annu. Rev. Psychol. 66, 115–142. doi: 10.1146/annurev-psych-010814-015031

Dikmen, S. S., Corrigan, J. D., Levin, H. S., Machamer, J., Stiers, W., and Weisskopf, M. G. (2009). Cognitive outcome following traumatic brain injury. J. Head Trauma Rehabil. 24, 430–438. doi: 10.1097/HTR.0b013e3181c133e9

Dima, D., Jogia, J., and Frangou, S. (2014). Dynamic causal modeling of load-dependent modulation of effective connectivity within the verbal working memory network. Hum. Brain Mapp. 35, 3025–3035. doi: 10.1002/hbm.22382

Dobryakova, E., Boukrina, O., and Wylie, G. R. (2015). Investigation of information flow during a novel working memory task in individuals with traumatic brain injury. Brain Connect. 5, 433–441. doi: 10.1089/brain.2014.0283

Duff, S. J., and Hampson, E. (2000). A beneficial effect of estrogen on working memory in postmenopausal women taking hormone replacement therapy. Horm. Behav. 38, 262–276. doi: 10.1006/hbeh.2000.1625

Dunning, D. L., Westgate, B., and Adlam, A.-L. R. (2016). A meta-analysis of working memory impairments in survivors of moderate-to-severe traumatic brain injury. Neuropsychology 30, 811–819. doi: 10.1037/neu0000285

Ellis, M. U., DeBoard Marion, S., McArthur, D. L., Babikian, T., Giza, C., Kernan, C. L., et al. (2016). The UCLA study of children with moderate-to-severe traumatic brain injury: event-related potential measure of interhemispheric transfer time. J. Neurotrauma 33, 990–996. doi: 10.1089/neu.2015.4023

Engle, R. W. (2002). Working memory capacity as executive attention. Curr. Dir. Psychol. Sci. 11, 19–23. doi: 10.1111/1467-8721.00160

Engle, R. W., and Kane, M. J. (2004). “Executive attention, working memory capacity, and a two-factor theory of cognitive control,” in The Psychology of Learning and Motivation: Advances in Research and Theory , ed. B. H. Ross (New York, NY: Elsevier), 145–199. doi: 10.1016/S0079-7421(03)44005-X

Farrer, T. J. (2017). Encyclopedia of Geropsychology , ed. N. A. Pachana. Singapore: Springer. doi: 10.1007/978-981-287-080-3

CrossRef Full Text

Fiebig, F., and Lansner, A. (2017). A spiking working memory model based on hebbian short-term potentiation. J. Neurosci. 37, 83–96. doi: 10.1523/JNEUROSCI.1989-16.2016

Friston, K., Moran, R., and Seth, A. K. (2013). Analysing connectivity with granger causality and dynamic causal modelling. Curr. Opin. Neurobiol. 23, 172–178. doi: 10.1016/j.conb.2012.11.010

Gorman, S., Barnes, M. A., Swank, P. R., Prasad, M., Cox, C. S., and Ewing-Cobbs, L. (2016). Does processing speed mediate the effect of pediatric traumatic brain injury on working memory? Neuropsychology 30, 263–273. doi: 10.1037/neu0000214

Gorman, S., Barnes, M. A., Swank, P. R., Prasad, M., and Ewing-Cobbs, L. (2012). The effects of pediatric traumatic brain injury on verbal and visual-spatial working memory. J. Int. Neuropsychol. Soc. 18, 29–38. doi: 10.1017/S1355617711001251

Gottwald, B., Wilde, B., Mihajlovic, Z., and Mehdorn, H. M. (2004). Evidence for distinct cognitive deficits after focal cerebellar lesions. J. Neurol. Neurosurg. Psychiatry 75, 1524–1531. doi: 10.1136/jnnp.2003.018093

Grot, S., Légaré, V. P., Lipp, O., Soulières, I., Dolcos, F., and Luck, D. (2017). Abnormal prefrontal and parietal activity linked to deficient active binding in working memory in schizophrenia. Schizophr. Res. 188, 68–74. doi: 10.1016/j.schres.2017.01.021

Guye, S., and von Bastian, C. C. (2017). Working memory training in older adults: bayesian evidence supporting the absence of transfer. Psychol. Aging 32, 732–746. doi: 10.1037/pag0000206

Haller, S., Montandon, M.-L., Rodriguez, C., Moser, D., Toma, S., Hofmeister, J., et al. (2017). Caffeine impact on working memory-related network activation patterns in early stages of cognitive decline. Neuroradiology 59, 387–395. doi: 10.1007/s00234-017-1803-5

Haller, S., Rodriguez, C., Moser, D., Toma, S., Hofmeister, J., Sinanaj, I., et al. (2013). Acute caffeine administration impact on working memory-related brain activation and functional connectivity in the elderly: a BOLD and perfusion MRI study. Neuroscience 250, 364–371. doi: 10.1016/j.neuroscience.2013.07.021

Hedden, T., and Gabrieli, J. D. E. (2004). Insights into the ageing mind: a view from cognitive neuroscience. Nat. Rev. Neurosci. 5, 87–96. doi: 10.1038/nrn1323

Heinzel, S., Rimpel, J., Stelzel, C., and Rapp, M. A. (2017). Transfer effects to a multimodal dual-task after working memory training and associated neural correlates in older adults – a pilot study. Front. Hum. Neurosci. 11:85. doi: 10.3389/fnhum.2017.00085

Hillary, F. G., Medaglia, J. D., Gates, K., Molenaar, P. C., Slocomb, J., Peechatka, A., et al. (2011). Examining working memory task acquisition in a disrupted neural network. Brain 134, 1555–1570. doi: 10.1093/brain/awr043

Hsu, H.-L., Chen, D. Y.-T., Tseng, Y.-C., Kuo, Y.-S., Huang, Y.-L., Chiu, W.-T., et al. (2015). Sex differences in working memory after mild traumatic brain injury: a functional MR imaging study. Radiology 276, 828–835. doi: 10.1148/radiol.2015142549

Humphreys, M. S., Bain, J. D., and Pike, R. (1989). Different ways to cue a coherent memory system: a theory for episodic, semantic, and procedural tasks. Psychol. Rev. 96, 208–233. doi: 10.1037/0033-295X.96.2.208

Janowsky, J. S., Chavez, B., and Orwoll, E. (2000). Sex steroids modify working memory. J. Cogn. Neurosci. 12, 407–414. doi: 10.1162/089892900562228

Jimura, K., Chushak, M. S., Westbrook, A., and Braver, T. S. (2017). Intertemporal decision-making involves prefrontal control mechanisms associated with working memory. Cereb. Cortex doi: 10.1093/cercor/bhx015 [Epub ahead of print].

Joseph, J. E., Swearingen, J. E., Corbly, C. R., Curry, T. E., and Kelly, T. H. (2012). Influence of estradiol on functional brain organization for working memory. Neuroimage 59, 2923–2931. doi: 10.1016/j.neuroimage.2011.09.067

Karbach, J., and Verhaeghen, P. (2014). Making working memory work: a meta-analysis of executive control and working memory training in younger and older adults. Psychol. Sci. 25, 2027–2037. doi: 10.1177/0956797614548725

Kim, C., Kroger, J. K., Calhoun, V. D., and Clark, V. P. (2015). The role of the frontopolar cortex in manipulation of integrated information in working memory. Neurosci. Lett. 595, 25–29. doi: 10.1016/j.neulet.2015.03.044

Klaassen, E. B., De Groot, R. H. M., Evers, E. A. T., Snel, J., Veerman, E. C. I., Ligtenberg, A. J. M., et al. (2013). The effect of caffeine on working memory load-related brain activation in middle-aged males. Neuropharmacology 64, 160–167. doi: 10.1016/j.neuropharm.2012.06.026

Kovacs, K., and Conway, A. R. A. (2016). Process overlap theory: a unified account of the general factor of intelligence. Psychol. Inq. 27, 151–177. doi: 10.1080/1047840X.2016.1153946

Le, T. M., Borghi, J. A., Kujawa, A. J., Klein, D. N., and Leung, H.-C. (2017). Alterations in visual cortical activation and connectivity with prefrontal cortex during working memory updating in major depressive disorder. Neuroimage 14, 43–53. doi: 10.1016/j.nicl.2017.01.004

Liu, Z.-X., Glizer, D., Tannock, R., and Woltering, S. (2016). EEG alpha power during maintenance of information in working memory in adults with ADHD and its plasticity due to working memory training: a randomized controlled trial. Clin. Neurophysiol. 127, 1307–1320. doi: 10.1016/j.clinph.2015.10.032

Ma, L., Steinberg, J. L., Hasan, K. M., Narayana, P. A., Kramer, L. A., and Moeller, F. G. (2012). Working memory load modulation of parieto-frontal connections: evidence from dynamic causal modeling. Hum. Brain Mapp. 33, 1850–1867. doi: 10.1002/hbm.21329

Maehler, C., and Schuchardt, K. (2016). Working memory in children with specific learning disorders and/or attention deficits. Learn. Individ. Differ. 49, 341–347. doi: 10.1016/j.lindif.2016.05.007

Mandalis, A., Kinsella, G., Ong, B., and Anderson, V. (2007). Working memory and new learning following pediatric traumatic brain injury. Dev. Neuropsychol. 32, 683–701. doi: 10.1080/87565640701376045

Manktelow, A. E., Menon, D. K., Sahakian, B. J., and Stamatakis, E. A. (2017). Working memory after traumatic brain injury: the neural basis of improved performance with methylphenidate. Front. Behav. Neurosci. 11:58. doi: 10.3389/fnbeh.2017.00058

Miller, G. A., Galanter, E., and Pribram, K. H. (1960). Plans and the Structure of Behavior. New York, NY: Henry Holt and Company. doi: 10.1037/10039-000

Miyake, A., and Shah, P. (eds). (1999). Models of Working Memory: Mechanisms of Active Maintenance and Executive Control. New York, NY: Cambridge University Press. doi: 10.1017/CBO9781139174909

Mongillo, G., Barak, O., and Tsodyks, M. (2008). Synaptic theory of working memory. Science 319, 1543–1546. doi: 10.1126/science.1150769

Moore, A. B., Li, Z., Tyner, C. E., Hu, X., and Crosson, B. (2013). Bilateral basal ganglia activity in verbal working memory. Brain Lang. 125, 316–323. doi: 10.1016/j.bandl.2012.05.003

Murty, V. P., Sambataro, F., Radulescu, E., Altamura, M., Iudicello, J., Zoltick, B., et al. (2011). Selective updating of working memory content modulates meso-cortico-striatal activity. Neuroimage 57, 1264–1272. doi: 10.1016/j.neuroimage.2011.05.006

Nadebaum, C., Anderson, V., and Catroppa, C. (2007). Executive function outcomes following traumatic brain injury in young children: a five year follow-up. Dev. Neuropsychol. 32, 703–728. doi: 10.1080/87565640701376086

Nadler, R. T., and Archibald, L. M. D. (2014). The assessment of verbal and visuospatial working memory with school age canadian children. Can. J. Speech Lang. Pathol. Audiol. 38, 262–279.

Nakagawa, S., Takeuchi, H., Taki, Y., Nouchi, R., Sekiguchi, A., Kotozaki, Y., et al. (2016). Sex-related differences in the effects of sleep habits on verbal and visuospatial working memory. Front. Psychol. 7:1128. doi: 10.3389/fpsyg.2016.01128

Nissim, N. R., O’Shea, A. M., Bryant, V., Porges, E. C., Cohen, R., and Woods, A. J. (2017). Frontal structural neural correlates of working memory performance in older adults. Front. Aging Neurosci. 8:328. doi: 10.3389/fnagi.2016.00328

Oren, N., Ash, E. L., Tarrasch, R., Hendler, T., Giladi, N., and Shapira-Lichter, I. (2017). Neural patterns underlying the effect of negative distractors on working memory in older adults. Neurobiol. Aging 53, 93–102. doi: 10.1016/j.neurobiolaging.2017.01.020

Osaka, M., Osaka, N., Kondo, H., Morishita, M., Fukuyama, H., Aso, T., et al. (2003). The neural basis of individual differences in working memory capacity: an fMRI study. Neuroimage 18, 789–797. doi: 10.1016/S1053-8119(02)00032-0

Owen, A. M., McMillan, K. M., Laird, A. R., and Bullmore, E. (2005). N-back working memory paradigm: a meta-analysis of normative functional neuroimaging studies. Hum. Brain Mapp. 25, 46–59. doi: 10.1002/hbm.20131

Owens, J. A., Spitz, G., Ponsford, J. L., Dymowski, A. R., Ferris, N., and Willmott, C. (2017). White matter integrity of the medial forebrain bundle and attention and working memory deficits following traumatic brain injury. Brain Behav. 7:e00608. doi: 10.1002/brb3.608

Perbal, S., Couillet, J., Azouvi, P., and Pouthas, V. (2003). Relationships between time estimation, memory, attention, and processing speed in patients with severe traumatic brain injury. Neuropsychologia 41, 1599–1610. doi: 10.1016/S0028-3932(03)00110-6

Perlstein, W. M., Cole, M. A., Demery, J. A., Seignourel, P. J., Dixit, N. K., Larson, M. J., et al. (2004). Parametric manipulation of working memory load in traumatic brain injury: behavioral and neural correlates. J. Int. Neuropsychol. Soc. 10, 724–741. doi: 10.1017/S1355617704105110

Pham, A. V., and Hasson, R. M. (2014). Verbal and visuospatial working memory as predictors of children’s reading ability. Arch. Clin. Neuropsychol. 29, 467–477. doi: 10.1093/arclin/acu024

Phillips, N. L., Parry, L., Mandalis, A., and Lah, S. (2017). Working memory outcomes following traumatic brain injury in children: a systematic review with meta-analysis. Child Neuropsychol. 23, 26–66. doi: 10.1080/09297049.2015.1085500

Rees, K., Allen, D., and Lader, M. (1999). The influences of age and caffeine on psychomotor and cognitive function. Psychopharmacology 145, 181–188. doi: 10.1007/s002130051047

Reuter-Lorenz, P. A., and Cappell, K. A. (2008). Neurocognitive ageing and the compensation hypothesis. Curr. Dir. Psychol. Sci. 17, 177–182. doi: 10.1111/j.1467-8721.2008.00570.x

Reuter-Lorenz, P. A., and Park, D. C. (2010). Human neuroscience and the aging mind : a new look at old problems. J. Gerontol. Psychol. Sci. 65, 405–415. doi: 10.1093/geronb/gbq035

Rieck, J. R., Rodrigue, K. M., Boylan, M. A., and Kennedy, K. M. (2017). Age-related reduction of BOLD modulation to cognitive difficulty predicts poorer task accuracy and poorer fluid reasoning ability. Neuroimage 147, 262–271. doi: 10.1016/j.neuroimage.2016.12.022

Rodriguez Merzagora, A. C., Izzetoglu, M., Onaral, B., and Schultheis, M. T. (2014). Verbal working memory impairments following traumatic brain injury: an fNIRS investigation. Brain Imaging Behav. 8, 446–459. doi: 10.1007/s11682-013-9258-8

Rose, N. S., LaRocque, J. J., Riggall, A. C., Gosseries, O., Starrett, M. J., Meyering, E. E., et al. (2016). Reactivation of latent working memories with transcranial magnetic stimulation. Science 354, 1136–1139. doi: 10.1126/science.aah7011

Rotzer, S., Loenneker, T., Kucian, K., Martin, E., Klaver, P., and von Aster, M. (2009). Dysfunctional neural network of spatial working memory contributes to developmental dyscalculia. Neuropsychologia 47, 2859–2865. doi: 10.1016/j.neuropsychologia.2009.06.009

Schneider-Garces, N. J., Gordon, B. A., Brumback-Peltz, C. R., Shin, E., Lee, Y., Sutton, B. P., et al. (2010). Span, CRUNCH, and beyond: working memory capacity and the aging brain. J. Cogn. Neurosci. 22, 655–669. doi: 10.1162/jocn.2009.21230

Schöning, S., Engelien, A., Kugel, H., Schäfer, S., Schiffbauer, H., Zwitserlood, P., et al. (2007). Functional anatomy of visuo-spatial working memory during mental rotation is influenced by sex, menstrual cycle, and sex steroid hormones. Neuropsychologia 45, 3203–3214. doi: 10.1016/j.neuropsychologia.2007.06.011

Silvanto, J. (2017). Working memory maintenance: sustained firing or synaptic mechanisms? Trends Cogn. Sci. 21, 152–154. doi: 10.1016/j.tics.2017.01.009

Stegmayer, K., Usher, J., Trost, S., Henseler, I., Tost, H., Rietschel, M., et al. (2015). Disturbed cortico–amygdalar functional connectivity as pathophysiological correlate of working memory deficits in bipolar affective disorder. Eur. Arch. Psychiatry Clin. Neurosci. 265, 303–311. doi: 10.1007/s00406-014-0517-5

Treble, A., Hasan, K. M., Iftikhar, A., Stuebing, K. K., Kramer, L. A., Cox, C. S., et al. (2013). Working memory and corpus callosum microstructural integrity after pediatric traumatic brain injury: a diffusion tensor tractography study. J. Neurotrauma 30, 1609–1619. doi: 10.1089/neu.2013.2934

Vallat-Azouvi, C., Weber, T., Legrand, L., and Azouvi, P. (2007). Working memory after severe traumatic brain injury. J. Int. Neuropsychol. Soc. 13, 770–780. doi: 10.1017/S1355617707070993

Vartanian, O., Jobidon, M.-E., Bouak, F., Nakashima, A., Smith, I., Lam, Q., et al. (2013). Working memory training is associated with lower prefrontal cortex activation in a divergent thinking task. Neuroscience 236, 186–194. doi: 10.1016/j.neuroscience.2012.12.060

Wang, S., and Gathercole, S. E. (2013). Working memory deficits in children with reading difficulties: memory span and dual task coordination. J. Exp. Child Psychol. 115, 188–197. doi: 10.1016/j.jecp.2012.11.015

Wylie, G. R., Freeman, K., Thomas, A., Shpaner, M., OKeefe, M., Watts, R., et al. (2015). Cognitive improvement after mild traumatic brain injury measured with functional neuroimaging during the acute period. PLoS One 10:e0126110. doi: 10.1371/journal.pone.0126110

Ziaei, M., Salami, A., and Persson, J. (2017). Age-related alterations in functional connectivity patterns during working memory encoding of emotional items. Neuropsychologia 94, 1–12. doi: 10.1016/j.neuropsychologia.2016.11.012

Ziemus, B., Baumann, O., Luerding, R., Schlosser, R., Schuierer, G., Bogdahn, U., et al. (2007). Impaired working-memory after cerebellar infarcts paralleled by changes in bold signal of a cortico-cerebellar circuit. Neuropsychologia 45, 2016–2024. doi: 10.1016/j.neuropsychologia.2007.02.012

Zylberberg, J., and Strowbridge, B. W. (2017). Mechanisms of persistent activity in cortical circuits: possible neural substrates for working memory. Annu. Rev. Neurosci. 40, 603–627. doi: 10.1146/annurev-neuro-070815-014006

Keywords : working memory, neuroscience, psychology, cognition, brain, central executive, prefrontal cortex, review

Citation: Chai WJ, Abd Hamid AI and Abdullah JM (2018) Working Memory From the Psychological and Neurosciences Perspectives: A Review. Front. Psychol. 9:401. doi: 10.3389/fpsyg.2018.00401

Received: 24 November 2017; Accepted: 09 March 2018; Published: 27 March 2018.

Reviewed by:

Copyright © 2018 Chai, Abd Hamid and Abdullah. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Aini Ismafairus Abd Hamid, [email protected]

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

Information

• Author Services

Initiatives

You are accessing a machine-readable page. In order to be human-readable, please install an RSS reader.

All articles published by MDPI are made immediately available worldwide under an open access license. No special permission is required to reuse all or part of the article published by MDPI, including figures and tables. For articles published under an open access Creative Common CC BY license, any part of the article may be reused without permission provided that the original article is clearly cited. For more information, please refer to https://www.mdpi.com/openaccess .

Feature papers represent the most advanced research with significant potential for high impact in the field. A Feature Paper should be a substantial original Article that involves several techniques or approaches, provides an outlook for future research directions and describes possible research applications.

Feature papers are submitted upon individual invitation or recommendation by the scientific editors and must receive positive feedback from the reviewers.

Editor’s Choice articles are based on recommendations by the scientific editors of MDPI journals from around the world. Editors select a small number of articles recently published in the journal that they believe will be particularly interesting to readers, or important in the respective research area. The aim is to provide a snapshot of some of the most exciting work published in the various research areas of the journal.

Original Submission Date Received: .

• Active Journals
• Find a Journal
• Proceedings Series
• For Authors
• For Reviewers
• For Editors
• For Librarians
• For Publishers
• For Societies
• For Conference Organizers
• Open Access Policy
• Institutional Open Access Program
• Special Issues Guidelines
• Editorial Process
• Research and Publication Ethics
• Article Processing Charges
• Testimonials
• Preprints.org
• SciProfiles
• Encyclopedia

Article Menu

• Subscribe SciFeed
• Recommended Articles
• Google Scholar
• on Google Scholar
• Table of Contents

Find support for a specific problem in the support section of our website.

Please let us know what you think of our products and services.

Visit our dedicated information section to learn more about MDPI.

JSmol Viewer

Innovative load forecasting models and intelligent control strategy for enhancing distributed load levelling techniques in resilient smart grids.

1. Introduction

2. related work, 3. materials and methods, 3.1. dataset collection, 3.1.1. data source, 3.1.2. feature descriptions, 3.2. data pre-processing, 3.2.1. data normalization, 3.2.2. min–max scaling, 3.3. proposed model, 3.3.1. model architectures, 3.3.2. predicting dynamic loads, 3.3.3. intelligent control strategy for load levelling, 3.4. mathematical model, 3.4.1. mathematical model for gated recurrent unit (gru).

• Mathematical Model of LSTM

3.4.2. Optimization Model

3.5. intelligent control strategy, 3.6. evaluation metrics, 3.6.1. mean squared error, 3.6.2. mean absolute percentage error, 4. results and discussion, 4.1. performance of lstm, 4.2. performance of gru, 4.3. impact of intelligent control strategy, 5. conclusions, author contributions, data availability statement, conflicts of interest.

• Ahmad, N.; Ghadi, Y.; Adnan, M.; Ali, M. Load Forecasting Techniques for Power System: Research Challenges and Survey. IEEE Access 2022 , 10 , 71054–71090. [ Google Scholar ] [ CrossRef ]
• Abdelaziz, A.Y.; Biswal, M. Load Forecasting Models in Smart Grid. Encyclopedia. Available online: https://encyclopedia.pub/entry/41526 (accessed on 27 July 2024).
• Khan, A.R.; Mahmood, A.; Safdar, A.; Khan, Z.A.; Khan, N.A. Load forecasting, dynamic pricing and DSM in smart grid: A review. Renew. Sustain. Energy Rev. 2016 , 54 , 1311–1322. [ Google Scholar ] [ CrossRef ]
• Kuster, C.; Rezgui, Y.; Mourshed, M. Electrical load forecasting models: A critical systematic review. Sustain. Cities Soc. 2017 , 35 , 257–270. [ Google Scholar ] [ CrossRef ]
• Salehimehr, S.; Taheri, B.; Sedighizadeh, M. Short-term load forecasting in smart grids using artificial intelligence methods: A survey. J. Eng. 2022 , 2022 , 1133–1142. [ Google Scholar ] [ CrossRef ]
• Wang, X.; Zhang, M.; Ren, F. Load forecasting in a smart grid through customer behaviour learning using L1-regularized continuous conditional random fields. In Proceedings of the 2016 International Conference on Autonomous Agents & Multiagent Systems, Singapore, 9–13 May 2016; pp. 817–825. [ Google Scholar ]
• Ferreira, A.; Leitão, P.; Barata, J. Prediction models for short-term load and production forecasting in smart electrical grids. In Proceedings of the Industrial Applications of Holonic and Multi-Agent Systems: 8th International Conference, HoloMAS 2017, Lyon, France, 28–30 August 2017; Proceedings 8. Springer: Berlin/Heidelberg, Germany, 2017; Volume 10444, pp. 186–199. [ Google Scholar ] [ CrossRef ]
• Ali, S. Short Term Load Forecasting using Smart Meter Data. In Proceedings of the Tenth ACM International Conference on Future Energy Systems, Phoenix, AZ, USA, 25–28 June 2019; pp. 419–421. [ Google Scholar ]
• Zheng, J.; Xu, C.; Zhang, Z.; Li, X. Electric load forecasting in smart grids using Long-Short-Term-Memory based Recurrent Neural Network. In Proceedings of the 2017 51st Annual Conference on Information Sciences and Systems (CISS), Baltimore, MD, USA, 22–24 March 2017. [ Google Scholar ] [ CrossRef ]
• Roy, K.; Ishmam, A.; Taher, K.A. Demand forecasting in smart grid using long short-term memory. In Proceedings of the 2021 International Conference on Automation, Control and Mechatronics for Industry 4.0 (ACMI), Rajshahi, Bangladesh, 8–9 July 2021; Volume 2021, pp. 8–9. [ Google Scholar ] [ CrossRef ]
• Luo, D.S.; Wang, X.K. Research on the framework of smart grid load forecasting platform based on cloud computing. Hunan Daxue Xuebao/J. Hunan Univ. Nat. Sci. 2016 , 43 , 101–108. [ Google Scholar ]
• Islam, B.U.; Rasheed, M.; Ahmed, S.F. Review of Short-Term Load Forecasting for Smart Grids Using Deep Neural Networks and Metaheuristic Methods. Math. Probl. Eng. 2022 , 2022 , 4049685. [ Google Scholar ] [ CrossRef ]
• Dhumale, S. Smart Grid experimentation Plan for Load Forecasting—A Case study of a large power distribution company. Proc. Int. Jt. J. Conf. Eng. Technol. 2010 , 3 , 78–83. [ Google Scholar ]
• Chemetova, S.; Santos, P.; Ventim-Neves, M. Load forecasting in electrical distribution grid of medium voltage. IFIP Adv. Inf. Commun. Technol. 2016 , 470 , 340–349. [ Google Scholar ] [ CrossRef ]
• Lee, D.; Park, J.; Kundur, D. Trends in short-term renewable and load forecasting for applications in smart grid. Lect. Notes Inst. Comput. Sci. Soc. Telecommun. Eng. 2016 , 166 , 292–300. [ Google Scholar ] [ CrossRef ]
• Ibrahim, B.; Rabelo, L.; Gutierrez-Franco, E.; Clavijo-Buritica, N. Machine Learning for Short-Term Load Forecasting in Smart Grids. Energies 2022 , 15 , 8079. [ Google Scholar ] [ CrossRef ]
• Habbak, H.; Mahmoud, M.; Metwally, K.; Mohamed, I. Smart Load Forecasting Techniques and Their Applications in Smart Grids Hany. Green Energy Technol. 2012 , 101 , 119–121. [ Google Scholar ]
• Zhang, D.; Jin, X.; Shi, P.; Chew, X.Y. Real-time load forecasting model for the smart grid using bayesian optimized CNN-BiLSTM. Front. Energy Res. 2023 , 11 , 1193662. [ Google Scholar ] [ CrossRef ]
• Vazquez, R.; Amaris, H.; Alonso, M.; Lopez, G.; Moreno, J.I.; Olmeda, D.; Coca, J. Assessment of an Adaptive Load Forecasting Methodology in a Smart Grid Demonstration Project. Energies 2017 , 10 , 190. [ Google Scholar ] [ CrossRef ]
• Ann, O.N.; Forest, R. Short Term Electricity Load Forecasting for Tumakuru City Based Short Term Electricity Load Forecasting for Tumakuru City Based on Ann, Lstm and Random Forest. Int. J. Creat. Res. Thoughts 2022 , 10 , 2023. [ Google Scholar ]
• Fotopoulou, M.; Pediaditis, P.; Skopetou, N.; Rakopoulos, D.; Christopoulos, S.; Kartalidis, A. A Review of the Energy Storage Systems of Non-Interconnected European Islands. Sustainability 2024 , 16 , 1572. [ Google Scholar ] [ CrossRef ]
• Esnaola-Gonzalez, I.; Jelić, M.; Pujić, D.; Diez, F.J.; Tomašević, N. An AI-Powered System for Residential Demand Response. Electronics 2021 , 10 , 693. [ Google Scholar ] [ CrossRef ]
• Rai, S.; De, M. Load forecasting using Two-level Heterogeneous Ensemble Method for Smart Metered Distribution System. Sci. Iran. 2023 . [ Google Scholar ] [ CrossRef ]
• Azeem, A.; Ismail, I.; Jameel, S.M.; Romlie, F.; Danyaro, K.U.; Shukla, S. Deterioration of Electrical Load Forecasting Models in a Smart Grid Environment. Sensors 2022 , 22 , 4363. [ Google Scholar ] [ CrossRef ]
• Madrid, E.A.; Antonio, N. Short-term electricity load forecasting with machine learning. Information 2021 , 12 , 50. [ Google Scholar ] [ CrossRef ]
• Patro, S.G.K.; Sahu, K.K. Normalization: A preprocessing stage. arXiv 2015 , arXiv:1503.06462. [ Google Scholar ] [ CrossRef ]
• Hochreiter, S.; Schmidhuber, J. Long short-term memory. Neural Comput. 1997 , 9 , 1735–1780. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Learning phrase representations using RNN encoder-decoder for statistical machine translation. arXiv 2014 , arXiv:1406.1078.
• Willmott, C.T.; Matsuura, K. Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance. Clim. Res. 2005 , 30 , 79–82. [ Google Scholar ] [ CrossRef ]
• Evaluating accuracy (or error) measures. Am. Stat. 1995 , 49 , 290–293. [ CrossRef ]

Click here to enlarge figure

Share and Cite

Fangzong, W.; Nishtar, Z. Innovative Load Forecasting Models and Intelligent Control Strategy for Enhancing Distributed Load Levelling Techniques in Resilient Smart Grids. Electronics 2024 , 13 , 3552. https://doi.org/10.3390/electronics13173552

Fangzong W, Nishtar Z. Innovative Load Forecasting Models and Intelligent Control Strategy for Enhancing Distributed Load Levelling Techniques in Resilient Smart Grids. Electronics . 2024; 13(17):3552. https://doi.org/10.3390/electronics13173552

Fangzong, Wang, and Zuhaib Nishtar. 2024. "Innovative Load Forecasting Models and Intelligent Control Strategy for Enhancing Distributed Load Levelling Techniques in Resilient Smart Grids" Electronics 13, no. 17: 3552. https://doi.org/10.3390/electronics13173552

Article Metrics

Article access statistics, further information, mdpi initiatives, follow mdpi.

Subscribe to receive issue release notifications and newsletters from MDPI journals

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Review Article
• Published: 30 March 2020

Memory devices and applications for in-memory computing

• Abu Sebastian   ORCID: orcid.org/0000-0001-5603-5243 1 ,
• Manuel Le Gallo   ORCID: orcid.org/0000-0003-1600-6151 1 ,
• Riduan Khaddam-Aljameh 1 &
• Evangelos Eleftheriou 1  

Nature Nanotechnology volume  15 ,  pages 529–544 ( 2020 ) Cite this article

51k Accesses

1106 Citations

98 Altmetric

Metrics details

• Electrical and electronic engineering
• Electronic devices
• Information storage

A Publisher Correction to this article was published on 16 July 2020

This article has been updated

Traditional von Neumann computing systems involve separate processing and memory units. However, data movement is costly in terms of time and energy and this problem is aggravated by the recent explosive growth in highly data-centric applications related to artificial intelligence. This calls for a radical departure from the traditional systems and one such non-von Neumann computational approach is in-memory computing. Hereby certain computational tasks are performed in place in the memory itself by exploiting the physical attributes of the memory devices. Both charge-based and resistance-based memory devices are being explored for in-memory computing. In this Review, we provide a broad overview of the key computational primitives enabled by these memory devices as well as their applications spanning scientific computing, signal processing, optimization, machine learning, deep learning and stochastic computing.

This is a preview of subscription content, access via your institution

Access options

Access Nature and 54 other Nature Portfolio journals

Get Nature+, our best-value online-access subscription

24,99 € / 30 days

cancel any time

Subscribe to this journal

Receive 12 print issues and online access

251,40 € per year

only 20,95 € per issue

Buy this article

• Purchase on SpringerLink
• Instant access to full article PDF

Prices may be subject to local taxes which are calculated during checkout

Similar content being viewed by others

A compute-in-memory chip based on resistive random-access memory

A 64-core mixed-signal in-memory compute chip based on phase-change memory for deep neural network inference

Self-rectifying resistive memory in passive crossbar arrays

Change history, 16 july 2020.

A Correction to this paper has been published: https://doi.org/10.1038/s41565-020-0756-8

Mutlu, O., Ghose, S., Gómez-Luna, J. & Ausavarungnirun, R. Processing data where it makes sense: Enabling in-memory computation. Microprocess. Microsyst. 67 , 28–41 (2019).

Google Scholar  

Horowitz, M. Computing’s energy problem (and what we can do about it). In Proc. International Solid-state Circuits Conference (ISSCC) 10–14 (IEEE, 2014).

Keckler, S. W., Dally, W. J., Khailany, B., Garland, M. & Glasco, D. GPUs and the future of parallel computing. IEEE Micro 31 , 7–17 (2011).

Jouppi, N. P. et al. In-datacenter performance analysis of a tensor processing unit. In Proc. International Symposium on Computer Architecture (ISCA) 1–12 (IEEE, 2017).

Sze, V., Chen, Y.-H., Yang, T.-J. & Emer, J. S. Efficient processing of deep neural networks: A tutorial and survey. Proc. IEEE 105 , 2295–2329 (2017).

Patterson, D. et al. A case for intelligent RAM. IEEE Micro 17 , 34–44 (1997).

Farooq, M. et al. 3D copper TSV integration, testing and reliability. In Proc. International Electron Devices Meeting 7–1 (IEEE, 2011).

Pawlowski, J. T. Hybrid memory cube (HMC). In Proceedings of the Hot Chips Symposium (HCS) 1–24 (IEEE, 2011).

Kim, J. & Kim, Y. HBM: Memory solution for bandwidth-hungry processors. In Proc. Hot Chips Symposium (HCS) 1–24 (IEEE, 2014).

Shulaker, M. M. et al. Three-dimensional integration of nanotechnologies for computing and data storage on a single chip. Nature 547 , 74 (2017).

CAS   Google Scholar  

Di Ventra, M. & Pershin, Y. V. The parallel approach. Nat. Phys. 9 , 200 (2013).

Indiveri, G. & Liu, S.-C. Memory and information processing in neuromorphic systems. Proc. The IEEE 103 , 1379–1397 (2015).

Zhirnov, V. V. & Marinella, M. J. in Emerging Nanoelectronic Devices (eds Chen, A.) Ch. 3 (Wiley Online Library, 2015).

Wong, H.-S. P. & Salahuddin, S. Memory leads the way to better computing. Nat. Nanotechnol. 10 , 191 (2015).

Chua, L. Resistance switching memories are memristors. Appl. Phys. A Mater. Sci. Process. 102 , 765–783 (2011).

Li, S. et al. DRISA: A DRAM-based reconfigurable in-situ accelerator. In Proc. International Symposium on Microarchitecture (MICRO) 288–301 (IEEE, 2017).

Seshadri, V. et al. Ambit: In-memory accelerator for bulk bitwise operations using commodity DRAM technology. In Proc. International Symposium on Microarchitecture 273–287 (IEEE, 2017).

Jeloka, S., Akesh, N. B., Sylvester, D. & Blaauw, D. A 28 nm configurable memory (TCAM/BCAM/SRAM) using push-rule 6T bit cell enabling logic-in-memory. IEEE J. Solid-State Circuits 51 , 1009–1021 (2016).

Aga, S. et al. Compute caches. In Proc. International Symposium on High Performance Computer Architecture (HPCA) 481–492 (IEEE, 2017).

Wang, J. et al. A compute SRAM with bit-serial integer/floating-point operations for programmable in-memory vector acceleration. In Proc. International Solid- State Circuits Conference (ISSCC) 224–226 (IEEE, 2019).

Jiang, Z., Yin, S., Seok, M. & Seo, J. XNOR-SRAM: In-memory computing SRAM macro for binary/ternary deep neural networks. In Proc. Symposium on VLSI Technology 173–174 (IEEE, 2018).

Biswas, A. & Chandrakasan, A. P. CONV-SRAM: an energy-efficient SRAM with in-memory dot-product computation for low-power convolutional neural networks. IEEE J. Solid-State Circuits 54 , 217–230 (2019).

Valavi, H., Ramadge, P. J., Nestler, E. & Verma, N. A 64-tile 2.4-Mb in-memory-computing CNN accelerator employing charge-domain compute. IEEE J. Solid-State Circuits 54 , 1789–1799 (2019).

Verma, N. et al. In-memory computing: Advances and prospects. IEEE J. Solid-State Circuits 11 , 43–55 (2019).

Gonugondla, S. K., Kang, M. & Shanbhag, N. R. A variation-tolerant in-memory machine learning classifier via on-chip training. IEEE J. Solid-State Circuits 53 , 3163–3173 (2018).

Bankman, D., Yang, L., Moons, B., Verhelst, M. & Murmann, B. An always-on 3.8 μ J/86% CIFAR-10 mixed-signal binary CNN processor with all memory on chip in 28-nm CMOS. IEEE J. Solid-State Circuits 54 , 158–172 (2019).

Diorio, C., Hasler, P., Minch, A. & Mead, C. A. A single-transistor silicon synapse. IEEE Transactions on Electron Devices 43 , 1972–1980 (1996).

Merrikh-Bayat, F. et al. High-performance mixed-signal neurocomputing with nanoscale floating-gate memory cell arrays. IEEE Trans. Neural Netw. Learn. Syst. 29 , 4782–4790 (2018).

Wang, P. et al. Three-dimensional NAND flash for vector-matrix multiplication. EEE Trans. Very Large Scale Integr. VLSI Syst. 27 , 988–991 (2019).

Burr, G. W. et al. Access devices for 3D crosspoint memory. J. Vac. Sci. Technol. B Nanotechnol. Microelectron. 32 , 040802 (2014).

Hickmott, T. Low-frequency negative resistance in thin anodic oxide films. J. Appl. Phys. 33 , 2669–2682 (1962).

Beck, A., Bednorz, J., Gerber, C., Rossel, C. & Widmer, D. Reproducible switching effect in thin oxide films for memory applications. Applied Physics Letters 77 , 139–141 (2000).

Waser, R. & Aono, M. Nanoionics-based resistive switching memories. Nat. Mater. 6 , 833–840 (2007).

Strukov, D. B., Snider, G. S., Stewart, D. R. & Williams, R. S. The missing memristor found. Nature 453 , 80 (2008).

Ovshinsky, S. R. Reversible electrical switching phenomena in disordered structures. Phys. Rev. Lett. 21 , 1450 (1968).

Wong, H.-S. P. et al. Phase change memory. Proc. IEEE 98 , 2201–2227 (2010).

Burr, G. W. et al. Recent progress in phase-change memory technology. IEEE J. Emerg. Sel. Top. Circuits Syst. 6 , 146–162 (2016).

Khvalkovskiy, A. et al. Basic principles of STT-MRAM cell operation in memory arrays. J. Phys. D Appl. Phys. 46 , 074001 (2013).

Kent, A. D. & Worledge, D. C. A new spin on magnetic memories. Nat. Nanotechnol. 10 , 187 (2015).

Vourkas, I. & Sirakoulis, G. C. Emerging memristor-based logic circuit design approaches: A review. IEEE Circuits and Systems Magazine 16 , 15–30 (2016).

Borghetti, J. et al. Memristive switches enable stateful logic operations via material implication. Nature 464 , 873 (2010).

Linn, E., Rosezin, R., Tappertzhofen, S., Böttger, U. & Waser, R. Beyond von neumann-logic operations in passive crossbar arrays alongside memory operations. Nanotechnology 23 , 305205 (2012).

Jeong, D. S., Kim, K. M., Kim, S., Choi, B. J. & Hwang, C. S. Memristors for energy-efficient new computing paradigms. Adv. Electron. Mater. 2 , 1600090 (2016).

Kvatinsky, S. et al. MAGIC-memristor-aided logic IEEE Trans . Circuits Syst. II Express Briefs 61 , 895–899 (2014).

Mahmoudi, H., Windbacher, T., Sverdlov, V. & Selberherr, S. Implication logic gates using spin-transfer-torque-operated magnetic tunnel junctions for intrinsic logic-in-memory. Solid State Electron. 84 , 191–197 (2013).

Kim, K. M. et al. Single-cell stateful logic using a dual-bit memristor. Phys. Status Solidi Rapid Res. Lett. 13 , 1800629 (2019).

Xu, N., Fang, L., Kim, K. M. & Hwang, C. S. Time-efficient stateful dual-bit-memristor logic. Phys. Status Solidi Rapid Res. Lett. 13 , 1900033 (2019).

Li, S. et al. Pinatubo: A processing-in-memory architecture for bulk bitwise operations in emerging non-volatile memories. In Proc. The Design Automation Conference (DAC) 173 (ACM, 2016).

Xie, L. et al. Scouting logic: A novel memristor-based logic design for resistive computing. In Proc. The IEEE Symposium on VLSI (ISVLSI) 176–181 (IEEE, 2017).

Maan, A. K., Jayadevi, D. A. & James, A. P. A survey of memristive threshold logic circuits. IEEE Trans. Neural Netw. Learn. Syst. 28 , 1734–1746 (2016).

Burr, G. W. et al. Neuromorphic computing using non-volatile memory. Adv Phys X 2 , 89–124 (2017).

Ielmini, D. & Wong, H.-S. P. In-memory computing with resistive switching devices. Nat. Electron. 1 , 333 (2018).

Wang, Z. et al. Resistive switching materials for information processing. Nat. Rev. Mater. https://doi.org/10.1038/s41578-019-0159-3 (2020).

Article   Google Scholar  

Wright, C. D., Hosseini, P. & Diosdado, J. A. V. Beyond von-neumann computing with nanoscale phase-change memory devices. Adv. Funct. Mater. 23 , 2248–2254 (2013).

Sebastian, A. et al. Brain-inspired computing using phase-change memory devices. J. Appl. Phys. 124 , 111101 (2018).

Godse, A. P. & Godse, D. A. Computer Organization and Architecture (Technical Publications, 2008).

Bojnordi, M. N. & Ipek, E. Memristive boltzmann machine: A hardware accelerator for combinatorial optimization and deep learning. In Proc. The International Symposium on High Performance Computer Architecture (HPCA) 1–13 (IEEE, 2016).

Shafiee, A. et al. ISAAC: A convolutional neural network accelerator with in-situ analog arithmetic in crossbars. Comput. Archit. News 44 , 14–26 (2016).

Chi, P. et al. PRIME: A novel processing-in-memory architecture for neural network computation in ReRAM-based main memory. In Proc. 43rd Annual International Symposium on Computer Architecture (ISCA) News 27–39 (IEEE, 2016).

Song, L., Qian, X., Li, H. & Chen, Y. PIPELAYER: A pipelined ReRAM-based accelerator for deep learning. In Proc. The International Symposium on High Performance Computer Architecture (HPCA) , 541–552 (IEEE, 2017).

Zidan, M. A. et al. A general memristor-based partial differential equation solver. Nat. Electron. 1 , 411 (2018).

Higham, N. J. Accuracy and Stability of Numerical Algorithms , Vol. 80 (Society for Industrial and Applied Mathematics, 2002).

Bekas, C., Curioni, A. & Fedulova, I. Low cost high performance uncertainty quantification. In Proc. 2nd Workshop on High Performance Computational Finance 1–8 (ACM, 2009).

Le Gallo, M. et al. Mixed-precision in-memory computing. Nat. Electron. 1 , 246–253 (2018).

Liu, S., Wang, Y., Fardad, M. & Varshney, P. K. A memristor-based optimization framework for artificial intelligence applications. IEEE Circuits and Systems Magazine 18 , 29–44 (2018).

Sun, Z. et al. Solving matrix equations in one step with cross-point resistive arrays. Proc. Natl. Acad. Sci. USA 116 , 4123–4128 (2019).

Sturges, R. H. Analog matrix inversion (robot kinematics). IEEE Journal on Robotics and Automation 4 , 157–162 (1988).

Feinberg, B., Wang, S. & Ipek, E. Making memristive neural network accelerators reliable. In Proc. The International Symposium on High Performance Computer Architecture (HPCA) 52–65 (IEEE, 2018).

Feinberg, B., Vengalam, U. K. R., Whitehair, N., Wang, S. & Ipek, E. Enabling scientific computing on memristive accelerators. In Proc. International Symposium on Computer Architecture (ISCA) 367–382 (IEEE, 2018).

Li, C. et al. Analogue signal and image processing with large memristor crossbars. Nat. Electron. 1 , 52–59 (2018).

Le Gallo, M., Sebastian, A., Cherubini, G., Giefers, H. & Eleftheriou, E. Compressed sensing with approximate message passing using in-memory computing. IEEE Trans. Electron Devices 65 , 4304–4312 (2018).

Cai, F. et al. Harnessing intrinsic noise in memristor hopfield neural networks for combinatorial optimization. Preprint at https://arxiv.org/abs/1903.11194 (2019).

Mostafa, H., Müller, L. K. & Indiveri, G. An event-based architecture for solving constraint satisfaction problems. Nat. Commun. 6 , 8941 (2015).

Parihar, A., Shukla, N., Jerry, M., Datta, S. & Raychowdhury, A. Vertex coloring of graphs via phase dynamics of coupled oscillatory networks. Sci. Rep. 7 , 911 (2017).

Kumar, S., Strachan, J. P. & Williams, R. S. Chaotic dynamics in nanoscale NbO 2 Mott memristors for analogue computing. Nature 548 , 318 (2017).

Torrejon, J. et al. Neuromorphic computing with nanoscale spintronic oscillators. Nature 547 , 428–431 (2017).

Seo, J. et al. On-chip sparse learning acceleration with CMOS and resistive synaptic devices. IEEE Trans. Nanotechnol. 14 , 969–979 (2015).

Sheridan, P. M. et al. Sparse coding with memristor networks. Nat. Nanotechnol. 12 , 784–789 (2017).

Sheridan, P. M., Du, C. & Lu, W. D. Feature extraction using memristor networks. IEEE Trans. Neural Netw. Learn. Syst. 27 , 2327–2336 (2016).

Choi, S., Sheridan, P. & Lu, W. D. Data clustering using memristor networks. Sci. Rep. 5 , 10492 (2015).

Karam, R., Puri, R., Ghosh, S. & Bhunia, S. Emerging trends in design and applications of memory-based computing and content-addressable memories. Proc. IEEE 103 , 1311–1330 (2015).

Rahimi, A. et al. High-dimensional computing as a nanoscalable paradigm. IEEE Trans. Circuits Syst. I Regul. Pap. 64 , 2508–2521 (2017).

Wu, T. F. et al. Hyperdimensional computing exploiting carbon nanotube FETs, resistive RAM, and their monolithic 3D integration. IEEE J. Solid-State Circuits 53 , 3183–3196 (2018).

Graves, A. et al. Hybrid computing using a neural network with dynamic external memory. Nature 538 , 471 (2016).

Ni, K. et al. Ferroelectric ternary content-addressable memory for one-shot learning. Nat. Electron. 2 , 521–529 (2019).

Eryilmaz, S. B. et al. Brain-like associative learning using a nanoscale non-volatile phase change synaptic device array. Front. Neurosci. 8 , 205 (2014).

Hu, S. et al. Associative memory realized by a reconfigurable memristive Hopfield neural network. Nat. Commun. 6 , 7522 (2015).

Kavehei, O. et al. An associative capacitive network based on nanoscale complementary resistive switches for memory-intensive computing. Nanoscale 5 , 5119–5128 (2013).

Du, C. et al. Reservoir computing using dynamic memristors for temporal information processing. Nat. Commun. 8 , 2204 (2017).

Sebastian, A. et al. Temporal correlation detection using computational phase-change memory. Nat. Commun. 8 , 1115 (2017).

LeCun, Y., Bengio, Y. & Hinton, G. Deep learning. Nature 521 , 436 (2015).

He, K., Zhang, X., Ren, S. & Sun, J. Deep residual learning for image recognition. In Proc. Conference on Computer Vision and Pattern Recognition (CVPR) 770–778 (IEEE, 2016).

LeCun, Y. Deep learning hardware: Past, present, and future. In Proc. International Solid-State Circuits Conference (ISSCC) 12–19 (IEEE, 2019).

Chen, Y., Yang, T., Emer, J. & Sze, V. Eyeriss v2: A flexible accelerator for emerging deep neural networks on mobile devices. IEEE J. Em. Sel. Top. C 9 , 292–308 (2019).

Dazzi, M. et al. 5 parallel prism: A topology for pipelined implementations of convolutional neural networks using computational memory. In Proc. NeurIPS MLSys Workshop (NeurIPS, 2019); http://learningsys.org/neurips19/acceptedpapers.html

Jia, Z., Maggioni, M., Smith, J. & Scarpazza, D. P. Dissecting the NVidia Turing T4 GPU via microbenchmarking. Preprint at https://arxiv.org/abs/1903.07486 (2019).

Hubara, I., Courbariaux, M., Soudry, D., El-Yaniv, R. & Bengio, Y. Quantized neural networks: Training neural networks with low precision weights and activations. J. Mach. Learn. Res. 18 , 6869–6898 (2017).

Xue, C. et al. 24.1 a 1mb multibit ReRAM computing-in-memory macro with 14.6ns parallel MAC computing time for CNN based AI edge processors. In Proc. The International Solid-State Circuits Conference (ISSCC) 388–390 (IEEE, 2019).

Hu, M. et al. Memristor-based analog computation and neural network classification with a dot product engine. Advanced Materials 30 , 1705914 (2018).

Yao, P. et al. Fully hardware-implemented memristor convolutional neural network. Nature 577 , 641–646 (2020).

Suri, M. et al. Phase change memory as synapse for ultra-dense neuromorphic systems: Application to complex visual pattern extraction. In Proc. The International Electron Devices Meeting (IEDM) 4.4.1–44.4 (IEEE, 2011).

Chen, W.-H. et al. CMOS-integrated memristive non-volatile computing-in-memory for AI edge processors. Nat. Electron. 2 , 420–428 (2019).

Murray, A. F. & Edwards, P. J. Enhanced mlp performance and fault tolerance resulting from synaptic weight noise during training. IEEE T. Neural Networ. 5 , 792–802 (1994).

Liu, B. et al. Vortex: Variation-aware training for memristor X-bar. In Proc. The Design Automation Conference (DAC) 1–6 (DAC, 2015).

Sebastian, A. et al. Computational memory-based inference and training of deep neural networks. In Proc. The Symposium on VLSI Technology T168–T169 (IEEE, 2019).

Gokmen, T., Onen, M. & Haensch, W. Training deep convolutional neural networks with resistive cross-point devices. Front. Neurosci. 11 , 538 (2017).

Alibart, F., Zamanidoost, E. & Strukov, D. B. Pattern classification by memristive crossbar circuits using ex situ and in situ training. Nat. Commun. 4 , 2072 (2013).

Burr, G. W. et al. Experimental demonstration and tolerancing of a large-scale neural network (165 000 synapses) using phase-change memory as the synaptic weight element. IEEE T. Electron Dev. 62 , 3498–3507 (2015).

Gokmen, T. & Vlasov, Y. Acceleration of deep neural network training with resistive cross-point devices: design considerations. Front. Neurosci. 10 , 333 (2016).

Agarwal, S. et al. Achieving ideal accuracies in analog neuromorphic computing using periodic carry. In Proc. The Symposium on VLSI Technology T174–T175 (IEEE, 2017).

Ambrogio, S. et al. Equivalent-accuracy accelerated neural-network training using analogue memory. Nature 558 , 60–67 (2018).

Yu, S. Neuro-inspired computing with emerging nonvolatile memory. Proc. IEEE 106 , 260–285 (2018).

Prezioso, M. et al. Training and operation of an integrated neuromorphic network based on metal-oxide memristors. Nature 521 , 61–64 (2015).

Yao, P. et al. Face classification using electronic synapses. Nat. Commun. 8 , 15199 (2017).

Li, C. et al. Efficient and self-adaptive in-situ learning in multilayer memristor neural networks. Nat. Commun. 9 , 2385 (2018).

Nandakumar, S. et al. Mixed-precision architecture based on computational memory for training deep neural networks. In Proc. The International Symposium on Circuits and Systems (ISCAS) 1–5 (IEEE, 2018).

Pfeiffer, M. & Pfeil, T. Deep learning with spiking neurons: Opportunities and challenges. Front. Neurosci. 12 , 774 (2018).

Diehl, P. U. et al. Fast-classifying, high-accuracy spiking deep networks through weight and threshold balancing. In Proc. International Joint Conference on Neural Networks (IJCNN) 1–8 (IEEE, 2015).

Sengupta, A., Ye, Y., Wang, R., Liu, C. & Roy, K. Going deeper in spiking neural networks: VGG and residual architectures. Front. Neurosci. 13 , 95 (2019).

Esser, S. K., Appuswamy, R., Merolla,P., Arthur, J. V. & Modha, D. S. Backpropagation for energy-efficient neuromorphic computing. In Proc. Advances in Neural Information Processing Systems (Eds. Cortes, C. et al) 1117–1125 (NIPS, 2015).

Lee, J. H., Delbruck, T. & Pfeiffer, M. Training deep spiking neural networks using backpropagation. Front. Neurosci. 10 , 508 (2016).

Woźniak, S., Pantazi, A. & Eleftheriou, E. Deep networks incorporating spiking neural dynamics. Preprint at https://arxiv.org/abs/1812.07040 (2018).

Benjamin, B. V. et al. Neurogrid: A mixed-analog-digital multichip system for large-scale neural simulations. Proc. IEEE 102 , 699–716 (2014).

Qiao, N. et al. A reconfigurable on-line learning spiking neuromorphic processor comprising 256 neurons and 128k synapses. Front. Neurosci. 9 , 141 (2015).

Kuzum, D., Jeyasingh, R. G., Lee, B. & Wong, H.-S. P. Nanoelectronic programmable synapses based on phase change materials for brain-inspired computing. Nano Letters 12 , 2179–2186 (2011).

Kim, S. et al. NVM neuromorphic core with 64k-cell (256-by-256) phase change memory synaptic array with on-chip neuron circuits for continuous in-situ learning. In Proc. The International Electron Devices Meeting (IEDM) 17–1 (IEEE, 2015).

Tuma, T., Le Gallo, M., Sebastian, A. & Eleftheriou, E. Detecting correlations using phase-change neurons and synapses. IEEE Electr. Device L. 37 , 1238–1241 (2016).

Pantazi, A., Woźniak, S., Tuma, T. & Eleftheriou, E. All-memristive neuromorphic computing with level-tuned neurons. Nanotechnology 27 , 355205 (2016).

Covi, E. et al. Analog memristive synapse in spiking networks implementing unsupervised learning. Front. Neurosci. 10 , 482 (2016).

Serb, A. et al. Unsupervised learning in probabilistic neural networks with multi-state metal-oxide memristive synapses. Nat. Commun. 7 , 12611 (2016).

Kheradpisheh, S. R., Ganjtabesh, M., Thorpe, S. J. & Masquelier, T. STDP-based spiking deep convolutional neural networks for object recognition. Neural Networks 99 , 56–67 (2018).

Moraitis, T., Sebastian, A. & Eleftheriou, E. The role of short-term plasticity in neuromorphic learning: Learning from the timing of rate-varying events with fatiguing spike-timing-dependent plasticity. IEEE Nanotechnology Magazine 12 , 45–53 (2018).

Wang, Z. et al. Memristors with diffusive dynamics as synaptic emulators for neuromorphic computing. Nat. Mater. 16 , 101 (2017).

Carboni, R. & Ielmini, D. Stochastic memory devices for security and computing. Adv. Electron. Mater . 1900198 (2019).

Jo, S. H., Kim, K.-H. & Lu, W. Programmable resistance switching in nanoscale two-terminal devices. Nano letters 9 , 496–500 (2008).

Le Gallo, M., Athmanathan, A., Krebs, D. & Sebastian, A. Evidence for thermally assisted threshold switching behavior in nanoscale phase-change memory cells. J. Appl. Phys. 119 , 025704 (2016).

Le Gallo, M., Tuma, T., Zipoli, F., Sebastian, A. & Eleftheriou, E. Inherent stochasticity in phase-change memory devices. In Proc. 2016 46th European Solid-State Device Research Conference (ESSDERC) 373–376 (IEEE, 2016).

Alaghi, A. & Hayes, J. P. Survey of stochastic computing. ACM T Embed. Comput. S. 12 , 92 (2013).

Gupta, S., Agrawal, A., Gopalakrishnan, K. & Narayanan, P. Deep learning with limited numerical precision. In Proc. International Conference on Machine Learning 1737–1746 (2015).

Yang, K. et al. 16.3 a 23mb/s 23pj/b fully synthesized true-random-number generator in 28nm and 65nm CMOS. In Proc. Proceedings of the International Solid-State Circuits Conference (ISSCC) 280–281 (IEEE, 2014).

Jiang, H. et al. A novel true random number generator based on a stochastic diffusive memristor. Nat. Commun. 8 , 882 (2017).

Gaba, S., Sheridan, P., Zhou, J., Choi, S. & Lu, W. Stochastic memristive devices for computing and neuromorphic applications. Nanoscale 5 , 5872–5878 (2013).

Balatti, S. et al. Physical unbiased generation of random numbers with coupled resistive switching devices. IEEE T. Electron Dev. 63 , 2029–2035 (2016).

Choi, W. H. et al. A magnetic tunnel junction based true random number generator with conditional perturb and real-time output probability tracking. In Proc. The International Electron Devices Meeting 12–5 (IEEE, 2014).

Carboni, R. et al. Random number generation by differential read of stochastic switching in spin-transfer torque memory. IEEE Electr. Device L. 39 , 951–954 (2018).

Shim, Y., Chen, S., Sengupta, A. & Roy, K. Stochastic spin-orbit torque devices as elements for bayesian inference. Sci. Rep. 7 , 14101 (2017).

Tuma, T., Pantazi, A., Le Gallo, M., Sebastian, A. & Eleftheriou, E. Stochastic phase-change neurons. Nat. Nanotechnol. 11 , 693–699 (2016).

Mizrahi, A. et al. Neural-like computing with populations of superparamagnetic basis functions. Nat. Commun. 9 , 1533 (2018).

Bichler, O. et al. Visual pattern extraction using energy-efficient 2-PCM synapse neuromorphic architecture. IEEE T. Electron Dev. 59 , 2206–2214 (2012).

Holcomb, D. E., Burleson, W. P. & Fu, K. Power-up SRAM state as an identifying fingerprint and source of true random numbers. IEEE T. Comput. 58 , 1198–1210 (2009).

Gao, L., Chen, P.-Y., Liu, R. & Yu, S. Physical unclonable function exploiting sneak paths in resistive cross-point array. IEEE Transactions on Electron Devices 63 , 3109–3115 (2016).

Nili, H. et al. Hardware-intrinsic security primitives enabled by analogue state and nonlinear conductance variations in integrated memristors. Nat. Electron. 1 , 197 (2018).

Jiang, H. et al. A provable key destruction scheme based on memristive crossbar arrays. Nat. Electron. 1 , 548 (2018).

Talati, N., Gupta, S., Mane, P. & Kvatinsky, S. Logic design within memristive memories using memristor-aided logic (MAGIC). IEEE T. Nanotechnol. 15 , 635–650 (2016).

Cheng, L. et al. Functional demonstration of a memristive arithmetic logic unit (MemALU) for in-memory computing. Adv. Funct. Mater . (2019).

Haj-Ali, A., Ben-Hur, R., Wald, N., Ronen, R. & Kvatinsky, S. IMAGING: In-memory algorithms for image processing. IEEE T. Circuits Systems-I 65 , 4258–4271 (2018).

Hamdioui, S. et al. Applications of computation-in-memory architectures based on memristive devices. In Proc. The Design, Automation & Test in Europe Conference & Exhibition (DATE) 486–491 (IEEE, 2019).

Xiong, F., Liao, A. D., Estrada, D. & Pop, E. Low-power switching of phase-change materials with carbon nanotube electrodes. Science 332 , 568–570 (2011).

Li, K.-S. et al. Utilizing sub-5 nm sidewall electrode technology for atomic-scale resistive memory fabrication. In Proc. Symposium on VLSI Technology 1–2 (IEEE, 2014).

Salinga, M. et al. Monatomic phase change memory. Nat. Mater. 17 , 681–685 (2018).

Pi, S. et al. Memristor crossbar arrays with 6-nm half-pitch and 2-nm critical dimension. Nat. Nanotechnol. 14 , 35 (2019).

Brivio, S., Frascaroli, J. & Spiga, S. Role of Al doping in the filament disruption in HfO 2 resistance switches. Nanotechnology 28 , 395202 (2017).

Choi, S. et al. SiGe epitaxial memory for neuromorphic computing with reproducible high performance based on engineered dislocations. Nat. Mater. 17 , 335 (2018).

Boybat, I. et al. Neuromorphic computing with multi-memristive synapses. Nat. Commun. 9 , 2514 (2018).

Koelmans, W. W. et al. Projected phase-change memory devices. Nat. Commun. 6 , 8181 (2015).

Giannopoulos, I. et al. 8-bit precision in-memory multiplication with projected phase-change memory. In Proc. The International Electron Devices Meeting (IEDM) 27–7 (IEEE, 2018).

Chen, Y. et al. DaDianNao: A machine-learning supercomputer. In Proc. The 47th Annual IEEE/ACM International Symposium on Microarchitecture 609–622 (IEEE Computer Society, 2014).

Ankit, A. et al. PUMA: A programmable ultra-efficient memristor-based accelerator for machine learning inference. In Proc. The International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) , 715–73 (ACM, 2019).

Eleftheriou, E. et al. Deep learning acceleration based on in-memory computing. IBM Journal of Research and Development (2019).

Yoon, K. J., Bae, W., Jeong, D.-K. & Hwang, C. S. Comprehensive writing margin analysis and its application to stacked one diode-one memory device for high-density crossbar resistance switching random access memory. Adv. Electron. Mater. 2 , 1600326 (2016).

Le Gallo, M., Sebastian, A., Cherubini, G., Giefers, H. & Eleftheriou, E. Compressed sensing recovery using computational memory. In Proc. The International Electron Devices Meeting (IEDM) 28–3 (IEEE, 2017).

van de Burgt, Y. et al. A non-volatile organic electrochemical device as a low-voltage artificial synapse for neuromorphic computing. Nat. Mater. 16 , 414 (2017).

Tang, J. et al. ECRAM as scalable synaptic cell for high-speed, low-power neuromorphic computing. In Proc. The International Electron Devices Meeting (IEDM) 13–1 (IEEE, 2018).

Fuller, E. J. et al. Parallel programming of an ionic floating-gate memory array for scalable neuromorphic computing. Science 364 , 570–574 (2019).

Kimura, H. et al. Complementary ferroelectric-capacitor logic for low-power logic-in-memory VLSI. IEEE Journal of Solid-State Circuits 39 , 919–926 (2004).

Aziz, A. et al . Computing with ferroelectric FETs: Devices, models, systems, and applications. In Proc. The Design, Automation & Test in Europe Conference & Exhibition (DATE) 1289–1298 (IEEE, 2018).

Chanthbouala, A. et al. A ferroelectric memristor. Nat. Mater. 11 , 860 (2012).

Ríos, C. et al. Integrated all-photonic non-volatile multi-level memory. Nat. Photon. 9 , 725 (2015).

Wuttig, M., Bhaskaran, H. & Taubner, T. Phase-change materials for non-volatile photonic applications. Nat. Photon. 11 , 465 (2017).

Ríos, C. et al. In-memory computing on a photonic platform. Sci. Adv. 5 , eaau5759 (2019).

Download references

Acknowledgements

We would like to thank T. Tuma for technical discussions and assistance with scientific illustrations, G. Sarwat and I. Boybat for critical review of the manuscript, and L. Rudin and N. Gustafsson for editorial help. A.S. acknowledges funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement number 682675).

Author information

Authors and affiliations.

IBM Research – Zurich, Rüschlikon, Switzerland

Abu Sebastian, Manuel Le Gallo, Riduan Khaddam-Aljameh & Evangelos Eleftheriou

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Abu Sebastian .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Peer review information Nature Nanotechnology thanks Cheol Seong Hwang and the other, anonymous, reviewers for their contribution to the peer review of this work.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Cite this article.

Sebastian, A., Le Gallo, M., Khaddam-Aljameh, R. et al. Memory devices and applications for in-memory computing. Nat. Nanotechnol. 15 , 529–544 (2020). https://doi.org/10.1038/s41565-020-0655-z

Download citation

Received : 16 September 2019

Accepted : 10 February 2020

Published : 30 March 2020

Issue Date : 02 July 2020

DOI : https://doi.org/10.1038/s41565-020-0655-z

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

This article is cited by

Ferroelectric compute-in-memory annealer for combinatorial optimization problems.

• Xunzhao Yin

Nature Communications (2024)

Binarized neural network of diode array with high concordance to vector–matrix multiplication

• Yunwoo Shin
• Kyoungah Cho
• Sangsig Kim

Scientific Reports (2024)

High-throughput screening to identify two-dimensional layered phase-change chalcogenides for embedded memory applications

• Xiaozhe Wang

npj Computational Materials (2024)

Low power nanoscale S-FED based single ended sense amplifier applied in integrate and fire neuron circuit

• SeyedMohamadJavad Motaman
• Tara Ghafouri
• Negin Manavizadeh

Memristor-based storage system with convolutional autoencoder-based image compression network

• Yizhou Zhang
• Jinfeng Kang

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• HHS Author Manuscripts

Learning and memory

Anna-katharine brem.

1 Berenson-Allen Center for Noninvasive Brain Stimulation, Division of Cognitive Neurology, Department of Neurology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA

ALVARO PASCUAL-LEONE

2 Institut Guttman de Neurorehabilitació, Universitat Autonoma, Barcelona, Spain

INTRODUCTION

A fairly large number of studies to date have investigated the nature of learning and memory processes in brain-injured and healthy subjects with noninvasive brain stimulation (NBS) methods. NBS techniques, such as transcranial magnetic stimulation (TMS) and transcranial direct current stimulation (tDCS), can alter brain activity in targeted cortical areas and distributed brain networks. The effects depend on the stimulation parameters. TMS and tDCS can be used to interfere with ongoing brain activity (“virtual lesion”) and thus help to characterize brain–behavior relations, give information about the chronometry of cognitive processes, and reveal causal relationships. Particularly in real-time combination with electroencephalography (EEG) or functional magnetic resonance imaging (fMRI), TMS and tDCS are valuable tools for neuropsychological research. They offer the combination of interference methods (TMS, tDCS) with techniques to record ongoing brain activity with high temporal (EEG) and spatial (MRI) resolution. This can: (1) shed unique insights into physiological and behavioral interactions, and (2) test, refine, and improve cognitive models; and (3) might ultimately lead to better neurorehabilitative methods.

The main goals of research with NBS in learning and memory have been to: (1) identify underlying neuropsychological processes and neurobiological components; (2) find out how this knowledge can be used to diagnose and restore dysfunctions of learning and memory in various patient populations; and (3) assess the use of NBS for enhancement purposes in healthy subjects.

In the present chapter, we first review and define memory and learning processes from a neuropsychological perspective. Then we provide a systematic and comprehensive summary of available research that investigates the neurobiological substrates of memory and aims to improve memory functions in patient populations, as well as in healthy subjects. Finally, we discuss methodological considerations and limitations, as well as the promise of the approach.

FRAMING APPLICATION OF NONINVASIVE BRAIN STIMULATION IN THE CONTEXT OF NEUROPSYCHOLOGICAL DEFINITIONS

Learning and memory are cognitive functions that encompass a variety of subcomponents. These components can be structured in different ways. For example, we can focus on their temporal dimension, or differentiate various forms of memory by virtue of their content or mechanisms of acquisition ( Fig. 55.1 ). It seems clear that the cognitive structure of learning and memory is complex, and that, given the many interactions and overlaps between key subcomponents, neither neuropsychological nor neurobiological models can give us a fully satisfying taxonomy.

Classification of different types of memory process.

A key advance in the study of the neurobiological substrates of memory was Squire’s (1987 , 2004 ) distinction between declarative and nondeclarative memory functions related to their differential reliance on distinct neural structures ( Cohen and Squire, 1980 ). Declarative memory incorporates semantic and episodic memory, and refers to everyday memory functions, which are typically impaired in amnesic patients. Declarative memory is thought to rely primarily on medial temporal lobe structures, including the hippocampus. Nondeclarative memory includes various subcomponents, of which procedural memory or formation of motor memories is the most prominent. Nondeclarative memory is thought to depend mostly on striatum, cerebellum, and cortical association areas ( Cohen and Squire, 1980 ). However, procedural memory also includes associative learning forms, such as classical and operant conditioning, and nonassociative learning forms such as priming, habituation, and learning of perceptual and cognitive routines. Notably, motor learning has been regarded as a less cognitive form of memory functions, and most research makes a clear distinction between motor and nonmotor memory functions. Thus, it seems clear that declarative and nondeclarative memory processes are interactive and partly overlapping domains.

Historically, the distinction between explicit and implicit memory has been associated with declarative and nondeclarative memory. It is often argued that declarative memory (semantic and episodic memory) corresponds to explicit memories that are conscious and verbally transmittable. On the other hand, nondeclarative memory is thought to represent an implicit and nonverbal type of memory that is acquired subconsciously. Although most declarative memory contents seem to be acquired explicitly, and most nondeclarative memory contents appear to be acquired implicitly, this dichotomy is an oversimplification and ultimately not accurate. For example, declarative memories can be acquired subconsciously (e.g., memories of an emotionally intense event or subliminal priming effects), and nondeclarative memories can be acquired with conscious engagement (e.g., learning of motor movements playing sports or a musical instrument).

Another important dichotomy, first proposed by William James (1890) , differentiates memory subcomponents along a temporal dimension of duration (short-versus long-term memory, STM versus LTM). Since then researchers have proposed that STM and LTM are dependent on different neural substrates. More recently, however, it has been argued that the same representations that are active during encoding are also active during STM or during retrieval from LTM. According to these models, medial temporal lobe structures are responsible for the establishment of new representations independent of their duration, and the same binding processes are active in both STM and LTM ( Wheeler et al., 2000 ; Jonides et al., 2008 ). A related temporal dichotomy separates retrograde and anterograde memory processes ( Hartje and Poeck, 2002 ; Markowitsch and Staniloiu, 2013 ). Access to memories of the past enables us to improve current decisions, while mental time traveling and the imagination of future experiences helps us to follow long-term goals ( Boyer, 2008 ).

These are some of the complex and not mutually exclusive dichotomies of memory processes that NBS could help link to specific neural substrates. For example, one can conceive of experiments aimed at assessing whether disruption of specific brain regions affects one type of memory process and not another (e.g., Basso et al., 2010 ), or experiments evaluating the time at which disruption of a given brain region interferes with a specific memory step (e.g., Oliveri et al., 2001 ). One can use NBS to explore the nature of the relation between different processes within or across different dichotomies. Finally, one can compare the effects of NBS in healthy individuals and those with deficits in specific memory processes, and evaluate the impact on the deficit or even on other, apparently unaffected, memory and learning types.

It is also apparent that memory is tightly connected to time perception, attention, and emotional valence of memory contents, and there is evidence that brain circuits implicated with these functions are overlapping with areas involved in processing of memory functions. For example, with an increasing load of varying experiences stored in memory, time intervals are perceived to be longer ( Bailey and Areni, 2006 ), and the subjective perception of a long time interval recruits areas such as the medial temporal cortex, which is known to be involved in binding episodic memory features ( Noulhiane et al., 2007 ). State-dependent models have proposed that there is no “centralized clock,” but that there are time-dependent neural changes, such as short-term synaptic plasticity, accounting for the decoding of temporal information ( Karmarkar and Buonomano, 2007 ). It has been suggested that there is no linear metric of time, but that short time intervals are rather encoded in the context of (memory) events and therefore a state of local neural networks. In the same way as long-term plasticity may provide a memory of a learning experience ( Martin et al., 2000 ), state-dependent networks may use short-term plasticity to provide a memory trace of the recent stimulus history of a network ( Buonomano, 2000 ). These are further examples of questions that NBS can help address. Pharmacological experimental interventions suggest that affecting working memory (WM) also interferes with temporal processing ( Rammsayer et al., 2001 ). However, NBS offers a promise of spatial and temporal precision that pharmacological agents lack.

Currently, researchers are trying to integrate findings in the memory domain into comprehensive models aiming to account for the wealth of data on functional characteristics of memory networks. There are debates over the implication of attention functions to memory and specifically, for example, of the role of parietal regions to retrieval of episodic memory. For instance, the Attention to Memory (AtoM) model postulates that the dorsal parietal cortex mediates top-down attention processes guided by retrieval goals (orienting), while ventral parietal cortex mediates automatic bottom-up attention processes captured by retrieved memory output (detection) ( Ciaramelli et al., 2008 ; Cabeza et al., 2011 ). Cabeza and colleagues (2011) have proposed that parietal regions control attention in a similar way to perception processes. While orienting-related activity for memory and perception are thought to overlap in dorsoparietal cortex (DPC), detection-related activity is believed to overlap in ventroparietal cortex (VPC). Furthermore, both DPC and VPC show strong connectivity with medial–temporal lobe (MTL) during a memory task, which can, however, shift to strong connectivity with visual cortex during a perception task. Accordingly, the DPC appears to be collaborating with the prefrontal cortex (PFC) to induce top-down attention to salient retrieval paths, while the VPC seems to be involved in the activation of episodic features in alliance with the MTL. Thus, current models of memory processes integrate dynamic concepts of distributed network interactions and plasticity. These and other conclusions are derived from brain imaging studies, which, although extremely valuable, cannot offer insights into causality ( Silvanto and Pascual-Leone, 2012 ). Here again, NBS offers the promise of a transformative approach.

Procedural memory

Motor learning and the formation of motor memories can be defined as an improvement of motor skills through practice, which are associated with long-lasting neuronal changes. They rely primarily on the primary motor cortex, premotor and supplementary motor cortices, cerebellum, thalamus, and striatal areas ( Karni et al., 1998 ; Muellbacher et al., 2002 ; Seidler et al., 2002 ; Ungerleider et al., 2002 ). As learned from patients with apraxia, the parietal cortex is furthermore implicated in accessing long-term stored motor skills and contributes to visuospatial processing during motor learning ( Halsband and Lange, 2006 ). Frontoparietal networks may become important after learning has been established, and play key roles in consolidation and storage of skill ( Wheaton and Hallett, 2007 ).

Motor learning and memory take a special place within the memory domain and have been studied extensively. However, procedural memories build on subprocesses similar to those of nonmotor memories: they are divided into encoding, consolidation and long-term stability, retrieval ( Karni et al., 1998 ; Robertson et al., 2005 ), and even a short-term memory system has been suggested to exist in the primary motor cortex ( Classen et al., 1998 ). Robertson (2009) has further proposed that motor and nonmotor memory processes may be fully or partially supported by the same neuronal resources during wakefulness, but not during sleep. Indeed, the MTL – which is known to support declarative memory formation – also contributes to implicit procedural learning ( Schendan et al., 2003 ; Robertson, 2007 ; Albouy et al., 2008 ). During sleep, motor and nonmotor memory systems may be functionally disengaged, which may promote independent offline consolidation within systems ( Robertson, 2009 ). As we shall see, key aspects of such insights have been derived from recent studies using NBS.

Short-term memory

STM is an essential component of cognition and is defined as the maintenance of information over a short period of time (seconds). Multistore models differentiate between STM and LTM. STM can remain unimpaired in amnesic patients who show distinct LTM impairments ( Scoville and Milner, 1957 ; Cave and Squire, 1992 ). However, STM can be impaired while LTM functions remain intact ( Shallice and Warrington, 1970 ). According to William James (1890) , STM (primary memory) involves a conscious maintenance of sensory stimuli over a short period of time after which they are not present anymore. On the other hand, LTM (secondary memory) involves the reactivation of past experiences that were not consciously available between the time of encoding and retrieval. This led to the assumption, going back to Hebb (1940s), that STM and LTM are based on separate neural systems. While STM engages repeated excitation of a cellular compound, LTM leads to structural changes on the synaptic level, which are preceded by consolidation processes that are thought to be highly dependent on hippocampal functions. NBS, particularly TMS combined with EEG, MRI, or other brain imaging methods, has provided valuable insights on such neurobiological questions.

Baddeley also proposed a multistore architecture of STM and LTM ( Baddeley and Hitch, 1974 ; Baddeley, 1986 ). In his model, STM consists of a “verbal buffer” (phonological loop) and a “visuospatial sketchpad” (maintenance of visual information). He later added an “episodic buffer” that is supposed to draw on the other buffers and LTM ( Baddeley, 2000 ). Finally, a “central executive” is argued to be responsible for orchestrating all components. As we shall see, such cognitive models lend themselves exquisitely well to hypothesis testing with NBS.

Unitary store models assume that the MTL is engaged in both STM and LTM, and that its function is the establishment of new representations independent of their duration. Accordingly, information that does not require binding processes can be preserved in amnesic patients, which might also explain often preserved retrieval of consolidated preinjury memories. In a comprehensive review, Jonides and colleagues (2008) concluded that STM and LTM are not separable, but that STM consists of temporarily activated LTM representations. Several studies have confirmed these assumptions ( Ranganath and D’Esposito, 2001 ; Hannula et al., 2006 ; Olson et al., 2006a , b ). According to their assumptions, initial neural representations are also the repository of long-term representations, as they are active during encoding, as well as during STM, or the retrieval from LTM into STM ( Wheeler et al., 2000 ). Chronometric brain stimulation experimental designs can be applied to explore such questions (e.g., Mottaghy et al., 2003a ).

Long-term memory

LTM refers to the mechanism by which acquired memories gain stability or are strengthened over time, and become resistant to interference ( Brashers-Krug et al., 1996 ; McGaugh, 2000 ; Dudai, 2004 ). Consolidation is assessed as a change in performance between testing and retesting ( Robertson et al., 2004 ; Walker, 2005 ) and provides a direct measure of “offline” changes.

Mainly two components of LTM are described in the literature and frequently included under the term “declarative memory” – episodic and semantic memory. They rely mostly on MTL structures. Episodic memory refers to contents that can be located within a spatiotemporal context, such as holiday memories or autobiographical events. On the other hand, semantic memories are independent of context and are not personally relevant. They consist of general and factual world knowledge, such as “Dakar is the capital city of Senegal.” However, “nondeclarative” memory functions, such as procedural memory (see above), also involve LTM consolidation processes, such as knowing how to ride a bike.

Successful long-term storage includes several steps starting with the encoding of information, followed by short-term storage and consolidation from STM to LTM, as well as repeated reconsolidation. Consolidation is thought to occur in a structured way allowing for prompt and precise retrieval. Elegant work from Muellbacher and colleagues (2002) pioneered the use of NBS approaches to explore the neurobiology of such processes in humans. During consolidation, memories can undergo changes that can be quantitative (enhancement, strengthening) as well as qualitative in nature (e.g., awareness of underlying sequences) ( Wagner et al., 2004 ; Walker, 2005 ; Robertson and Cohen, 2006 ). Chronometric brain stimulation paradigms are contributing to clarify some of these issues. Consolidation mechanisms may depend on neuronal reactivation (signal increase), on the removal of noise-inducing synaptic changes (noise decrease), or their combination, all of which can be examined with NBS. For example, offline performance changes seem to be causally associated with neuronal reactivation ( Rasch et al., 2007 ). However, it remains to be shown that disruption of reactivation would impair consolidation processes, a problem that seems experimentally approachable with TMS.

It has been shown that sleep plays an important role in the consolidation of memories ( Walker et al., 2002 ; Korman et al., 2007 ), and it has been argued (synaptic homeostasis hypothesis) that a net increase in the efficacy and number of synapses during wakefulness may add noise to the network. The reduction of noise would therefore improve the signal-to-noise ratio. Slow-wave sleep is thought to be responsible for downscaling synaptic strength and therefore noise reduction ( Tononi and Cirelli, 2003 , 2006 ), and has been associated with learning and the induction of brain plasticity ( Huber et al., 2004 , 2006 ; De Gennaro et al., 2008 ). NBS, in this case, particularly tDCS, is being elegantly employed to test some of these notions, while TMS–EEG studies are providing experimental support for the underlying hypotheses (e.g., Marshall et al., 2004 , 2011 ).

Encoding and retrieval

During encoding, various event features distributed across neocortical areas are held actively online through processes guided by the PFC ( Miller and Cohen, 2001 ; D’Esposito, 2007 ). TMS and tDCS lend themselves well to experimentally test such notions and evaluate precise spatial and temporal aspects of the hypothesized neural substrates.

The MTL is thought to be responsible for binding these representations in a highly structured way to enable optimal retrieval at a later timepoint ( Cohen and Eichenbaum, 1991 ; Squire and Zola, 1998 ), and activity in PFC and MTL during encoding is correlated with successful retrieval ( Paller and Wagner, 2002 ). Moreover, intermediate processes such as additional encoding or consolidation processes, are relevant for further stabilization of memories ( Squire, 1984 ; Nadel et al., 2000 ; Paller, 2002 ). Critical encoding components include bottom-up sensory processes as well as top-down processes that select/engage, maintain, and update relevant features ( Shimamura, 2011 ). Here again, NBS is a valuable experimental tool, thanks to the opportunity of interference with ongoing neural activity in a spatially and temporally controlled manner.

Retrieval of episodic memories depends on the recollection of encoded contextual features of a past event, such as time, place, people, sights, thoughts, and emotions ( Mitchell and Johnson, 2009 ). Source memory is therefore an important element of episodic memory ( Tulving, 2002 ; Shimamura and Wickens, 2009 ). MTL plays its part in memory retrieval by reinstating these features ( Eldridge et al., 2005 ; Moscovitch et al, 2006 ). Successful retrieval has also been associated with the PFC ( Buckner et al., 1998 ; Dobbins et al., 2002 ; Simons and Spiers, 2003 ), which is involved in top-down executive control. The HERA (Hemispheric Encoding/Retrieval Asymmetry) model proposed by Tulving and colleagues (1994) postulates that both prefrontal lobes subserve memory processes, but play different roles. While the left PFC is believed to be more involved in encoding and semantic retrieval, the right PFC is thought to be more important in episodic memory retrieval. Early functional imaging studies proposed an asymmetry in memory processes irrespective of modality, with encoding and retrieval being associated with left and right/bilateral PFC respectively ( Cabeza and Nyberg, 2000 ; Haxby et al., 2000 ; Fletcher and Henson, 2001 ). The HAROLD (Hemispheric Asymmetry Reduction in OLDer adults) model suggests that prefrontal activity during cognitive performance becomes less lateralized with advancing age ( Cabeza, 2002 ). In particular, the role of the PFC can be evaluated with TMS or tDCS, as the PFC is easily accessible to modulation with NBS (e.g., Gagnon et al., 2010 , 2011 ).

Besides MTL, PFC, and cortical sites that store contextual features, brain imaging studies suggest that parietal areas also play an important role in episodic memory retrieval ( Wagner et al., 2005 ; Cabeza et al., 2008 ). For instance, according to a recently proposed theory (“COrtical Binding of Relational Activity”, CoBRA), the VPC acts as a binding zone for episodic features and linking these to long-term memory networks ( Shimamura, 2011 ). Both the CoBRA model and the AtoM model (see above) share some similarities, as both suggest that MTL and VPC are linked. Although the role assigned to the VPC differs between the AtoM model (bottom-up processes) and the CoBRA model (integration of event-related activity), they might complement each other. Paired-pulse TMS and the combination of TMS with brain imaging are well suited to examine such notions of corticocortical interactions.

Prospective memory

Prospective memory involves an intention to carry out a psychological or physical act and is related to future-oriented behaviors. In order to realize a goal in the future, it is necessary to retain intentions and activate them at the right time and/or in the appropriate context ( Ellis et al., 1999 ). Depending on the time that passes in between the creation of the intention and the action, and depending on whether the action is triggered externally (context feature) or internally (internal pacemaker), prospective memory involves working and long-term memory processes, as well as attentional processes ( Wittmann, 2009 ). Within this context it has been proposed that, during encoding, prospective memory contents obtain a special status, where they are tagged as not being achieved yet. During the presentation of prospective memory cues, temporal areas are active, possibly representing stimulus-driven attentional processes ( Reynolds et al., 2009 ). The delay period between encoding the intention and the actual act is filled with cognitive activity that prevents active and conscious rehearsal, which differentiates prospective memory from WM or vigilance ( Reynolds et al., 2009 ; Burgess et al., 2011 ). Prospective memory and WM take a special place within the memory domain as they rely strongly on executive processes. However, prospective memory and WM engage different brain areas. Whereas WM demands dorsolateral prefrontal cortex (DLPFC) activity, prospective memory has been associated mainly with activation in the rostral PFC ( Okuda et al., 1998 , 2007 ; Reynolds et al., 2009 ), which is implicated in “future thinking” ( Atance and O’Neill, 2001 ). Such, largely theoretical, considerations derived from careful task analysis and psychological and cognitive model formation can be tested experimentally using NBS.

Working memory

WM refers to the temporary, active maintenance and manipulation of information necessary for complex tasks, while ignoring irrelevant information. It involves the temporary manipulation of external (experienced) or internal (retrieved) stimuli. Like other memory components, it also involves an encoding and retrieval stage. The PFC is an integral component for successful WM performance ( Missonnier et al., 2003 , 2004 ; Jaeggi et al., 2007 ), and NBS offers experimental approaches that were previously limited to animal models.

WM takes a special place within the memory functions, as it is highly dependent on top-down processing and selective attention. Top-down modulation allows us to focus attention on relevant stimuli and ignore irrelevant distractors. This is achieved through an improvement of the signal-to-noise ratio by increasing sensory activity for relevant items and decreasing activity for irrelevant items ( Gazzaley and Nobre, 2012 ). Successful manipulation of information is necessary for encoding as well as the integration of memory functions with other so-called higher cognitive functions associated with conscious processing, such as decision-making, mental imagery, interference control, or language functions. State-dependency experimental designs with NBS ( Silvanto and Pascual-Leone, 2008 ) might allow selective modulation of different items of information and thus shifting of the signal-to-noise ratio. This offers intriguing promises for translational applications of such NBS to populations with WM deficits, such as the elderly or patients with attention-deficit disorders, Parkinson’s disease, or schizophrenia.

UNDERSTANDING THE NEURAL MECHANISMS OF LEARNING AND MEMORY

Learning and memory processes are investigated with a wealth of methods. In the literature we find studies that use brain imaging during memory tasks, analyze the number of remembered items correlated with EEG activity, look at the influence of state changes as captured by various brain imaging and neurophysiological measures, or “borrow patients’ illnesses” to investigate the impact of serendipitous lesions. The application of all these methods has led to valuable information about the neural mechanisms of memory. However, cause–effect relationships are difficult to establish. NBS is uniquely suited to provide this ( Silvanto and Pascual-Leone, 2012 ).

Although TMS and tDCS both promote changes in excitability, they do not rely on the same processes ( Wagner et al., 2007 ; Nitsche et al., 2008 ) and behavioral effects can be different. Neuronavigated TMS can serve to probe the spatio-temporal contribution of certain structures and processes important for learning and memory. It can reveal where and when certain memory processes happen and can shed light on the interplay of multiple processes. On the other hand, the temporal and spatial resolution is lower for tDCS, which is a reason why the utility of tDCS to study spatiotemporal properties of learning and memory is limited. In the following section we concentrate on studies applying TMS as a means to induce so-called “virtual lesions” in the healthy brain ( Pascual-Leone et al., 2000 ). In recent years, research in this field has grown immensely.

Assessing memory functions by induction of virtual lesions in healthy subjects

The first systematic investigation of the contribution of certain brain areas to cognitive functions took place during World War I. Soldiers with circumscribed brain lesions after gunshots provided information about how certain brain regions are associated with cognitive functions ( Lepore, 1994 ). Later, Luria’s work with brain-damaged war veterans contributed strongly to rekindling of the interest in neuropsychology during World War II ( Luria, 1972 ).

Although lesion studies with patients have been widely used since then to investigate learning and memory, they have some disadvantages. Important variables, such as, for example, lesion size, comorbidities, and age, cannot be controlled easily. On the other hand, modern brain imaging methods, such as positron emission tomography (PET) and fMRI, are able to detect regional activation changes with an excellent spatial resolution, and allow for controlled, test–retest experimental designs, but their low temporal resolution does not allow investigation of the organization of distributed memory networks, and they cannot provide information on facilitatory or inhibitory effects or cause–effect relationships. EEG offers a direct measure of brain activity with exquisite temporal resolution, but spatial resolution is in turn limited.

Many of these disadvantages can be overcome when using TMS to induce a “virtual lesion” in an otherwise healthy brain ( Pascual-Leone et al., 1999 ; Walsh and Pascual-Leone, 2003 ). Instead of studying cognitive functions in patients with brain lesions, we can use TMS as a means to induce virtual lesions in healthy subjects and, therefore, reproduce neurobehavioral patterns of patients with brain lesions. TMS is a method that interferes with brain activity and thereby allows probing the chronological contribution of underlying cortical areas. However, it is important to note that our understanding on the neural mechanisms underlying such “virtual lesions” is rather limited, and that a functional disruption is not simply dependent on a mere modification of cortical excitability in the targeted brain area, but appears to involve a complex interplay of inhibitory and excitatory mechanisms, disruption of oscillators, and modification of functional connectivity and synaptic efficacy across distributed neural networks.

TMS has been used in a vast number of studies investigating mechanisms of motor learning and memory ( Bütefisch et al., 2004 ; Censor and Cohen, 2011 ), whereas studies looking at nonmotor memory functions are less numerous. However, recent technical advances allowing the combination of TMS with EEG and fMRI are promising and will allow further exploration of nonmotor memory processes ( Miniussi and Thut, 2010 ; Thut and Pascual-Leone, 2010 ). The combination of methods has, furthermore, the advantage of helping to unravel local and distant effects of brain stimulation and give us insights into functional connectivity.

Most research groups that study WM or STM with NBS methods have focused on the DLPFC or the parietal cortex, believed to be core cortical structures for memory processes. Typically, these studies have used delayed response tasks or n -back tasks to measure STM or WM performance, respectively. A classical example of a delayed match-to-sample task is the Sternberg task ( Sternberg, 1966 ), where the subject is shown a list of numbers or letters and is asked to memorize them. After the delay period, a probe number or letter is shown and the subject has to indicate whether the probe was in the list. Researchers have used several versions of this test using different stimuli and parameters. In “ n -back tasks” a string of visual or auditory stimuli is presented, and subjects have to compare each new stimulus with a stimulus presented n trials back. n -back tasks with n = 1 involve a continuous maintenance and matching of stimuli, whereas n -back tasks with n > 1 furthermore require concurrent engagement of manipulation processes. The reallocation of attention and processing capacity away from mere matching to actual WM processes (by increasing n ) is reflected in decreasing P300 amplitudes ( Watter et al., 2001 ). As these tasks draw on different processes, we will address them in separate sections. Studies using delayed match-to-sample tasks will therefore be summarized under the STM section, whereas studies using the n -back task, or other tasks requiring the online manipulation and integration of stimuli, will be summarized under the WM section. Another major section gives an overview for studies that have investigated encoding, consolidation, and retrieval.

The number of studies that apply TMS and tDCS to address questions regarding the underlying neurobiological structure and modulation of memory functions has grown rapidly in past years. The studies presented in Table 55.1 have applied single-pulse TMS, paired-pulse TMS, repetitive TMS (rTMS), and theta-burst stimulation (TBS). The tasks that were used draw on various processes (attentional, sensory, motor, verbal/nonverbal, spatial/nonspatial, maintenance/manipulation) and stimulation parameters, such as pattern, timing, duration, intensity, and location, vary across studies. It is important to realize that memory tasks vary greatly regarding their specific cognitive demands. In addition, it is important to recognize TMS methodological factors. For example, online stimulation differs from offline stimulation in that underlying brain areas are concomitantly activated through TMS as well as through task performance. This combined activation may affect stimulation outcome. Finally, note that some studies report effects on accuracy, whereas others focus on response times (see Table 55.1 ). It is important to note, though, that the amount of time it takes to recognize an already encountered stimulus or to recall a memorized representation is far less important than the accuracy of this process. Finally, we have to keep in mind that the act of receiving TMS may have an influence on attentional processes that should be carefully controlled for.

Synopsis of peer-reviewed, published studies applying noninvasive brain stimulation in the memory domain

A, C, S, anodal, cathodal, sham; ADAS-Cog, Alzheimer’s Disease Assessment Scale – Cognitive subscale; aMT, active motor threshold; ATL, anterior temporal lobe; BA, Brodmann’s area; BBR, brain-behavior relationship; Cb, cerebellum; CGIC, Clinical Global Impression of Change; CPM, colored progressive matrices; cTBS, continuous theta-burst stimulation; Cz, vertex; DLPFC, dorsolateral prefrontal cortex; DMS, delayed match-to-sample; DMPFC, dorsomedial prefrontal cortex; DR, discrimination rate; EF, executive functions; ERP, event-related potential; Exp., experiment; FEF, frontal eye fields; FL, frontal lobe; fO, frontal operculum; Fz, frontal midline; GDS, Geriatric Depression Scale; HF, high frequency; HVOT, Hooper Visual Organization Test; IADL, Instrumental Activities of Daily Living; IFG, inferior frontal gyrus; IFJ, inferior frontal junction; IPL, inferior parietal lobule; IPS, intraparietal sulcus; ITI, intertrain interval; L, left; LA/RA, left anodal/right anodal; LC/RC, left cathodal/ right cathodal; LF, low frequency; LH, left hemisphere; LOC, lateral occipital cortex; M1, primary motor cortex; MFG, middle frontal gyrus; MMSE, Mini Mental State Examination; MSO, maximum stimulator output; NP, neuropsychological; NPI, neuropsychiatric inventory; OC, occipital cortex; OFC, orbitofrontal cortex; p, pulse; PANAS, positive and negative symptoms scale; PC, parietal cortex; PCG, postcentral gyrus; PD, Parkinson’s disease; PFC, prefrontal cortex; PL, parietal lobule; PM, prospective memory; PMC, premotor cortex; PPC, posterior parietal cortex; ppTMS, paired-pulse transcranial magnetic stimulation; R, right; RBMT, Rivermead Behavioural Memory Test; rCBF, regional cerebral blood flow; ref, reference; RH, right hemisphere; rMT, resting motor threshold; R-pSAC and L-pSAC, right and left parietal somatosensory association cortex; RT, reaction time; rTMS, repetitive transcranial magnetic stimulation; S1, primary somatosensory cortex; SFG, superior frontal gyrus; sign., significant; SMG, supramarginal gyrus; SOA, supraorbital area; sp, single pulse; SPL, superior parietal lobule; stim, stimulation; STM, short-term memory; T, tesla; TL, temporal lobe; TMT, trail making test; TPC, temporoparietal cortex; tRNS, transcranial random noise stimulation; TSOS, transcranial slow oscillation stimulation; VAT, visual attention task; VLPFC, ventrolateral prefrontal cortex; VPFC, ventral prefrontal cortex; VFT, verbal fluency test; VRT, visual recognition task; WAIS, Wechsler Adult Intelligence Scale; WCST, Wisconsin Card Sorting Test; WM, working memory; y, years.

Despite the many differences between studies, the growing literature summarized in Table 55.1 is providing important novel insights in the neurobiology of human learning and memory, and illustrates the power of NBS in this area of cognitive neuroscience.

S hort-term memory

Prefrontal areas undoubtedly play an important role in STM processes. However, one of the questions that NBS studies are helping address relates to the organization of information processing streams. Is processing of STM supported through a domain-specific segregation (spatial, object, verbal processing) or rather through a processing segregation (encoding, maintenance, storage)?

Processing segregation

Most studies examining this question have used a delayed match-to-sample task and applied stimulation during either the delay period or the decision period ( Fig. 55.2 ). High-frequency TMS applied over the parietal cortex during the delay period can improve STM function ( Kessels et al., 2000 ; Kirschen et al., 2006 ; Luber et al., 2007 ; Yamanaka et al., 2010 ), but some studies found it to impair STM ( Koch et al., 2005 ; Postle et al., 2006 ). In either case, the effects seem specific to the delay period, since parietal TMS during the decision phase has not been found to impact STM ( Luber et al., 2007 ; Hamidi et al., 2009 ). The question whether DLPFC also plays a role during the delay phase has not been answered yet. Although some TMS studies support DLPFC participation ( Pascual-Leone and Hallett, 1994 ; Koch et al., 2005 ), others have found no impact when stimulating DLPFC during the delay phase ( Herwig et al., 2003 ; Postle et al., 2006 ; Hamidi et al., 2008 ; Sandrini et al., 2008 ). On the other hand, high-frequency TMS over the DLPFC during the decision period impairs STM functions ( Koch et al., 2005 ; Hamidi et al., 2009 ). Therefore, although further studies are needed, findings suggest a dissociation between parietal and prefrontal areas, playing primary roles in delay and decision phases, respectively. These findings are supportive of the notions of posteroanterior temporal gradient in memory processing: parietal regions coming online first and prefrontal regions contributing to later subprocesses. Chronometric TMS experimental designs enable such notions to be directly tested further.

Schematic summary of findings from studies investigating the impact on short-term memory after stimulation over the left or right prefrontal cortex, parietal areas, or the cerebellum during the delay (green) or the decision period (orange).

Mottaghy et al. (2003a) conducted the first such experiment ( Fig. 55.3 ), albeit focusing on verbal WM. They used single-pulse TMS to explore the temporal dynamics of left and right inferior parietal and DLPFC involvement in verbal WM in six healthy volunteers. TMS was applied at 10 different time points 140–500 ms into the delay period of a 2-back verbal WM task. Precise and consistent targeting of a given cortical brain region was assured by using frameless stereotactic neuronavigation. A choice reaction task was used as a control task. Interference with task accuracy was induced by TMS earlier in the parietal cortex than in the PFC, and earlier over the right than the left hemisphere. This suggests a propagation of information flow from posterior to anterior cortical sites, converging in the left PFC. Significant interference with reaction time was observed after 180 ms with left PFC stimulation. These effects were not observed in the control task, underlining the task specificity of our results. Hamidi and colleagues (2009) also examined the roles of right and left DLPFC in recall and recognition. They found that right DLPFC stimulation impaired accuracy in delayed recall, while enhancing accuracy in delayed recognition. On the other hand, left DLPFC stimulation impaired delayed recognition. Therefore, it seems clear that TMS, in repetitive and chronometric single-pulse experimental designs, can provide valuable insights into the functional segregation of core subprocesses of STM.

Accuracy in the 2-back task as a function of the time of transcranial magnetic stimulation (TMS). TMS interference peaked at 180 ms at the right inferior parietal cortex, at 220 ms at the left inferior parietal cortex and right middle frontal gyrus (MFG), and at 260 ms at the left MFG (all p < 0.05). This study illustrates the chronometry of causal contributions of different brain regions to memory processing. (Modified from Mottaghy et al., 2003a , by permission of the authors.)

Domain-specific segregation

Mottaghy et al. (2002b) , in another pioneering study ( Fig. 55.4 ) used TMS to show that functional and modality-specific segregation need not be mutually exclusive. They applied low-frequency rTMS to explore the functional organization of STM by selectively disrupting the left dorsomedial PFC (DMPFC), DLPFC, or ventral PFC (VPFC). They applied a 10-min 1-Hz rTMS train before assessing spatial or nonspatial (face recognition) delayed-response performance. Spatial task performance was impaired after rTMS to DMPFC, whereas nonspatial task performance was impaired after rTMS to VPFC. Disruption of the DLPFC affected the performance in both tasks. This finding reveals a task-related segregation of processing streams along prefrontal structures. More recent studies have confirmed the utility of TMS to offer empirical support for modality-specific segregation. For example, Soto et al. (2012) combined evidence from fMRI and rTMS to demonstrate that verbal and nonverbal memories interact with attention functions independently: whereas rTMS to the superior frontal gyrus disrupted STM effects from colored shapes, rTMS to the lateral occipital cortex disrupted effects from written words. Finally, Morgan and colleagues (2013) used TMS to reveal the neural substrates for integration of segregated features of STM processes. They investigated STM for colors, orientations, and combinations of these, and found that continuous TBS (cTBS) over the right parietal cortex or left inferior frontal gyrus selectively impaired STM for combinations but not for single features. Therefore, functional coupling between frontal and parietal areas appears to be critical to bind modality-specific segregated processes.

Study exploring the segregation of memory processes in prefrontal cortex. Two alternative models were proposed based on the data. ( A ) There might be two different, nonoverlapping, functionally segregated regions within the prefrontal cortex (PFC) that are domain-specific (S, spatial domain; F, face domain). Repetitive transcranial magnetic stimulation (TMS) over the dorsomedial PFC (DMPFC) interferes only with the processing of the spatial information. Dorsolateral PFC (DLPFC) stimulation might have induced overlapping interference of two adjacent domain-specific areas, whereas the ventral PFC (VPFC) led only to interference with the processing of the face stimuli. ( B ) The DMPFC and the VPFC interference effects can be explained in the same manner as in proposal ( A ); however, the performance deterioration over the DLPFC in this model might be explained by the interference with information processing of a common module (C) that is employed during both types of stimulus. (Modified from Mottaghy et al., 2002b , by permission of the authors.)

Frontoparietal binding

Frontoparietal interactions in memory formation and maintenance appear to be dynamic and NBS studies – particularly studies combining TMS with MRI or EEG – help gaining critical insights in this regard.

In the motor domain, frontoparietal interactions seem to be particularly important in the early phase of learning, as has been shown in a recent study combining TMS and EEG ( Karabanov et al., 2012 ). In the nonmotor domain, a recent TMS–fMRI study ( Feredoes et al., 2011 ) found that DLPFC contribution to maintenance of stimuli in STM is highly dynamic depending on the presence or absence of distractors. In the presence of distractors, DLPFC changes its communication with posterior regions to support maintenance. These results are supported by tDCS studies that assign the DLPFC an important role in STM in the presence of distractors ( Gladwin et al., 2012 ; Meiron and Lavidor, 2013 ). Zanto and colleagues (2011) combined EEG with 1-Hz rTMS to the right inferior frontal junction to investigate the contribution of the prefrontal cortex in top-down modulation of visual processing and STM in a delayed-match-to-sample task. They found that EEG patterns from posterior electrodes, which are associated with the distinction of task-relevant and -irrelevant stimuli during early encoding, were diminished after TMS, which again predicted a subsequent decrease in STM accuracy. Subjects with stronger frontoposterior functional connectivity furthermore showed greater disruption. Higo and colleagues (2011) combined offline TMS over the frontal junction with subsequent fMRI to explore the same question. They also observed a TMS-induced decrease of effects in posterior regions depending on task relevance/irrelevance. The inferior frontal junction may therefore control the causal connection between early attentional processes and subsequent STM performance, and may regulate the level of activity of representations in posterior brain areas depending on their relevance/irrelevance for response selection.

It could be hypothesized that the interaction between frontal and posterior areas during the delay period secures the maintenance of information, especially if this information needs to be protected from distracting information. These processes may be related to the regulation and reactivation of patterns that were active during encoding. Accordingly, frontal areas might recruit neuronal assemblies and regulate their activity in posterior areas in order to protect and actively maintain information. Such activations may be most prominent at the beginning of the delay period and decrease gradually.

Other brain regions involved in STM

Frontal and parietal areas are undoubtedly the most explored areas in STM. Although it has been debated in the literature, there is some evidence that the cerebellum may also be involved in STM. When Desmond and colleagues (2005) applied single-pulse TMS (at 120% resting motor threshold) over the right superior cerebellum at the beginning of the delay phase, they found an increased reaction time but no change in accuracy for correct trials in the Sternberg task. This is in agreement with a tDCS study that probed the cerebellum and found an abolishment of practice-dependent improvements in reaction time after anodal as well as cathodal tDCS in a Sternberg task ( Ferrucci et al., 2008 ).

Last, but not least, cortical areas implied in sensory processing are also believed to be involved in STM of sensory information, which may be guided through attentional processes. A number of TMS studies have shown a role of visual cortex with visual STM and WM (see review by Postle et al., 2006 ). A few studies have furthermore investigated the tactile domain. Application of TMS to the visual cortex during the delay phase of STM tasks results in a decrease of accuracy in the targeted visual field for high memory loads ( Van de Ven et al., 2012 ) or a decrease in memory scanning rates ( Beckers and Hömberg, 1991 ). The effect of TMS was furthermore shown to be different if applied at the beginning (inhibitory) or at the end (facilitatory) of the delay period in both a visual STM task and imagery ( Cattaneo et al., 2009 ). This is an elegant application of state-dependency TMS experimental designs ( Silvanto and Pascual-Leone, 2008 ). Although neurons implicated in encoding are highly active at the onset of the retention period, TMS might preferentially activate neurons not involved in encoding, thereby reducing the signal-to-noise ratio of the memory trace, and impair behavior.

In the tactile domain, a TMS study using single-pulse stimulation over the middle frontal gyrus (MFG) during the early maintenance period led to a decrease in reaction time in a tactile STM task, even in the presence of a distracting stimulus ( Hannula et al., 2010 ). In a follow-up study, the same group investigated whether this improvement only occurs when the interference is tactile, or whether MFG creates a more general top-down suppression ( Savolainen et al., 2011 ). Their results showed that TMS did not lead to facilitation when a visual interference was presented, but only when the interfering stimulus was also tactile.

These and other findings (e.g., Silvanto and Cattaneo, 2010 ) suggest that sensory brain areas involved in early, modality-specific, processing of perceptual stimuli contribute to the formation and maintenance of STM representations through an interaction with the attentional system. In this context, TMS can help elaborate the chronology of memory processes and contributions of state-dependent processes.

W orking memory

WM has been investigated with NBS in a growing number of studies. As for STM, most of these studies have explored the roles of DLPFC and parietal areas, trying to find an answer to the question of whether information is separately processed with regard to domain or functional subprocess ( Fig. 55.5 ). In addition, some studies have examined the question of whether the same areas that participate in STM are also active in WM tasks.

Schematic summary of findings from studies investigating the impact on verbal (red) or nonverbal (blue) working memory (WM) after stimulation over the left or right prefrontal cortex (PFC), parietal areas, or cerebellum.

Verbal and nonverbal WM in DLPFC

Again building on pioneering work from Mottaghy et al. (2000) , most researchers have found an impairment of verbal WM after stimulation of the left DLPFC ( Mull and Seyal, 2001 ; Mottaghy et al., 2000 , 2003a ; Postle et al., 2006 ; Osaka et al., 2007 ) and after stimulation of the right DLPFC ( Mottaghy et al., 2003a ; Postle et al., 2006 ; Sandrini et al., 2008 ). However, some studies failed to find such effects ( Mull and Seyal, 2001 ; Rami et al., 2003 ; Imm et al., 2008 ; Sandrini et al., 2008 ).

The role of DLPFC in nonverbal WM has been studied much less ( Oliveri et al., 2001 ; Imm et al., 2008 ; Sandrini et al., 2008 ). Sandrini and colleagues (2008) tried to clarify domain- and process-specific contributions of the DLPFC. They presented physically identical stimuli (letters in different spatial locations) in a 1-back task (STM) and a 2-back task (WM). Furthermore, they presented the 2-back task with stimuli of both or just one domain. A short train of 10-Hz rTMS was applied at the end of the delay period between stimuli. They found interference only during the 2-back task, and only when stimuli from both domains were presented. Interestingly, performance in the letter task was impaired after rTMS over the right DLPFC, whereas performance in the location task was impaired after rTMS over the left DLPFC. These results were interpreted as an interference effect on control mechanisms (central executive) in the sense of the suppression of task-irrelevant information. The same hypothesis has been put forward with regard to the protection of memory contents in STM ( Feredoes et al., 2011 ; Higo et al., 2011 ; Zanto et al., 2011 ), according to which an interaction between frontal and posterior areas during the delay period secures the maintenance of information, especially in the presence of distractors.

Further experiments have aimed at dissecting the role of DLPFC in WM in order to find out whether domain- or process-specific models should be favored, and others have examined the role of interactions between DLPFC and other brain areas. Combination of TMS with brain imaging has proven quite valuable in this context. Mottaghy and colleagues (2000) found that performance in a verbal WM (2-back) task was significantly diminished after rTMS (30-second train of 4-Hz rTMS) to the left but also the right DLPFC (F3/F4). Importantly, by combining TMS with PET, they showed that TMS-altered performance in the WM task was associated with a reduction in regional cerebral blood flow (rCBF) at the stimulation site and in distant areas as assessed with PET. In an elegant follow-up TMS–PET study, the same authors ( Mottaghy et al., 2003b ) showed that at baseline (in the absence of TMS) there was a negative correlation between rCBF in the left (but not the right) DLPFC and WM task performance. Application of rTMS to the left or the right DLPFC could disrupt WM performance, but appeared to do so on the basis of different distributed impact on a bihemispheric network of frontal and parietal regions: whereas rTMS over the left DLPFC led to changes in rCBF in the directly targeted left DLPFC and the contralateral right PFC, rTMS over the right DLPFC led to more distributed changes involving not only bihemispheric prefrontal, but also parietal areas ( Fig. 55.6B ). Regardless of the differential network impact of the right or left stimulation, the behavioral consequences of rTMS were always related to the impact onto left DLPFC rCBF. This study highlights a number of important findings of relevance for future studies on NBS in memory and learning. First, it shows that rTMS to different nodes of a given brain network can exert differential impact onto said brain network. More recently, Eldaief et al. (2011) have expanded on this line of inquiry combining resting-state fMRI with TMS to examine brain network dynamics. Second, the study shows that network dynamics are modified by behavioral engagement. In other words, it might be possible to learn about mechanisms of memory and learning by examining how the impact of TMS onto a given brain network is modulated by the behavioral state. Finally, the study illustrates that brain stimulation can affect behavior by disrupting a computation in the targeted brain region (as in the case of left DLPFC rTMS) or by disrupting function of a brain regions reached via trans-synaptic network impact (as in the case of rTMS to the right DLPFC altering left DLPFC via interhemispheric connections). This later finding is important in the interpretation of brain stimulation results in general, and illustrates the power of studies integrating brain stimulation with neuroimaging in exploring causal relations between brain activity and behavior ( Fig. 55.6A ). In a later study, Mottaghy and colleagues (2003a) applied single-pulse TMS at different time points after stimulus presentation to probe the temporal dynamics of parietal and prefrontal contributions in verbal WM. With this approach they were able to add chronometric information to their prior findings. They showed that single-pulse TMS could interfere with task accuracy earlier in the parietal than in the PFC, and earlier over the right than left hemisphere. This indicates an information flow from posterior to anterior converging in the left PFC. These series of studies reveal that both hemispheres contribute to WM, but that the computation performed by the left PFC is critical in verbal WM.

Transcranial magnetic stimulation–positron emission tomography (TMS–PET) study of the neurobiological substrates of working memory. ( A ) The impact of TMS on behavior relies on activity changes in local and distributed brain networks. The combination of TMS with brain imaging techniques, such as EEG, fMRI, PET, and EMG, allows us to detect correlations between these activity changes and behavior. Moreover, it allows study of the impact of state dependency on stimulation outcome. ( B ) Positive (green) and negative (red) correlation between regional cerebral blood flow (rCBF) and performance in the 2-back working memory (WM) task (1) without application of repetitive TMS (rTMS), (2) with rTMS delivered over the left middle frontal gyrus (MFG), and (3) with rTMS applied over the right MFG. While rTMS over the left MFG has a local impact, which is correlated with behavior, rTMS over the right MFG has an impact on a distributed network, including homologous areas. Importantly, also in the case of stimulation over the right MFG, activity changes in the left MFG but not right MFG are correlated with behavioral output. This key finding shows that the effect of TMS can be achieved by a direct effect on underlying areas, but also through trans-synaptic effects (e.g., in homologous areas). The combination of TMS with imaging techniques is crucial in order to identify neural substrates associated with behavioral output. (Modified from Mottaghy et al., 2003b , by permission of the authors.)

Interestingly, involvement of DLPFC, regardless of stimulus modality, has been shown in an often-cited study using bilateral single-pulse stimulation during a 2-back task ( Oliveri et al., 2001 ). Early temporal stimulation (300 ms) increased reaction time for object-related WM, whereas early parietal stimulation and late stimulation (600 ms) over the superior frontal gyrus increased reaction time for spatial WM. However, late DLPFC stimulation interfered with both tasks and not only with RT, but also with accuracy. These results relate to the discrimination of a dorsal (“where”) and ventral (“what”) pathway and again information flow from parietotemporal to frontal areas. They indicate that there might not only exist a bilateral involvement of the DLPFC in verbal WM, but that DLPFC might be active irrespective of stimulus material, unlike other prefrontal regions that may be segregated (see e.g., Mottaghy et al., 2002b ). Segregation in posterior areas seems to be easier to pinpoint, and is concordant with the view that both hemispheres are implicated in spatial and object WM tasks ( Smith and Jonides, 1997 ). The research that has been done up to date generally points into the direction of favoring a process-specific model for DLPFC, whereas other areas of the prefrontal or parietal cortex may be modality-based. Possibly, WM operations relying on DLPFC, such as selective attention and other executive processes (e.g., the inhibition of task-irrelevant stimuli), are independent of modality ( Smith and Jonides, 1999 ) and play a role in both STM and WM. The combination of fMRI and EEG with TMS may help us to disentangle further the interactions of DLPFC and other prefrontal and parietal areas to WM functions.

P rospective memory

Prospective memory is tightly connected with other memory subcomponents (see Fig. 55.1 ), which makes it difficult to single out its processes. Perhaps this challenge accounts for the fact that few studies to date have explored prospective memory using NBS. One study ( Basso et al., 2010 ) investigated whether verbal WM and prospective memory are based on common or separate processes. In a first experiment participants had to accomplish tasks with low, medium, or high WM load. In the prospective condition, subjects had to react whenever a specific word appeared. In a second experiment the prospective conditions included 1 or 3 prospective words. A higher prospective memory demand interfered with the WM task only at higher loads, whereas WM activity did not affect prospective memory performance. If both processes were part of the same system one might expect a trade-off. In a third experiment single-pulse TMS was applied to the left and right DLPFC in order to test the notion that WM and prospective memory rely on distinct systems. TMS to the DLPFC increased error rates in the prospective memory task, whereas the effect on the WM task was only marginal. No difference between hemispheres was detected. The authors concluded that WM and prospective memory may not be based on the same memory system. However, it is hard to rule out that prospective memory may require resources (including in part WM resources) and may thus be easier to disrupt with TMS. More complex TMS designs, such as input–output designs with TMS applied at various intensity and timings, seem necessary to explore this issue further.

Costa and coworkers (2011) investigated the effects of cTBS (80% active motor threshold) on prospective memory. Stimulation over left Brodmann area (BA) 10 (frontal pole) resulted in impaired accuracy as compared with stimulation over Cz. In a second experiment they did not find a significant difference after cTBS over left BA46 (DLPFC) and Cz. They concluded that the left BA10 is important for prospective memory processes. This is in accordance with a neuroimaging study ( Koechlin et al., 1999 ) that tried to dissociate the roles of frontopolar and DLPF cortices in prospective memory. Costa and colleagues employed a fairly novel TMS paradigm (cTBS) and tackled a complicated memory construct (prospective memory). However, this important, innovative study also illustrates one important challenge for all studies using NBS in memory: it is ultimately critical to have separate empirical demonstration of the impact of brain stimulation on brain function, and on behavior. In fact, ideally, one would want to apply TMS, measure the behavioral impact and the impact on brain physiology, and then correlate one with the other (see Fig. 55.6A ). Costa and coworkers (like most investigators using TMS or tDCS in studies of memory) placed the TMS coil over the scalp overlaying the brain regions they wanted to target (frontal pole or DLPFC). They then assumed that the TMS impact on brain activity would be maximal in the underlying cortex. They assessed the impact of TMS onto prospective memory and assumed that said impact must reflect the consequence of TMS-induced change in activity in the targeted brain region. There is a risk of circular logic in this approach: “If TMS over a given region has a predicted impact onto a given memory process, then I have shown that said brain region was affected by TMS and that it plays said role in memory.” Obviously, independent empirical demonstration of these two steps would be important and the use of NBS in studies of memory, or studies of cognitive functions in general, should aim to achieve such experimental discrimination.

E ncoding, consolidation, retrieval

Some studies have applied rTMS during the encoding phase and support the critical role of the PFC in such memory processes. Stimulation of the left DLPFC during the encoding phase has been found to affect both verbal ( Grafman et al., 1994 ; Rami et al., 2003 ; Sandrini et al., 2003 ; Flöel et al., 2004 ; Skrdlantová et al., 2005 ; Blanchet et al., 2010 ; Gagnon et al., 2010 , 2011 ) and nonverbal ( Rossi et al., 2001 , 2004 ; Blanchet et al., 2010 ; Gagnon et al., 2010 , 2011 ) memory. However, a few studies have reported an impact on memory functions after stimulating right frontal areas during the encoding phase of verbal ( Grafman et al., 1994 ; Sandrini et al., 2003 ; Kahn et al., 2005 ; Blanchet et al., 2010 ; Machizawa et al., 2010 ) or nonverbal ( Epstein et al., 2002 ; Flöel et al., 2004 ; Blanchet et al., 2010 ) memory functions. Some investigators did not find any effects after stimulating right frontal cortex ( Rami et al., 2003 ; Köhler et al., 2004 ). No effects have been found after stimulating parietal ( Köhler et al., 2004 ; Rossi et al., 2006 ) or occipital cortex ( Grafman et al., 1994 ). Only one study reported impairment after stimulating the temporal cortex ( Grafman et al., 1994 ).

Fewer studies have applied TMS during the retrieval phase of memories. Stimulation of the right DLPFC during the retrieval phase appears to be associated with an impact on both verbal ( Sandrini et al., 2003 ; Gagnon et al., 2010 , 2011 ) and nonverbal ( Rossi et al., 2001 , 2004 ; Gagnon et al., 2010 , 2011 ) memory. No studies have reported an effect after stimulation of the left hemisphere during the retrieval phase.

Several studies have used NBS to reveal the important role of the ventrolateral PFC (VLPFC) in the formation of long-term memory ( Grafman et al., 1994 ; Flöel et al., 2004 ; Köhler et al., 2004 ; Machizawa et al., 2010 ), and it has been suggested that VLPFC may be material-specific whereas DLPFC is not. Further studies are needed to shed light on these mechanisms.

Recent studies by Gagnon and colleagues explicitly addressed the assumptions of the HERA model ( Blanchet et al., 2010 ; Gagnon et al., 2010 , 2011 ) and tried to shed light on the contribution of left and right DLPFC in encoding and retrieval of verbal as well as nonverbal information. These are particularly important studies as they illustrate the value of TMS in the systematic testing of key aspects of a well formulated cognitive conceptual model. It is this type of experimental approach that can fully leverage the advantages of TMS in studies of memory and learning. In a first study, Gagnon et al. (2010) applied paired-pulse TMS (interstimulus interval (ISI) 3 ms) over the left or right DLPFC during encoding or retrieval of verbal (words) and nonverbal stimuli (random shapes). They found that left and right DLPFC play different roles in encoding and retrieval irrespective of stimulus type: stimulation of the left DLPFC during encoding resulted in discrimination deficits, whereas stimulation of the right DLFPC during retrieval resulted in a reduced hit and disrimination rate. In a follow-up study they applied paired-pulse TMS with a longer ISI (15 ms) to promote facilitation (rather than cortical suppression) to the left and right DLPFC during encoding or retrieval of verbal (words) and nonverbal stimuli (random shapes) ( Gagnon et al., 2011 ). They found a facilitation of reaction times during encoding (left DLPFC) and retrieval (right DLPFC) regardless of the type of material presented. These results are consistent with other TMS studies ( Rossi et al., 2001 , 2006 ; Rami et al., 2003 ) and provide experimental support for the HERA model, which proposes that the left PFC is more involved in semantic retrieval and episodic encoding than the right PFC, whereas the right PFC is involved in episodic retrieval ( Tulving et al., 1994 ). This hemispheric asymmetry seems to uphold for both verbal and nonverbal material ( Haxby et al., 2000 ; Blanchet et al., 2010 ).

USING NONINVASIVE BRAIN STIMULATION AS A DIAGNOSTIC TOOL

In addition to uses in cognitive neuroscience, it is worth considering the potential utility of NBS in clinical neuroscience as a diagnostic tool. Diagnostic applications of NBS are appealing as they are noninvasive and can be applied safely to various patient populations across the lifespan, if appropriate precautions are taken and guidelines are followed ( Rossi et al., 2009 ). TMS has an excellent temporal resolution and its spatial resolution is superior to tDCS, which are important advantages in diagnostic applications and make TMS a superior tool to probe brain reactivity and brain connectivity.

To date, TMS has not been established as a diagnostic tool. However, if we define carefully the areas of need in specific patient populations, we may be able to complement currently used test measures, which rely mainly on behavioral assessments ( Rost et al., 2008 ; Sigurdardottir et al., 2009 ; Gialanella, 2011 ; Wagle et al., 2011 ).

As for motor dysfunctions, nonmotor memory functions could be characterized by changes in the excitation/inhibition (E/I) balance and cortical plasticity in specific brain areas, which could be assessed with TMS–EEG measures ( Thut and Pascual-Leone, 2010 ). Changes of such neurophysiological measures over the time-course of cognitive rehabilitation, during normal and pathological aging, or in response to treatment of disease could help us establishing neurophysiological biomarkers indicative of functional improvements. Such measures could not only be helpful to differentiate across pathological entities, but may also disentangle underlying causes of memory dysfunctions on an individual level. Finally, this information could help develop novel and improve existing interventions in order to improve memory functions.

In the memory domain there are several questions worth exploring with TMS as a diagnostic tool: (1) What is the pathogenesis of present memory problems? (2) Who is at risk of developing memory problems and what kind of memory problems? (3) Who is likely to benefit from a given behavioral/physiologic/pharmacological intervention?

Identify the pathogenesis of memory problems

Depending on the etiology, the pathogenesis of an individual patient’s memory problem can be vastly different and be affected by many factors including age, environmental, and genetic predispositions. Regardless of etiology, though, one can also aim to identify the proximal, neural dysfunction that accounts for a given memory deficit. TMS can be applied to gain insights at both these levels of inquiry.

Single- and paired-pulse TMS measures may reveal changes in connectivity or altered network dynamics and link those to specific memory functions. Advanced combined technologies such as TMS–EEG or TMS–MRI allow us to utilize TMS-induced cortical evoked potentials or TMS-induced blood oxygen level-dependent (BOLD) fMRI changes as neural measures of brain activity in specific brain regions or networks to relate to behavioral memory measures.

rTMS paradigms, for example intermittent and continuous TBS stimulation (iTBS and cTBS), can be used to obtain indices of cortical plasticity that appear related to long-term potentiation and depression (LTP and LTD)- like induction of synaptic plasticity. Such paradigms can be used to evaluate cortical plasticity in neural structures thought to support memory processes and may allow us to draw conclusions regarding the pathogenesis of a memory problem. For example, a cortical lesion within a widespread memory network could not only have a direct impact on memory functions caused by this particular lesion but could also lead to indirect deficits due to disconnection of the lesioned area with another memory hub. TMS measures could inform us about acute processes as well as adaptive or maladaptive changes characteristic of chronic processes that lead to memory dysfunctions ( Pascual-Leone et al., 2011 ).

Identify risk for developing memory problems

Another major area of interest lies in the possible use of TMS as a physiological biomarker, which could indicate the individual risk of developing memory dysfunctions with age and predict what kind of memory problems could be expected in certain populations. Cognitive decline including memory functions presents a critical hallmark of aging ( Morrison and Baxter, 2012 ). Early changes in neuroplasticity and neurophysiological circuits indicated by TMS measures, such as short-latency afferent inhibition (SAI), could constitute biomarkers for the development of neurodegenerative disorders ( Freitas et al., 2011b ). Risk identification with this approach requires the integration of numerous factors associated with causal and formal pathomechanisms, including age-related changes, but also, for example, changes related to systemic diseases, such as diabetes mellitus, that may indirectly have an impact on brain physiology and plasticity. TMS could be a valuable tool to identify these factors and consequently help guide and implement early interventions in populations at risk.

Another approach is using TMS measures to identify risks related to interventions that could result in brain lesions or dysfunctions. For example, consider neurosurgical interventions: presurgical detailed knowledge about functional contributions of brain areas to be resected can critically inform surgical approaches and minimize the risk. In this context, the Wada test can be used to determine hemispheric language dominance prior to brain surgery ( Wada and Rasmussen, 1960 ). However, this test has a non-negligible risk of complications and discomfort for the patient and does not allow precise functional localization. Neuronavigated TMS can provide detailed information regarding functional anatomy of the targeted brain area and is potentially valuable for presurgical planning not only in regard to language dominance ( Pascual-Leone et al., 1991 ; Devlin and Watkins, 2007 ), but also in regard to memory ( Grafman et al., 1994 ). Such noninvasive neuronavigated TMS cortical mapping appears to correlate well with direct cortical stimulation (DCS) results and seems to be more precise than fMRI, which is the most widely used technique today ( Krieg et al., 2012 ). As DCS is limited to intraoperative use, presurgical TMS might also save operation time by guiding intraoperative DCS.

Predicting benefit from a given intervention/medication

Cognitive rehabilitation consists in assessment-based therapeutic interventions aiming to reduce disability and promote functional recovery. Functional changes are achieved through various intervention methods targeting restitution, compensation, and adaptation ( Cicerone et al., 2000 ). But how can we determine whether a given therapeutic intervention will have a beneficial effect for an individual patient?

TMS measures may be used not only to track but also to predict intervention-related neuroplastic changes within memory networks. Moreover, TMS measures can inform us about the functionality of specific neurophysiological circuits implicated in memory functions and may be indicative of how well an individual will profit from a given pharmacological intervention. For instance, acetylcholine (ACh) is a neurotransmitter that plays a crucial role in synaptic plasticity and memory functions, and ACh imbalances have been associated with memory deficits in patients with Alzheimer’s disease (AD) ( Davies and Maloney, 1976 ; Coyle et al., 1983 ). Deficits in cholinergic circuits can be counteracted with pharmacological interventions involving acetylcholine esterase (AChE) inhibitors. SAI is a TMS measure that is indicative of cholinergic circuits in the motor cortex ( Di Lazzaro et al., 2000 ) and is altered in patients with AD (for a review see Freitas et al., 2011a ). SAI may even be useful to differentiate dementia subtypes ( Di Lazzaro et al., 2006 , 2008 ) and may be used as an indicator of who will profit from AChE inhibitors. Short-latency intracortical inhibition (SICI) and the cortical silent period (cSP) are thought to reflect the excitability of inhibitory γ-aminobutyric acid (GABA)ergic circuits ( Hallett, 2000 ) and were also found to be abnormal in patients with AD. However, the relationship of these TMS measures with specific memory dysfunctions is less clear ( Freitas et al., 2011a ). Notably, studies up to date have relied on TMS measures from the motor cortex. However, the combination of TMS with EEG may enable us to find more precise TMS biomarkers by exploring neurophysiological changes outside the motor cortex.

MODULATING LEARNING AND MEMORY

The interest in the augmentation of cognitive functions reaches far back into the history of modern humanity. The use of memory techniques, for instance in order to improve rhetorical skills, was already promoted by Marcus Tullius Cicero (“De Oratore”, Book II, 55 bc ). One of these methods, the “Cicero Memory Method” (Method of loci), a simple memory enhancement method that uses visualization to structure information, is still in use today. The pursuit of cognitive augmentation has since led researchers to take advantage of technical developments in order to achieve a better outcome. In the past decade, scientists have therefore started investigating the impact of various NBS techniques on memory functions.

Learning is a prerequisite for the formation of memory traces and is thought to be dependent on synaptic plasticity mediated by LTP and LTD, which also represent key mechanisms in the effects of NBS on brain functions. This has not only rendered NBS valuable for the investigation of neuroplastic processes associated with learning and memory but also promotes it as a valuable tool to enhance memory functions.

Although TMS is used mostly for diagnostic purposes and the investigation of brain structures contributing to specific functions, tDCS is more often applied to enhance brain functions.

Healthy subjects

In the past decade, researchers have begun examining the effects of WM training on neural correlates and concomitant performance ( Jaeggi et al., 2008 ). These studies have shown that not only can WM capacity be increased via constructive training but also that said training increases the density of cortical D1 dopamine receptors in prefrontal regions ( McNab et al., 2009 ). The neurobiological substrate of WM is an ongoing topic of research; however, prefrontal regions are believed to be critically involved. Consistent with such notions, studies exploring the potential for NBS to enhance WM have focused on the prefrontal cortex, generally the DLPFC, and the majority have used verbal WM tasks. In most studies subjects were asked to practice STM or WM tasks concurrently to tDCS, and their WM abilities were assessed either during or afterwards.

Compared with sham stimulation, tDCS with the anode over the left DLPFC (and the cathode right supraorbitally) has been repeatedly reported to enhance WM in healthy subjects ( Fregni et al., 2005 ; Ohn et al., 2008 ; Mulquiney et al., 2011 ; Teo et al., 2011 ; Zaehle et al., 2011 ). Some researchers have suggested that increasing stimulation intensity ( Teo et al., 2011 ) or duration ( Ohn et al., 2008 ) might lead to more robust effects. Only one study has reported no memory improvement following tDCS with the anode over the left DLPFC ( Mylius et al., 2012 ), and one study reported improvement in STM but not in WM ( Andrews et al., 2011 ). The only study applying tDCS with the anode over the right DLPFC showed no WM effect ( Mylius et al., 2012 ). On the other hand, tDCS with the cathode over the left DLPFC (and the anode right supraorbitally) yielded diverse results in different studies, ranging from memory benefits ( Mylius et al., 2012 ), to no effects ( Fregni et al., 2005 ), and even negative effects ( Zaehle et al., 2011 ). The study by Zaehle et al. (2011) is of particular interest as the authors reported that the negative effects of tDCS with the cathode over the left DLPFC were associated with decreased electroencephalographic power in theta and alpha bands over posterior (parietal) regions. On the other hand, the authors found that improved WM following tDCS with the anode over the left DLPFC was associated with increased power in alpha and theta EEG bands over parietal regions. This study illustrates the potential of studies combining behavioral and neurophysiological outcome measures, and suggests the critical role of corticocortical interactions in memory enhancement. It has been proposed that a more distributed network may subserve WM functions with the posterior parietal cortex (PPC) playing an important role ( Mottaghy et al., 2002a ; Collette et al., 2006 ). Stimulation might disrupt activity in a given cortical region and thus release activity in a distant connected node, resulting in paradoxical facilitation ( Najib and Pascual-Leone, 2011 ). The specific nature of the stimulation seems important, although, for example, random noise stimulation over the left DLPFC showed no effects ( Mulquiney et al., 2011 ).

In order to explore further the role of parietal structures in WM, Sandrini and colleagues (2012) applied bilateral stimulation over the PPC during a 1-back (STM) or a 2-back (WM) task. They found a double dissociation, with STM being impaired after left-anodal/right-cathodal and WM being impaired after left-cathodal/right-anodal stimulation. They concluded that this dissociation might be due to differential processing strategies in STM and WM. However, the effects might have been mediated by impact on attentional (rather than memory) processes given the fact that only response time, and not accuracy, was affected. Future studies will need to investigate further the contribution of parietal areas and their interaction with prefrontal areas to WM enhancement.

Further studies could examine the duration of effects, the likely synergistic effect of cognitive training with tDCS, or the applicability of tDCS or other NBS methods to enhance WM across the age span, from children to elderly. However, all such studies need carefully to weigh risk–benefit considerations, and should be informed by a thoughtful discussion of the ethical connotations of such enhancement approaches ( Rossi et al., 2009 ; Hamilton et al, 2011 ; Horvath et al., 2011 ).

Whether NBS can enhance STM in normal subjects is less clear. Studies show less consistent results. This could in part be due to the fact that basic STM tasks are easy for healthy subjects, which leads to ceiling effects. More recent studies have applied adapted tasks, which, however, makes it difficult to compare across studies. Most studies, similar to the literature on WM, have targeted the DLPFC. Two recent studies reported beneficial effects of tDCS with the anode over the DLPFC for an STM task with additional distractors ( Gladwin et al., 2012 ; Meiron and Lavidor, 2013 ). One study found a gender-dependent improvement in accuracy, with male subjects profiting more from left DLPFC stimulation and female subjects profiting more from right DLPFC stimulation, but only if distractor loads were high ( Meiron and Lavidor, 2013 ). The other study used a modified Sternberg task, which introduced additional distractor stimuli during the delay period ( Gladwin et al., 2012 ). These workers found significant reaction time improvements after stimulation of the left DLPFC. Compared with these studies, Marshall et al. (2005) applied tDCS with either two anodes or two cathodes over DLPFC, with the reference electrodes positioned over the mastoids, and found deleterious effects of STM. This may indicate that the introduction of distractors to an STM task changes underlying neurobiological processes and enables enhancement effects. Improvements after tDCS may be due to either improved selective attention or more successful inhibition of distracting information. Indeed, a recent TMS study has shown that the role of the DLPFC in STM tasks seems to be dependent on the presence of distractors. The stronger the distraction, the more prominent the frontoparietal interactions become, in order to protect relevant memory representations ( Feredoes et al., 2011 ).

Studies in which investigators stimulated parietal areas have yielded partly opposing results. This is true of studies using tDCS and those employing TMS. Regarding TMS experiments, some show worsened STM ( Koch et al., 2005 ; Postle et al, 2006 ), while the other report improved STM ( Hamidi et al., 2008 ; Yamanaka et al., 2010 ) after high-frequency parietal stimulation during the delay period. As for tDCS experiments, Berryhill et al. (2010) found impairment in recognition, but not free recall, after tDCS with the cathode over the right parietal cortex (and the anode over the left cheek), whereas Heimrath and coworkers (2012) , positioning the cathode over the right parietal cortex (and the anode over the contralateral homologous area), found an improved capacity in a delayed match-to-sample task after tDCS when stimuli were presented in the left visual hemifield (STM for stimuli presented in the left hemifield decreased). Interestingly, Heimrath et al. used concurrent tDCS and EEG, and found a decrease in oscillatory power in the alpha band after cathodal stimulation. As alpha activity is assumed to reflect inhibition of distractors ( Klimesch, 1999 ), the authors suggested that this measure might indicate memory performance. This study again illustrates the potential of experiments combining behavioral and neurophysiological outcome measures with NBS.

Finally, one study probed the cerebellum and found an abolishment of practice-dependent improvements in response time in a Sternberg task, regardless of whether the anode or the cathode was placed over the cerebellum (and the other electrode over the vertex) ( Ferrucci et al., 2008 ). The contribution of the cerebellum to STM was also probed with single-pulse TMS by Desmond and colleagues (2005) , who also found a negative effect on response time in the Sternberg task. Whether other cerebellar stimulation paradigms can induce an enhancement of STM remains unexplored.

G eneral memory and learning

Researchers attempting to enhance learning processes have targeted various neural regions. Such diverse approaches again render it difficult to single out a pattern regarding stimulatory condition, mechanisms, and outcome. Most studies have applied tDCS during the learning phase, and most have targeted the left DLPFC or other left prefrontal areas. Generally, studies report memory improvement following tDCS with the anode over DLPFC ( Kincses et al., 2004 ; Javadi and Walsh, 2012 ; Javadi et al., 2012 ) or other prefrontal areas ( De Vries et al., 2010 ), and worsening memory after tDCS with the cathode over DLPFC ( Elmer et al., 2009 ; Hammer et al., 2011 ; Javadi and Walsh, 2012 ; Javadi et al., 2012 ) or other prefrontal areas ( Vines et al., 2006 ). However, in interpreting their results, investigators have often made the overly simplistic assumption that the effects of tDCS can be accounted for by the neurobiological effect of one of the electrodes, the anode enhancing and the cathode suppressing activity in the brain area under them. Yet, it is important to remember that tDCS is not monopolar and that all electrodes are active. Thus the brain is exposed to a flow of current with opposite faradizing effects of the anode and the cathode. Therefore, to speak of anodal tDCS or cathodal tDCS is inaccurate.

Few studies have targeted right prefrontal areas. One study reported no effects in an episodic verbal memory task after tDCS with either anode or cathode over the right prefrontal region ( Elmer et al., 2009 ). Two studies showed that the learning process of threat detection in a virtual reality environment and the time required to learn this skill can be improved following tDCS with the anode over the right prefrontal ( Bullard et al., 2011 ; Clark et al., 2012 ) or right parietal region ( Clark et al., 2012 ). Furthermore, Bullard and colleagues (2011) found that applying tDCS at the beginning of the learning phase significantly enhanced learning in comparison with findings in experienced learners (after 1 hour of training).

Bilateral stimulation (anode and cathode over homologous areas of either hemisphere) has been applied in a few studies ( Marshall et al., 2004 , 2011 ; Boggio et al., 2009 ; Chi et al., 2010 ; Cohen Kadosh et al., 2010 ; Penolazzi et al., 2010 ; Jacobson et al., 2012 ). Jacobson and coworkers (2012) applied bilateral tDCS (anodal left, cathodal right, or vice versa) over the parietal lobe during encoding. They found improved verbal memory only when the anode was placed over the left hemisphere and the cathode over the right hemisphere. Another study investigating the contribution of the parietal cortex to numerical learning applied bilateral tDCS during a training phase of 6 days ( Cohen Kadosh et al., 2010 ). While right-anodal/left-cathodal stimulation improved learning significantly, right-cathodal/left-anodal stimulation decreased learning compared with sham tDCS.

Penolazzi and colleagues (2010) applied bilateral tDCS (anode left and cathode right, or vice versa) over the frontotemporal cortex during encoding and found facilitated recall of pleasant images after right-anodal/left-cathodal tDCS, whereas left-anodal/right-cathodal tDCS facilitated recall of unpleasant images. These results support a theoretical model (specific valence hypothesis) according to which the right and left hemispheres are specialized in the processing of unpleasant and pleasant stimuli respectively. Another group applying bilateral stimulation (anodal left, cathodal right, or vice versa) over the anterior temporal lobe assessed visual memory ( Chi et al., 2010 ) and also reported an improvement in memorizing different types of shape after right-anodal/left-cathodal stimulation, but no effects when applying an inverse stimulation pattern.

One set of studies has investigated effects of bilateral anodal stimulation over DLPFC during sleep and wakefulness. In their first study, Marshall and colleagues (2004) reported an improvement of memory consolidation when applying intermittent (on/off 15 seconds) anodal tDCS simultaneously over both DLPFCs during slow-wave (nonrapid eye movement, non-REM) sleep but not during wakefulness. In a second study they investigated state-dependent effects, and found enhanced theta activity when transcranial slow oscillation stimulation (tSOS) was applied during wakefulness ( Kirov et al., 2009 ). Memory enhancement occurred only when tSOS was applied during learning, but not after learning. In their third study, Marshall and colleagues (2011) applied anodal theta-tDCS (tDCS oscillating at 5 Hz) during REM sleep and non-REM sleep, which led to increased gamma-band activity and decreased memory consolidation respectively. The data from these studies illustrate the potential of transcranial current stimulation at specific stimulation frequencies selectively to modulate specific brain oscillations. This NBS method provides an interesting approach for investigating the relation between cortical brain rhythms, sleep-related processes, and memory functions.

Some studies have reported apparently contradictory results, highlighting the need for further investigation of the mechanisms of action underlying tDCS and TMS. Boggio et al. (2009) found decreased “false memories” utilizing anodal tDCS over the left anterior temporal lobe, or bilateral (left-anodal/right-cathodal) tDCS. However, the same researchers reported a nearly identical effect after applying 1-Hz rTMS over the same region, a protocol that is believed to suppress activity of the targeted brain area ( Gallate et al., 2009 ). Of course, it is possible that the behavioral effect might be related to trans-synaptic network effects, rather than being mediated by the targeted brain region. Indeed, a study using single-pulse TMS reported a facilitatory effect on verbal memory after stimulating the right inferior PFC ( Kahn et al., 2005 ), presumably due to interhemispheric paradoxical facilitation effects. This would be consistent with another study that found an improvement in verbal memory after stimulating the left inferior PFC with 7-Hz rTMS bursts ( Köhler et al., 2004 ). Furthermore, a paired-pulse protocol known to induce facilitatory effects led to memory improvements after stimulation of the left and right DLPFC in verbal as well as nonverbal episodic memory. The combination of stimulation techniques and other methods, such as EEG and fMRI, allows their inherent advantages to be combined to help answer these open questions.

Elderly healthy subjects

Basic memory research includes mostly young and healthy subjects. However, one of the key topics in the domain of NBS research concerns the changes of interhemispheric balance and the increased compensatory recruitment of brain areas with aging. As memory represents an overarching topic for the elderly, it is crucial to promote research that investigates these changes and provides information as to how to enhance memory functions. Furthermore, research with healthy elderly subjects is vital if we want to translate it into the clinical setting, as patients with memory deficits are mostly older. A newly emerging field has started to investigate memory enhancement in elderly subjects and underlying models ( Rossi et al., 2004 ; Solé-Padullés et al., 2006 ; Manenti et al., 2011 ; Flöel et al., 2012 ).

The “Hemispheric Asymmetry Reduction in Older Adults” (HAROLD) model states that prefrontal activity during cognitive performance becomes less lateralized with advancing age ( Cabeza, 2002 ). Manenti and colleagues investigated the differential assumptions of the HERA model (young subjects) and the HAROLD model (elderly subjects), suggesting that hemispheric asymmetry is reduced with age. Interestingly, they could show that low-performing elderly subjects continue showing prefrontal asymmetry, whereas high-performing elderly individuals show reduced asymmetry indicative of compensatory mechanisms ( Manenti et al., 2011 ).

Although lateralized activations within the PFC can be observed in younger subjects during episodic memory tasks ( Rossi et al., 2001 ), this asymmetry vanishes progressively with advancing age, as indicated by bilateral interference effects ( Rossi et al., 2004 ).

Conversely, the predominance of left DLPFC effect during encoding was not abolished in older subjects, indicating its causal role for encoding along the lifespan. However, this study did not differentiate between high- and low-performing subjects. Another study supported the assumption that higher performance is associated with more bilateral recruitment of brain areas and that stimulation may be able to promote the recruitment of additional brain areas to compensate for age-related decline. Solé-Padullés and colleagues (2006) found improved performance in associative learning after 5-Hz offline rTMS, which was accompanied by additional recruitment of right prefrontal and bilateral posterior brain regions.

A tDCS study showed improvements in spatial learning and memory in elderly subjects (mean 62 years) when stimulating during encoding ( Flöel et al., 2012 ). Anodal stimulation over the right temporoparietal cortex improved free recall 1 week later compared with sham stimulation. No immediate learning differences were observed, which indicates that retention (less decay) rather than encoding was affected by the stimulation.

To summarize, several studies have found different results following the stimulation of the DLPFC in young and elderly healthy subjects in accordance with the HAROLD model ( Cabeza, 2002 ). These differences could be due to changes in interhemispheric balance and recruitment of different brain areas for the same tasks, which could arise due to compensatory mechanisms. It remains to be further elucidated whether these changes reflect local or distributed mechanisms, whether compensatory recruitment of additional brain areas is associated with higher performance levels and could be enhanced by NBS.

Compared with the wealth of studies that have been done with healthy and mostly young subjects, studies on patients are rather sparse (see Table 55.1 ). The evidence is encouraging and calls for further investigation of the combined application of NBS and neuropsychological therapy. Besides behavioral measures, these studies should ideally include other measurements, such as assessment of brain plasticity or memory-specific neurophysiological outcomes. The work on patients with stroke is very preliminary, and more studies with larger patient numbers and better control of lesion location are needed. In one crossover, sham-controlled study, Jo et al. (2009) applied tDCS with the anode over the left DLPFC (and the cathode over the contralateral supraorbital area) in a 2-back task to 10 patients with unilateral, right-hemispheric, ischemic, or hemorrhagic strokes (1–4 months poststroke). After a single stimulation session, performance accuracy but not reaction time improved significantly. Enhancement of memory functions has been more extensively investigated in patients with AD and Parkinson’s disease (PD). These findings provide evidence that NBS could be a safe and useful tool in restoring/compensating brain functions through activation of primary and compensatory networks that underlie memory functions.

A lzheimer’s disease

A few studies have demonstrated effects of NBS on cognitive functions in AD (6 TMS, 3 tDCS). The first studies that used NBS in AD looked primarily at language and not memory functions. Cotelli and colleagues used rTMS (20 Hz) over the left and right DLPFC and reported positive effects for both hemispheres. They applied a single online session of rTMS in two crossover, sham-controlled studies ( Cotelli et al., 2006 , 2008 ). In the first study they reported improved accuracy in action naming, but not object naming, for all patients ( Cotelli et al., 2006 ). In the second study they could replicate the positive results for action naming; however, object naming also improved significantly, although only in moderately to severely impaired patients ( Cotelli et al., 2008 ). The authors hypothesized that the lack of improvement in object naming may be due to a ceiling effect. Furthermore, the bilateral effect could have been due to compensatory activation of right hemispheric resources.

In a third placebo-controlled study the same authors tested various functions, including memory, executive functions, and language in patients with moderate AD ( Cotelli et al., 2011 ). This study entailed 4 weeks of daily sessions of 20-Hz rTMS to the left DLPFC. Although they found significant improvements in sentence comprehension after 10 sessions (with no further improvement after 20 sessions), they did not find any improvements in memory and executive functions ( Cotelli et al., 2011 ). This lack of improvement could be due to the fact that the patients were not doing any specific concomitant cognitive training. Alternatively, the lack of memory effects could be related to the targeted brain region.

Bentwich and colleagues (2011) interleaved cognitive training and rTMS (10 Hz) during 30 sessions while stimulating six different brain regions (Broca, Wernicke, right and left DLPFC and parietal cortices). During each session three of these regions were stimulated while patients did cognitive tasks that were developed to fit each of these regions. Improvements in cognitive functions were significant, as measured using the cognitive subscale of the Alzheimer’s Disease Assessment Scale (ADAS-Cog), and were maintained for 4.5 months after the training. A case report ( Haffen et al., 2012 ) showed an improvement in episodic memory (free recall) and processing speed following 10 sessions of rTMS (10 Hz) over the left DLPFC. These are open trials and, obviously, sham-controlled interventions are needed. However, the results are promising and warrant follow-up. In a sham-controlled trial, Ahmed and colleagues (2012) assigned 45 patients with AD to three different treatment groups to study the effects of high- or low-frequency rTMS (20 Hz, 1 Hz), or sham stimulation. Patients received treatment on 5 consecutive days without combined cognitive training. Mildly to moderately impaired patients receiving high-frequency rTMS improved significantly on all scales (Mini Mental State Examination (MMSE), Instrumental Daily Living Activity Scale, Geriatric Depression Scale), and maintained these improvements for 3 months. However, severely impaired patients did not respond to the treatment.

Two crossover studies applied tDCS for one session and reported improvements in visual recognition memory following stimulation of the left DLPFC and temporoparietal cortex (TPC) ( Boggio et al., 2009 ), and in word recognition following stimulation of the bilateral TPC ( Ferrucci et al., 2008 ). In the first study, the authors applied 15 minutes of anodal, cathodal, and sham stimulation over bilateral TPC on three different sessions in patients with mild AD. While anodal tDCS led to an improvement, cathodal stimulation led to impairments in word recognition. No effects were observed in a visual attention task ( Ferrucci et al., 2008 ). In the second study, mildly to moderately impaired AD patients received anodal tDCS over the left DLPFC, the left TPC, or sham stimulation. Stimulation over both DLFPC and TPC resulted in a significant improvement in visual recognition. No effects were observed on selective attention or a visual delayed match-to-sample task.

Possibly, tDCS-induced changes in cholinergic activity contributed to these improvements. A recent study reported a significant change of SAI (ISI 2 ms) in the motor cortex of healthy subjects after anodal stimulation, while the resting motor threshold and amplitudes of motor evoked potentials did not change ( Scelzo et al., 2011 ). This could explain the positive impact of tDCS on memory functions in the above-mentioned studies. Future studies measuring behavioral along with neurophysiological effects and exploring correlations between them would be desirable.

P arkinson’s disease

Two studies have applied TMS or tDCS with the aim of improving cognitive functioning in PD. The first study compared the effects of active or sham rTMS and fluoxetine or placebo in patients with PD with concurrent depression ( Boggio et al., 2005 ). The authors applied 15-Hz rTMS over the left DLPFC for 10 daily sessions, and assessed cognitive functions at baseline, and 2 and 8 weeks after the treatment. Treatments were not combined with cognitive training or psychotherapy. After 2 weeks both interventions led to similar improvements in the Stroop Test and the Wisconsin Card Sorting Test (executive functions), and the Hooper (visuospatial functions). Furthermore, depression rates improved significantly in both groups. However, no improvements were reported in STM or WM (digits forward and backward). Eight weeks after treatment, these improvements declined slightly but remained significant.

The second study found improved accuracy in a 3-back task during a single session of anodal tDCS over the left DLPFC. Improvement was significant at a stimulation intensity of 2 mA but not at 1 mA ( Boggio et al., 2006 ).

Cognitive impairments in PD are often associated with depression symptoms, which occur in about 35% of patients. Furthermore, dementia is common in these patients with a point prevalence of 30% ( Aarsland and Kurz, 2010 ). Further studies are needed to investigate underlying processes leading to cognitive impairments. Moreover, studies should evaluate the efficacy of repetitive NBS in combination with cognitive training for this patient population.

A quickly growing number of studies is using NBS applications to study the underlying neurobiological substrates of memory functions, to investigate the use of TMS as a diagnostic tool, and the application of NBS to enhance memory functions. To date, most studies have used TMS to probe underlying memory processes and their causal and temporal relationships, whereas TMS, tDCS, and other forms of transcranial current stimulation are being used to enhance memory functions in healthy as well as patient populations. The combination of NBS with other methods, such as EEG and fMRI, enables the measurement of behavioral along with neurophysiological effects; the exploration of correlations between them is desirable to advance our neurobiological understanding and optimize future interventions.

ACKNOWLEDGMENTS

A.P.-L. serves on the scientific advisory boards for Nexstim, Neuronix, Starlab Neuroscience, Neosync, and Novavision, and is listed as an inventor on several issued and pending patents on the real-time integration of transcranial magnetic stimulation (TMS) with electroencephalography (EEG) and magnetic resonance imaging (MRI). Work on this study was supported by grants from the National Center for Research Resources: Harvard Clinical and Translational Science Center/Harvard Catalyst (UL1 RR025758), and investigator-initiated grants from Nexstim Inc. and Neuronix. A.-K.B. was supported by the Young Academics Support of the University of Zurich, Switzerland. K.R. was supported by the Dean’s Summer Research Award Grant, Harvard University.

R eferences

• Aarsland D, Kurz MW. The epidemiology of dementia associated with Parkinson disease. J Neurol Sci. 2010; 289 :18–22. [ PubMed ] [ Google Scholar ]
• Ahmed MA, Darwish ES, Khedr EM, et al. Effects of low versus high frequencies of repetitive transcranial magnetic stimulation on cognitive function and cortical excitability in Alzheimer’s dementia. J Neurol. 2012; 259 :83–92. [ PubMed ] [ Google Scholar ]
• Albouy G, Sterpenich V, Balteau E, et al. Both the hippocampus and striatum are involved in consolidation of motor sequence memory. Neuron. 2008; 58 :261–272. [ PubMed ] [ Google Scholar ]
• Andrews SC, Hoy KE, Enticott PG, et al. Improving working memory: the effect of combining cognitive activity and anodal transcranial direct current stimulation to the left dorsolateral prefrontal cortex. Brain Stimul. 2011; 4 :84–89. [ PubMed ] [ Google Scholar ]
• Atance CM, O’Neill DK. Episodic future thinking. Trends Cogn Sci. 2001; 5 :533–539. [ PubMed ] [ Google Scholar ]
• Baddeley AD. Working Memory. Clarendon; Oxford: 1986. [ Google Scholar ]
• Baddeley AD. The episodic buffer: a new component of working memory? Trends Cogn Sci. 2000; 4 :417–423. [ PubMed ] [ Google Scholar ]
• Baddeley AD, Hitch G. Working memory. In: Bower GA, editor. Recent Advances in Learning and Motivation. Vol. 8. Academic Press; New York: 1974. pp. 47–90. [ Google Scholar ]
• Bailey N, Areni CS. Background music as a quasi clock in retrospective duration judgments. Percept Mot Skills. 2006; 102 :435–444. [ PubMed ] [ Google Scholar ]
• Basso D, Ferrari M, Palladino P. Prospective memory and working memory: asymmetrical effects during frontal lobe TMS stimulation. Neuropsychologia. 2010; 48 :3282–3290. [ PubMed ] [ Google Scholar ]
• Beckers G, Hömberg V. Impairment of visual perception and visual short term memory scanning by transcranial magnetic stimulation of occipital cortex. Exp Brain Res. 1991; 87 :421–432. [ PubMed ] [ Google Scholar ]
• Bentwich J, Dobronevsky E, Aichenbaum S, et al. Beneficial effect of repetitive transcranial magnetic stimulation combined with cognitive training for the treatment of Alzheimer’s disease: a proof of concept study. J Neural Transm. 2011; 118 :463–471. [ PubMed ] [ Google Scholar ]
• Berryhill ME, Wencil EB, Branch Coslett H, et al. A selective working memory impairment after transcranial direct current stimulation to the right parietal lobe. Neurosci Lett. 2010; 479 :312–316. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Blanchet S, Gagnon G, Schneider C. The contribution of the dorsolateral prefrontal cortex in full and divided encoding: a paired-pulse transcranial magnetic stimulation study. Behav Neurol. 2010; 23 :107–115. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Boggio PS, Fregni F, Bermpohl F, et al. Effect of repetitive TMS and fluoxetine on cognitive function in patients with Parkinson’s disease and concurrent depression. Mov Disord. 2005; 20 :1178–1184. [ PubMed ] [ Google Scholar ]
• Boggio PS, Ferrucci R, Rigonatti SP, et al. Effects of transcranial direct current stimulation on working memory in patients with Parkinson’s disease. J Neurol Sci. 2006; 249 :31–38. [ PubMed ] [ Google Scholar ]
• Boggio PS, Khoury LP, Martins DC, et al. Temporal cortex direct current stimulation enhances performance on a visual recognition memory task in Alzheimer disease. J Neurol Neurosurg Psychiatry. 2009; 80 :444–447. [ PubMed ] [ Google Scholar ]
• Boggio PS, Ferrucci R, Mameli F, et al. Prolonged visual memory enhancement after direct current stimulation in Alzheimer’s disease. Brain Stimul. 2012; 5 :223–230. [ PubMed ] [ Google Scholar ]
• Boyer P. Evolutionary economics of mental time travel? Trends Cogn Sci. 2008; 12 :219–224. [ PubMed ] [ Google Scholar ]
• Brashers-Krug T, Shadmehr R, Bizzi E. Consolidation in human motor memory. Nature. 1996; 382 :252–255. [ PubMed ] [ Google Scholar ]
• Buckner RL, Koutstaal W, Schacter DL, et al. Functional-anatomic study of episodic retrieval using fMRI. I. Retrieval effort versus retrieval success. Neuroimage. 1998; 7 :151–162. [ PubMed ] [ Google Scholar ]
• Bullard LM, Browning ES, Clark VP, et al. Transcranial direct current stimulation’s effect on novice versus experienced learning. Exp Brain Res. 2011; 213 :9–14. [ PubMed ] [ Google Scholar ]
• Buonomano DV. Decoding temporal information: a model based on short-term synaptic plasticity. J Neurosci. 2000; 20 :1129–1141. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Burgess PW, Gonen-Yaacovi G, Volle E. Functional neuroimaging studies of prospective memory: what have we learnt so far? Neuropsychologia. 2011; 49 :2246–2257. [ PubMed ] [ Google Scholar ]
• Bütefisch CM, Khurana V, Kopylev L, et al. Enhancing encoding of a motor memory in the primary motor cortex by cortical stimulation. J Neurophysiol. 2004; 91 :2110–2116. [ PubMed ] [ Google Scholar ]
• Cabeza R. Hemispheric asymmetry reduction in older adults: the HAROLD model. Psychol Aging. 2002; 17 :85–100. [ PubMed ] [ Google Scholar ]
• Cabeza R, Nyberg L. Neural bases of learning and memory: functional neuroimaging evidence. Curr Opin Neurol. 2000; 13 :415–421. [ PubMed ] [ Google Scholar ]
• Cabeza R, Ciaramelli E, Olson IR, et al. The parietal cortex and episodic memory: an attentional account. Nat Rev Neurosci. 2008; 9 :613–625. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Cabeza R, Mazuz YS, Stokes J, et al. Overlapping parietal activity in memory and perception: evidence for the attention to memory model. J Cogn Neurosci. 2011; 23 :3209–3217. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Cattaneo Z, Vecchi T, Pascual-Leone A, et al. Contrasting early visual cortical activation states causally involved in visual imagery and short-term memory. Eur J Neurosci. 2009; 30 :1393–1400. [ PubMed ] [ Google Scholar ]
• Cave CB, Squire LR. Intact verbal and nonverbal short-term memory following damage to the human hippocampus. Hippocampus. 1992; 2 :151–163. [ PubMed ] [ Google Scholar ]
• Censor N, Cohen LG. Using repetitive transcranial magnetic stimulation to study the underlying neural mechanisms of human motor learning and memory. J Physiol. 2011; 589 :21–28. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Chi RP, Fregni F, Snyder AW. Visual memory improved by non-invasive brain stimulation. Brain Res. 2010; 1353 :168–175. [ PubMed ] [ Google Scholar ]
• Ciaramelli E, Grady CL, Moscovitch M. Top-down and bottom-up attention to memory: a hypothesis (AtoM) on the role of the posterior parietal cortex in memory retrieval. Neuropsychologia. 2008; 46 :1828–1851. [ PubMed ] [ Google Scholar ]
• Cicerone KD, Dahlberg C, Kalmar K, et al. Evidence-based cognitive rehabilitation: recommendations for clinical practice. Arch Phys Med Rehabil. 2000; 81 :1596–1615. [ PubMed ] [ Google Scholar ]
• Clark VP, Coffman BA, Mayer AR, et al. TDCS guided using fMRI significantly accelerates learning to identify concealed objects. Neuroimage. 2012; 59 :117–128. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Classen J, Liepert J, Wise SP, et al. Rapid plasticity of human cortical movement representation induced by practice. J Neurophysiol. 1998; 79 :1117–1123. [ PubMed ] [ Google Scholar ]
• Cohen NJ, Eichenbaum H. The theory that wouldn’t die: a critical look at the spatial mapping theory of hippocampal function. Hippocampus. 1991; 1 :265–268. [ PubMed ] [ Google Scholar ]
• Cohen NJ, Squire LR. Preserved learning and retention of pattern-analyzing skill in amnesia: dissociation of knowing how and knowing that. Science. 1980; 210 :207–210. [ PubMed ] [ Google Scholar ]
• Cohen Kadosh R, Soskic S, Iuculano T, et al. Modulating neuronal activity produces specific and long-lasting changes in numerical competence. Curr Biol. 2010; 20 :2016–2020. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Collette F, Hogge M, Salmon E, et al. Exploration of the neural substrates of executive functioning by functional neuroimaging. Neuroscience. 2006; 139 :209–221. [ PubMed ] [ Google Scholar ]
• Costa A, Oliveri M, Barban F, et al. Keeping memory for intentions: a cTBS investigation of the frontopolar cortex. Cereb Cortex. 2011; 21 :2696–2703. [ PubMed ] [ Google Scholar ]
• Cotelli M, Manenti R, Cappa SF, et al. Effect of transcranial magnetic stimulation on action naming in patients with Alzheimer disease. Arch Neurol. 2006; 63 :1602–1604. [ PubMed ] [ Google Scholar ]
• Cotelli M, Manenti R, Cappa SF, et al. Transcranial magnetic stimulation improves naming in Alzheimer disease patients at different stages of cognitive decline. Eur J Neurol. 2008; 15 :1286–1292. [ PubMed ] [ Google Scholar ]
• Cotelli M, Calabria M, Manenti R, et al. Improved language performance in Alzheimer disease following brain stimulation. J Neurol Neurosurg Psychiatry. 2011; 82 :794–797. [ PubMed ] [ Google Scholar ]
• Coyle JT, Price DL, DeLong MR. Alzheimer’s disease: a disorder of cortical cholinergic innervation. Science. 1983; 219 :1184–1190. [ PubMed ] [ Google Scholar ]
• Davies P, Maloney AJ. Selective loss of central cholinergic neurons in Alzheimer’s disease. Lancet. 1976; 2 :1403. [ PubMed ] [ Google Scholar ]
• De Gennaro L, Fratello F, Marzano C, et al. Cortical plasticity induced by transcranial magnetic stimulation during wakefulness affects electroencephalogram activity during sleep. PLoS One. 2008; 3 :e2483. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Desmond JE, Chen SHA, Shieh PB. Cerebellar transcranial magnetic stimulation impairs verbal working memory. Ann Neurol. 2005; 58 :553–560. [ PubMed ] [ Google Scholar ]
• de Vries MH, Barth ACR, Maiworm S, et al. Electrical stimulation of Broca’s area enhances implicit learning of an artificial grammar. J Cogn Neurosci. 2010; 22 :2427–2436. [ PubMed ] [ Google Scholar ]
• De Weerd P, Reithler J, van de Ven V, et al. Posttraining transcranial magnetic stimulation of striate cortex disrupts consolidation early in visual skill learning. J Neurosci. 2012; 32 :1981–1988. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• D’Esposito M. From cognitive to neural models of working memory. Philos Trans R Soc Lond B Biol Sci. 2007; 362 :761–772. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Devlin JT, Watkins KE. Stimulating language: insights from TMS. Brain. 2007; 130 :610–622. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Di Lazzaro V, Oliviero A, Profice P, et al. Muscarinic receptor blockade has differential effects on the excitability of intracortical circuits in the human motor cortex. Exp Brain Res. 2000; 135 :455–461. [ PubMed ] [ Google Scholar ]
• Di Lazzaro V, Pilato F, Dileone M, et al. In vivo cholinergic circuit evaluation in frontotemporal and Alzheimer dementias. Neurology. 2006; 66 :1111–1113. [ PubMed ] [ Google Scholar ]
• Di Lazzaro V, Pilato F, Dileone M, et al. In vivo functional evaluation of central cholinergic circuits in vascular dementia. Clin Neurophysiol. 2008; 119 :2494–2500. [ PubMed ] [ Google Scholar ]
• Dobbins IG, Foley H, Schacter DL, et al. Executive control during episodic retrieval: multiple prefrontal processes subserve source memory. Neuron. 2002; 35 :989–996. [ PubMed ] [ Google Scholar ]
• Dudai Y. The neurobiology of consolidations, or, how stable is the engram? Annu Rev Psychol. 2004; 55 :51–86. [ PubMed ] [ Google Scholar ]
• Eldaief MC, Halko MA, Buckner RL, et al. Transcranial magnetic stimulation modulates the brain’s intrinsic activity in a frequency-dependent manner. Proc Natl Acad Sci U S A. 2011; 108 :21229–21234. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Eldridge LL, Engel SA, Zeineh MM, et al. A dissociation of encoding and retrieval processes in the human hippocampus. J Neurosci. 2005; 25 :3280–3286. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Ellis J, Kvavilashvili L, Milne A. Experimental tests of prospective remembering: the influence of cue-event frequency on performance. Br J Psychol. 1999; 90 :9–23. [ PubMed ] [ Google Scholar ]
• Elmer S, Burkard M, Renz B, et al. Direct current induced short-term modulation of the left dorsolateral prefrontal cortex while learning auditory presented nouns. Behav Brain Funct. 2009; 5 :29. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Epstein CM, Sekino M, Yamaguchi K, et al. Asymmetries of prefrontal cortex in human episodic memory: effects of transcranial magnetic stimulation on learning abstract patterns. Neurosci Lett. 2002; 320 :5–8. [ PubMed ] [ Google Scholar ]
• Feredoes E, Heinen K, Weiskopf N, et al. Causal evidence for frontal involvement in memory target maintenance by posterior brain areas during distracter interference of visual working memory. Proc Natl Acad Sci U S A. 2011; 108 :17510–17515. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Ferrucci R, Mameli F, Guidi I, et al. Transcranial direct current stimulation improves recognition memory in Alzheimer disease. Neurology. 2008; 71 :493–498. [ PubMed ] [ Google Scholar ]
• Fletcher PC, Henson RN. Frontal lobes and human memory: insights from functional neuroimaging. Brain. 2001; 124 :849–881. [ PubMed ] [ Google Scholar ]
• Flöel A, Poeppel D, Buffalo EA, et al. Prefrontal cortex asymmetry for memory encoding of words and abstract shapes. Cereb Cortex. 2004; 14 :404–409. [ PubMed ] [ Google Scholar ]
• Flöel A, Suttorp W, Kohl O, et al. Non-invasive brain stimulation improves object-location learning in the elderly. Neurobiol Aging. 2012; 33 :1682–1689. [ PubMed ] [ Google Scholar ]
• Fregni F, Boggio PS, Nitsche M, et al. Anodal transcranial direct current stimulation of prefrontal cortex enhances working memory. Exp Brain Res. 2005; 166 :23–30. [ PubMed ] [ Google Scholar ]
• Freitas C, Mondragón-Llorca H, Pascual-Leone A. Noninvasive brain stimulation in Alzheimer’s disease: systematic review and perspectives for the future. Exp Gerontol. 2011a; 46 :611–627. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Freitas C, Perez J, Knobel M, et al. Changes in cortical plasticity across the lifespan. Front Aging Neurosci. 2011b; 3 :5. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Gagnon G, Blanchet S, Grondin S, et al. Paired-pulse transcranial magnetic stimulation over the dorsolateral prefrontal cortex interferes with episodic encoding and retrieval for both verbal and non-verbal materials. Brain Res. 2010; 1344 :148–158. [ PubMed ] [ Google Scholar ]
• Gagnon G, Schneider C, Grondin S, et al. Enhancement of episodic memory in young and healthy adults: a paired-pulse TMS study on encoding and retrieval performance. Neurosci Lett. 2011; 488 :138–142. [ PubMed ] [ Google Scholar ]
• Gallate J, Chi R, Ellwood S, et al. Reducing false memories by magnetic pulse stimulation. Neurosci Lett. 2009; 449 :151–154. [ PubMed ] [ Google Scholar ]
• Gazzaley A, Nobre AC. Top-down modulation: bridging selective attention and working memory. Trends Cogn Sci. 2012; 16 :129–135. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Gialanella B. Aphasia assessment and functional outcome prediction in patients with aphasia after stroke. J Neurol. 2011; 258 :343–349. [ PubMed ] [ Google Scholar ]
• Gladwin TE, den Uyl TE, Fregni FF, et al. Enhancement of selective attention by tDCS: interaction with interference in a Sternberg task. Neurosci Lett. 2012; 512 :33–37. [ PubMed ] [ Google Scholar ]
• Grafman J, Pascual-Leone A, Alway D, et al. Induction of a recall deficit by rapid-rate transcranial magnetic stimulation. Neuroreport. 1994; 5 :1157–1160. [ PubMed ] [ Google Scholar ]
• Haffen E, Chopard G, Pretalli J-B, et al. A case report of daily left prefrontal repetitive transcranial magnetic stimulation (rTMS) as an adjunctive treatment for Alzheimer disease. Brain Stimul. 2012; 5 :264–266. [ PubMed ] [ Google Scholar ]
• Hallett M. Transcranial magnetic stimulation and the human brain. Nature. 2000; 406 :147–150. [ PubMed ] [ Google Scholar ]
• Halsband U, Lange RK. Motor learning in man: a review of functional and clinical studies. J Physiol Paris. 2006; 99 :414–424. [ PubMed ] [ Google Scholar ]
• Hamidi M, Tononi G, Postle BR. Evaluating frontal and parietal contributions to spatial working memory with repetitive transcranial magnetic stimulation. Brain Res. 2008; 1230 :202–210. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Hamidi M, Tononi G, Postle BR. Evaluating the role of prefrontal and parietal cortices in memory-guided response with repetitive transcranial magnetic stimulation. Neuropsychologia. 2009; 47 :295–302. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Hamilton R, Messing S, Chatterjee A. Rethinking the thinking cap: ethics of neural enhancement using noninvasive brain stimulation. Neurology. 2011; 76 :187–193. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Hammer A, Mohammadi B, Schmicker M, et al. Errorless and errorful learning modulated by transcranial direct current stimulation. BMC Neurosci. 2011; 12 :72. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Hannula DE, Tranel D, Cohen NJ. The long and the short of it: relational memory impairments in amnesia, even at short lags. J Neurosci. 2006; 26 :8352–8359. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Hannula H, Neuvonen T, Savolainen P, et al. Increasing top-down suppression from prefrontal cortex facilitates tactile working memory. Neuroimage. 2010; 49 :1091–1098. [ PubMed ] [ Google Scholar ]
• Hartje W, Poeck K. Klinische Neuropsychologie. 5th Thieme; Stuttgart: 2002. [ Google Scholar ]
• Haxby JV, Petit L, Ungerleider LG, et al. Distinguishing the functional roles of multiple regions in distributed neural systems for visual working memory. Neuroimage. 2000; 11 :145–156. [ PubMed ] [ Google Scholar ]
• Heimrath K, Sandmann P, Becke A, et al. Behavioral and electrophysiological effects of transcranial direct current stimulation of the parietal cortex in a visuo-spatial working memory task. Front Psychiatry. 2012; 3 :56. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Herwig U, Abler B, Schönfeldt-Lecuona C, et al. Verbal storage in a premotor-parietal network: evidence from fMRI-guided magnetic stimulation. Neuroimage. 2003; 20 :1032–1041. [ PubMed ] [ Google Scholar ]
• Higo T, Mars RB, Boorman ED, et al. Distributed and causal influence of frontal operculum in task control. Proc Natl Acad Sci U S A. 2011; 108 :4230–4235. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Horvath JC, Perez JM, Forrow L, et al. Transcranial magnetic stimulation: a historical evaluation and future prognosis of therapeutically relevant ethical concerns. J Med Ethics. 2011; 37 :137–143. [ PubMed ] [ Google Scholar ]
• Huber R, Ghilardi MF, Massimini M, et al. Local sleep and learning. Nature. 2004; 430 :78–81. [ PubMed ] [ Google Scholar ]
• Huber R, Ghilardi MF, Massimini M, et al. Arm immobilization causes cortical plastic changes and locally decreases sleep slow wave activity. Nat Neurosci. 2006; 9 :1169–1176. [ PubMed ] [ Google Scholar ]
• Imm J-H, Kang E, Youn T, et al. Different hemispheric specializations for pitch and audioverbal working memory. Neuroreport. 2008; 19 :99–103. [ PubMed ] [ Google Scholar ]
• Jacobson L, Goren N, Lavidor M, et al. Oppositional transcranial direct current stimulation (tDCS) of parietal substrates of attention during encoding modulates episodic memory. Brain Res. 2012; 1439 :66–72. [ PubMed ] [ Google Scholar ]
• Jaeggi SM, Buschkuehl M, Etienne A, et al. On how high performers keep cool brains in situations of cognitive overload. Cogn Affect Behav Neurosci. 2007; 7 :75–89. [ PubMed ] [ Google Scholar ]
• Jaeggi SM, Buschkuehl M, Jonides J, et al. Improving fluid intelligence with training on working memory. Proc Natl Acad Sci U S A. 2008; 105 :6829–6833. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• James W. Principles of Psychology. Henry Holt; New York: 1890. [ Google Scholar ]
• Javadi AH, Walsh V. Transcranial direct current stimulation (tDCS) of the left dorsolateral prefrontal cortex modulates declarative memory. Brain Stimul. 2012; 5 :231–241. [ PubMed ] [ Google Scholar ]
• Javadi AH, Cheng P, Walsh V. Short duration transcranial direct current stimulation (tDCS) modulates verbal memory. Brain Stimul. 2012; 5 :468–474. [ PubMed ] [ Google Scholar ]
• Jo JM, Kim YH, Ko MH, et al. Enhancing the working memory of stroke patients using tDCS. Am J Phys Med Rehabil. 2009; 88 :404–409. [ PubMed ] [ Google Scholar ]
• Jonides J, Lewis RL, Nee DE, et al. The mind and brain of short-term memory. Annu Rev Psychol. 2008; 59 :193–224. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Kahn I, Pascual-Leone A, Theoret H, et al. Transient disruption of ventrolateral prefrontal cortex during verbal encoding affects subsequent memory performance. J Neurophysiol. 2005; 94 :688–698. [ PubMed ] [ Google Scholar ]
• Karabanov A, Jin S-H, Joutsen A, et al. Timing-dependent modulation of the posterior parietal cortex–primary motor cortex pathway by sensorimotor training. J Neurophysiol. 2012; 107 :3190–3199. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Karmarkar UR, Buonomano DV. Timing in the absence of clocks: encoding time in neural network states. Neuron. 2007; 53 :427–438. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Karni A, Meyer G, Rey-Hipolito C, et al. The acquisition of skilled motor performance: fast and slow experience-driven changes in primary motor cortex. Proc Natl Acad Sci U S A. 1998; 95 :861–868. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Kessels RP, d’Alfonso AA, Postma A, et al. Spatial working memory performance after high-frequency repetitive transcranial magnetic stimulation of the left and right posterior parietal cortex in humans. Neurosci Lett. 2000; 287 :68–70. [ PubMed ] [ Google Scholar ]
• Kincses TZ, Antal A, Nitsche MA, et al. Facilitation of probabilistic classification learning by transcranial direct current stimulation of the prefrontal cortex in the human. Neuropsychologia. 2004; 42 :113–117. [ PubMed ] [ Google Scholar ]
• Kirov R, Weiss C, Siebner HR, et al. Slow oscillation electrical brain stimulation during waking promotes EEG theta activity and memory encoding. Proc Natl Acad Sci U S A. 2009; 106 :15460–15465. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Kirschen MP, Davis-Ratner MS, Jerde TE, et al. Enhancement of phonological memory following transcranial magnetic stimulation (TMS) Behav Neurol. 2006; 17 :187–194. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Klimesch W. EEG alpha and theta oscillations reflect cognitive and memory performance: a review and analysis. Brain Res Brain Res Rev. 1999; 29 :169–195. [ PubMed ] [ Google Scholar ]
• Koch G, Oliveri M, Torriero S, et al. rTMS evidence of different delay and decision processes in a fronto-parietal neuronal network activated during spatial working memory. Neuroimage. 2005; 24 :34–39. [ PubMed ] [ Google Scholar ]
• Koechlin E, Basso G, Pietrini P, et al. The role of the anterior prefrontal cortex in human cognition. Nature. 1999; 399 :148–151. [ PubMed ] [ Google Scholar ]
• Köhler S, Paus T, Buckner RL, et al. Effects of left inferior prefrontal stimulation on episodic memory formation: a two-stage fMRI-rTMS study. J Cogn Neurosci. 2004; 16 :178–188. [ PubMed ] [ Google Scholar ]
• Korman M, Doyon J, Doljansky J, et al. Daytime sleep condenses the time course of motor memory consolidation. Nat Neurosci. 2007; 10 :1206–1213. [ PubMed ] [ Google Scholar ]
• Krieg SM, Shiban E, Buchmann N, et al. Utility of presurgical navigated transcranial magnetic brain stimulation for the resection of tumors in eloquent motor areas. J Neurosurg. 2012; 116 :994–1001. [ PubMed ] [ Google Scholar ]
• Lepore FE. Harvey Cushing, Gordon Holmes, and the neurological lessons of World War I. Arch Neurol. 1994; 51 :711–722. [ PubMed ] [ Google Scholar ]
• Luber B, Kinnunen LH, Rakitin BC, et al. Facilitation of performance in a working memory task with rTMS stimulation of the precuneus: frequency- and time-dependent effects. Brain Res. 2007; 1128 :120–129. [ PubMed ] [ Google Scholar ]
• Luber B, Stanford AD, Bulow P, et al. Remediation of sleep-deprivation-induced working memory impairment with fMRI-guided transcranial magnetic stimulation. Cereb Cortex. 2008; 18 :2077–2085. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Luria AR. The Man With a Shattered World: The History of a Brain Wound. Harvard University Press; Cambridge, MA: 1972. [ Google Scholar ]
• Machizawa MG, Kalla R, Walsh V, et al. The time course of ventrolateral prefrontal cortex involvement in memory formation. J Neurophysiol. 2010; 103 :1569–1579. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Manenti R, Cotelli M, Miniussi C. Successful physiological aging and episodic memory: a brain stimulation study. Behav Brain Res. 2011; 216 :153–158. [ PubMed ] [ Google Scholar ]
• Markowitsch HJ, Staniloiu A. The impairment of recollection in functional amnesic states. Cortex. 2013; 49 :1494–1510. [ PubMed ] [ Google Scholar ]
• Marshall L, Mölle M, Hallschmid M, et al. Transcranial direct current stimulation during sleep improves declarative memory. J Neurosci. 2004; 24 :9985–9992. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Marshall L, Mölle M, Siebner HR, et al. Bifrontal transcranial direct current stimulation slows reaction time in a working memory task. BMC Neurosci. 2005; 6 :23. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Marshall L, Kirov R, Brade J, et al. Transcranial electrical currents to probe EEG brain rhythms and memory consolidation during sleep in humans. PloS One. 2011; 6 :e16905. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Martin SJ, Grimwood PD, Morris RG. Synaptic plasticity and memory: an evaluation of the hypothesis. Annu Rev Neurosci. 2000; 23 :649–711. [ PubMed ] [ Google Scholar ]
• McGaugh JL. Memory – a century of consolidation. Science. 2000; 287 :248–251. [ PubMed ] [ Google Scholar ]
• McNab F, Varrone A, Farde L, et al. Changes in cortical dopamine D1 receptor binding associated with cognitive training. Science. 2009; 323 :800–802. [ PubMed ] [ Google Scholar ]
• Meiron O, Lavidor M. Unilateral prefrontal direct current stimulation effects are modulated by working memory load and gender. Brain Stimul. 2013; 6 :440–447. [ PubMed ] [ Google Scholar ]
• Miller EK, Cohen JD. An integrative theory of prefrontal cortex function. Annu Rev Neurosci. 2001; 24 :167–202. [ PubMed ] [ Google Scholar ]
• Miniussi C, Thut G. Combining TMS and EEG offers new prospects in cognitive neuroscience. Brain Topogr. 2010; 22 :249–256. [ PubMed ] [ Google Scholar ]
• Missonnier P, Leonards U, Gold G, et al. A new electrophysiological index for working memory load in humans. Neuroreport. 2003; 14 :1451–1455. [ PubMed ] [ Google Scholar ]
• Missonnier P, Gold G, Leonards U, et al. Aging and working memory: early deficits in EEG activation of posterior cortical areas. J Neural Transm. 2004; 111 :1141–1154. [ PubMed ] [ Google Scholar ]
• Mitchell KJ, Johnson MK. Source monitoring 15 years later: what have we learned from fMRI about the neural mechanisms of source memory? Psychol Bull. 2009; 135 :638–677. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Morgan HM, Jackson MC, van Koningsbruggen MG, et al. Frontal and parietal theta burst TMS impairs working memory for visual-spatial conjunctions. Brain Stimul. 2013; 6 :122–129. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Morrison JH, Baxter MG. The ageing cortical synapse: hallmarks and implications for cognitive decline. Nat Rev Neurosci. 2012; 13 :240–250. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Moscovitch M, Nadel L, Winocur G, et al. The cognitive neuroscience of remote episodic, semantic and spatial memory. Curr Opin Neurobiol. 2006; 16 :179–190. [ PubMed ] [ Google Scholar ]
• Mottaghy FM, Krause BJ, Kemna LJ, et al. Modulation of the neuronal circuitry subserving working memory in healthy human subjects by repetitive transcranial magnetic stimulation. Neurosci Lett. 2000; 280 :167–170. [ PubMed ] [ Google Scholar ]
• Mottaghy FM, Döring T, Müller-Gärtner H-W, et al. Bilateral parieto-frontal network for verbal working memory: an interference approach using repetitive transcranial magnetic stimulation (rTMS) Eur J Neurosci. 2002a; 16 :1627–1632. [ PubMed ] [ Google Scholar ]
• Mottaghy FM, Gangitano M, Sparing R, et al. Segregation of areas related to visual working memory in the prefrontal cortex revealed by rTMS. Cereb Cortex. 2002b; 12 :369–375. [ PubMed ] [ Google Scholar ]
• Mottaghy FM, Gangitano M, Krause BJ, et al. Chronometry of parietal and prefrontal activations in verbal working memory revealed by transcranial magnetic stimulation. Neuroimage. 2003a; 18 :565–575. [ PubMed ] [ Google Scholar ]
• Mottaghy FM, Pascual-Leone A, Kemna LJ, et al. Modulation of a brain–behavior relationship in verbal working memory by rTMS. Brain Res Cogn Brain Res. 2003b; 15 :241–249. [ PubMed ] [ Google Scholar ]
• Muellbacher W, Ziemann U, Wissel J, et al. Early consolidation in human primary motor cortex. Nature. 2002; 415 :640–644. [ PubMed ] [ Google Scholar ]
• Mull BR, Seyal M. Transcranial magnetic stimulation of left prefrontal cortex impairs working memory. Clin Neurophysiol. 2001; 112 :1672–1675. [ PubMed ] [ Google Scholar ]
• Mulquiney PG, Hoy KE, Daskalakis ZJ, et al. Improving working memory: exploring the effect of transcranial random noise stimulation and transcranial direct current stimulation on the dorsolateral prefrontal cortex. Clin Neurophysiol. 2011; 122 :2384–2389. [ PubMed ] [ Google Scholar ]
• Mylius V, Jung M, Menzler K, et al. Effects of transcranial direct current stimulation on pain perception and working memory. Eur J Pain. 2012; 16 :974–982. [ PubMed ] [ Google Scholar ]
• Nadel L, Samsonovich A, Ryan L, et al. Multiple trace theory of human memory: computational, neuroimaging, and neuropsychological results. Hippocampus. 2000; 10 :352–368. [ PubMed ] [ Google Scholar ]
• Najib U, Pascual-Leone A. Paradoxical functional facilitation with noninvasive brain stimulation. In: Kapur N, Pascual-Leone A, Ramachandran V, et al., editors. The Paradoxical Brain. Cambridge University Press; Cambridge: 2011. pp. 234–260. [ Google Scholar ]
• Nitsche MA, Cohen LG, Wassermann EM, et al. Transcranial direct current stimulation: state of the art 2008. Brain Stimul. 2008; 1 :206–223. [ PubMed ] [ Google Scholar ]
• Noulhiane M, Pouthas V, Hasboun D, et al. Role of the medial temporal lobe in time estimation in the range of minutes. Neuroreport. 2007; 18 :1035–1038. [ PubMed ] [ Google Scholar ]
• Ohn SH, Park C-I, Yoo W-K, et al. Time-dependent effect of transcranial direct current stimulation on the enhancement of working memory. Neuroreport. 2008; 19 :43–47. [ PubMed ] [ Google Scholar ]
• Okuda J, Fujii T, Yamadori A, et al. Participation of the prefrontal cortices in prospective memory: evidence from a PET study in humans. Neurosci Lett. 1998; 253 :127–130. [ PubMed ] [ Google Scholar ]
• Okuda J, Fujii T, Ohtake H, et al. Differential involvement of regions of rostral prefrontal cortex (Brodmann area 10) in time- and event-based prospective memory. Int J Psychophysiol. 2007; 64 :233–246. [ PubMed ] [ Google Scholar ]
• Oliveri M, Turriziani P, Carlesimo GA, et al. Parietofrontal interactions in visual-object and visual-spatial working memory: evidence from transcranial magnetic stimulation. Cereb Cortex. 2001; 11 :606–618. [ PubMed ] [ Google Scholar ]
• Olson IR, Moore KS, Stark M, et al. Visual working memory is impaired when the medial temporal lobe is damaged. J Cogn Neurosci. 2006a; 18 :1087–1097. [ PubMed ] [ Google Scholar ]
• Olson IR, Page K, Moore KS, et al. Working memory for conjunctions relies on the medial temporal lobe. J Neurosci. 2006b; 26 :4596–4601. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Osaka N, Otsuka Y, Hirose N, et al. Transcranial magnetic stimulation (TMS) applied to left dorsolateral prefrontal cortex disrupts verbal working memory performance in humans. Neurosci Lett. 2007; 418 :232–235. [ PubMed ] [ Google Scholar ]
• Paller KA. Cross-cortical consolidation as the core defect in amnesia: prospects for hypothesis-testing with neuropsychology and neuroimaging. In: Squire LR, Schacter DL, editors. Neuropsychology of Memory. 3rd Guilford Press; New York: 2002. pp. 73–87. [ Google Scholar ]
• Paller KA, Wagner AD. Observing the transformation of experience into memory. Trends Cogn Sci. 2002; 6 :93–102. [ PubMed ] [ Google Scholar ]
• Pascual-Leone A, Hallett M. Induction of errors in a delayed response task by repetitive transcranial magnetic stimulation of the dorsolateral prefrontal cortex. Neuroreport. 1994; 5 :2517–2520. [ PubMed ] [ Google Scholar ]
• Pascual-Leone A, Gates JR, Dhuna A. Induction of speech arrest and counting errors with rapid-rate transcranial magnetic stimulation. Neurology. 1991; 41 :697–702. [ PubMed ] [ Google Scholar ]
• Pascual-Leone A, Bartres-Faz D, Keenan JP. Transcranial magnetic stimulation: studying the brain–behaviour relationship by induction of “virtual lesions” Philos Trans R Soc Lond B Biol Sci. 1999; 354 :1229–1238. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Pascual-Leone A, Walsh V, Rothwell J. Transcranial magnetic stimulation in cognitive neuroscience – virtual lesion, chronometry, and functional connectivity. Curr Opin Neurobiol. 2000; 10 :232–237. [ PubMed ] [ Google Scholar ]
• Pascual-Leone A, Freitas C, Oberman L, et al. Characterizing brain cortical plasticity and network dynamics across the age-span in health and disease with TMS-EEG and TMS-fMRI. Brain Topogr. 2011; 24 :302–315. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Penolazzi B, Di Domenico A, Marzoli D, et al. Effects of transcranial direct current stimulation on episodic memory related to emotional visual stimuli. PloS One. 2010; 5 :e10623. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Postle BR, Ferrarelli F, Hamidi M, et al. Repetitive transcranial magnetic stimulation dissociates working memory manipulation from retention functions in the prefrontal, but not posterior parietal, cortex. J Cogn Neurosci. 2006; 18 :1712–1722. [ PubMed ] [ Google Scholar ]
• Preston G, Anderson E, Silva C, et al. Effects of 10 Hz rTMS on the neural efficiency of working memory. J Cogn Neurosci. 2009; 22 :447–456. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Rami L, Gironell A, Kulisevsky J, et al. Effects of repetitive transcranial magnetic stimulation on memory subtypes: a controlled study. Neuropsychologia. 2003; 41 :1877–1883. [ PubMed ] [ Google Scholar ]
• Rammsayer TH, Hennig J, Haag A, et al. Effects of noradrenergic activity on temporal information processing in humans. Q J Exp Psychol B. 2001; 54 :247–258. [ PubMed ] [ Google Scholar ]
• Ranganath C, D’Esposito M. Medial temporal lobe activity associated with active maintenance of novel information. Neuron. 2001; 31 :865–873. [ PubMed ] [ Google Scholar ]
• Rasch B, Büchel C, Gais S, et al. Odor cues during slow-wave sleep prompt declarative memory consolidation. Science. 2007; 315 :1426–1429. [ PubMed ] [ Google Scholar ]
• Reynolds JR, West R, Braver T. Distinct neural circuits support transient and sustained processes in prospective memory and working memory. Cereb Cortex. 2009; 19 :1208–1221. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Robertson EM. The serial reaction time task: implicit motor skill learning? J Neurosci. 2007; 27 :10073–10075. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Robertson EM. From creation to consolidation: a novel framework for memory processing. PLoS Biol. 2009; 7 :e19. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Robertson EM, Cohen DA. Understanding consolidation through the architecture of memories. Neuroscientist. 2006; 12 :261–271. [ PubMed ] [ Google Scholar ]
• Robertson EM, Pascual-Leone A, Miall RC. Current concepts in procedural consolidation. Nat Rev Neurosci. 2004; 5 :576–582. [ PubMed ] [ Google Scholar ]
• Robertson EM, Press DZ, Pascual-Leone A. Off-line learning and the primary motor cortex. J Neurosci. 2005; 25 :6372–6378. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Rossi S, Cappa SF, Babiloni C, et al. Prefrontal cortex in long-term memory: an “interference” approach using magnetic stimulation. Nat Neurosci. 2001; 4 :948–952. [ PubMed ] [ Google Scholar ]
• Rossi S, Miniussi C, Pasqualetti P, et al. Age-related functional changes of prefrontal cortex in long-term memory: a repetitive transcranial magnetic stimulation study. J Neurosci. 2004; 24 :7939–7944. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Rossi S, Pasqualetti P, Zito G, et al. Prefrontal and parietal cortex in human episodic memory: an interference study by repetitive transcranial magnetic stimulation. Eur J Neurosci. 2006; 23 :793–800. [ PubMed ] [ Google Scholar ]
• Rossi S, Hallett M, Rossini PM, et al. Safety, ethical considerations, and application guidelines for the use of transcranial magnetic stimulation in clinical practice and research. Clin Neurophysiol. 2009; 120 :2008–2039. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Rost NS, Smith EE, Chang Y, et al. Prediction of functional outcome in patients with primary intracerebral hemorrhage: the FUNC score. Stroke. 2008; 39 :2304–2309. [ PubMed ] [ Google Scholar ]
• Sandrini M, Cappa SF, Rossi S, et al. The role of prefrontal cortex in verbal episodic memory: rTMS evidence. J Cogn Neurosci. 2003; 15 :855–861. [ PubMed ] [ Google Scholar ]
• Sandrini M, Rossini PM, Miniussi C. Lateralized contribution of prefrontal cortex in controlling task-irrelevant information during verbal and spatial working memory tasks: rTMS evidence. Neuropsychologia. 2008; 46 :2056–2063. [ PubMed ] [ Google Scholar ]
• Sandrini M, Fertonani A, Cohen LG, et al. Double dissociation of working memory load effects induced by bilateral parietal modulation. Neuropsychologia. 2012; 50 :396–402. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Sauseng P, Klimesch W, Heise KF, et al. Brain oscillatory substrates of visual short-term memory capacity. Curr Biol. 2009; 19 :1846–1852. [ PubMed ] [ Google Scholar ]
• Savolainen P, Carlson S, Boldt R, et al. Facilitation of tactile working memory by top-down suppression from prefrontal to primary somatosensory cortex during sensory interference. Behav Brain Res. 2011; 219 :387–390. [ PubMed ] [ Google Scholar ]
• Scelzo E, Giannicola G, Rosa M, et al. Increased short latency afferent inhibition after anodal transcranial direct current stimulation. Neurosci Lett. 2011; 498 :167–170. [ PubMed ] [ Google Scholar ]
• Schendan HE, Searl MM, Melrose RJ, et al. An FMRI study of the role of the medial temporal lobe in implicit and explicit sequence learning. Neuron. 2003; 37 :1013–1025. [ PubMed ] [ Google Scholar ]
• Schutter DJLG, van Honk J. Increased positive emotional memory after repetitive transcranial magnetic stimulation over the orbitofrontal cortex. J Psychiatry Neurosci. 2006; 31 :101–104. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Scoville WB, Milner B. Loss of recent memory after bilateral hippocampal lesions. J Neurol Neurosurg Psychiatry. 1957; 20 :11–21. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Seidler RD, Purushotham A, Kim S-G, et al. Cerebellum activation associated with performance change but not motor learning. Science. 2002; 296 :2043–2046. [ PubMed ] [ Google Scholar ]
• Shallice T, Warrington EK. Independent functioning of verbal memory stores: a neuropsychological study. Q J Exp Psychol. 1970; 22 :261–273. [ PubMed ] [ Google Scholar ]
• Shimamura AP. Episodic retrieval and the cortical binding of relational activity. Cogn Affect Behav Neurosci. 2011; 11 :277–291. [ PubMed ] [ Google Scholar ]
• Shimamura AP, Wickens TD. Superadditive memory strength for item and source recognition: the role of hierarchical relational binding in the medial temporal lobe. Psychol Rev. 2009; 116 :1–19. [ PubMed ] [ Google Scholar ]
• Sigurdardottir S, Andelic N, Roe C, et al. Cognitive recovery and predictors of functional outcome 1 year after traumatic brain injury. J Int Neuropsychol Soc. 2009; 15 :740–750. [ PubMed ] [ Google Scholar ]
• Silvanto J, Cattaneo Z. Transcranial magnetic stimulation reveals the content of visual short-term memory in the visual cortex. Neuroimage. 2010; 50 :1683–1689. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Silvanto J, Pascual-Leone A. State-dependency of transcranial magnetic stimulation. Brain Topogr. 2008; 21 :1–10. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Silvanto J, Pascual-Leone A. Why the assessment of causality in brain–behavior relations requires brain stimulation. J Cogn Neurosci. 2012; 24 :775–777. [ PubMed ] [ Google Scholar ]
• Simons JS, Spiers HJ. Prefrontal and medial temporal lobe interactions in long-term memory. Nat Rev Neurosci. 2003; 4 :637–648. [ PubMed ] [ Google Scholar ]
• Skrdlantová L, Horácek J, Dockery C, et al. The influence of low-frequency left prefrontal repetitive transcranial magnetic stimulation on memory for words but not for faces. Physiol Res. 2005; 54 :123–128. [ PubMed ] [ Google Scholar ]
• Smith EE, Jonides J. Working memory: a view from neuroimaging. Cogn Psychol. 1997; 33 :5–42. [ PubMed ] [ Google Scholar ]
• Smith EE, Jonides J. Storage and executive processes in the frontal lobes. Science. 1999; 283 :1657–1661. [ PubMed ] [ Google Scholar ]
• Solé-Padullés C, Bartrés-Faz D, Junqué C, et al. Repetitive transcranial magnetic stimulation effects on brain function and cognition among elders with memory dysfunction. A randomized sham-controlled study. Cereb Cortex. 2006; 16 :1487–1493. [ PubMed ] [ Google Scholar ]
• Soto D, Rotshtein P, Hodsoll J, et al. Common and distinct neural regions for the guidance of selection by visuo-verbal information held in memory: converging evidence from fMRI and rTMS. Hum Brain Mapp. 2012; 33 :105–120. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Squire LR. The medial temporal region and memory consolidation: a new hypothesis. In: Weingartner H, Parder E, editors. Memory Consolidation. Erlbaum; Hillsdale, NJ: 1984. pp. 185–210. [ Google Scholar ]
• Squire LR. Memory and Brain. Oxford University Press; New York: 1987. [ Google Scholar ]
• Squire LR. Memory systems of the brain: a brief history and current perspective. Neurobiol Learn Mem. 2004; 82 :171–177. [ PubMed ] [ Google Scholar ]
• Squire LR, Zola SM. Episodic memory, semantic memory, and amnesia. Hippocampus. 1998; 8 :205–211. [ PubMed ] [ Google Scholar ]
• Sternberg S. High-speed scanning in human memory. Science. 1966; 153 :652–654. [ PubMed ] [ Google Scholar ]
• Teo F, Hoy KE, Daskalakis ZJ, et al. Investigating the role of current strength in tDCS modulation of working memory performance in healthy controls. Front Psychiatry. 2011; 2 :45. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Thut G, Pascual-Leone A. A review of combined TMS-EEG studies to characterize lasting effects of repetitive TMS and assess their usefulness in cognitive and clinical neuroscience. Brain Topogr. 2010; 22 :219–232. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Tononi G, Cirelli C. Sleep and synaptic homeostasis: a hypothesis. Brain Res Bull. 2003; 62 :143–150. [ PubMed ] [ Google Scholar ]
• Tononi G, Cirelli C. Sleep function and synaptic homeostasis. Sleep Med Rev. 2006; 10 :49–62. [ PubMed ] [ Google Scholar ]
• Tulving E. Episodic memory: from mind to brain. Annu Rev Psychol. 2002; 53 :1–25. [ PubMed ] [ Google Scholar ]
• Tulving E, Kapur S, Craik FI, et al. Hemispheric encoding/ retrieval asymmetry in episodic memory: positron emission tomography findings. Proc Natl Acad Sci U S A. 1994; 91 :2016–2020. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Ungerleider LG, Doyon J, Karni A. Imaging brain plasticity during motor skill learning. Neurobiol Learn Mem. 2002; 78 :553–564. [ PubMed ] [ Google Scholar ]
• van de Ven V, Jacobs C, Sack AT. Topographic contribution of early visual cortex to short-term memory consolidation: a transcranial magnetic stimulation study. J Neurosci. 2012; 32 :4–11. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Vines BW, Schnider NM, Schlaug G. Testing for causality with transcranial direct current stimulation: pitch memory and the left supramarginal gyrus. Neuroreport. 2006; 17 :1047–1050. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Wada J, Rasmussen T. Intracarotid injection of sodium amytal for the lateralization of cerebral speech dominance. J Neurosurg. 1960; 106 :1117–1133. 1960. [ PubMed ] [ Google Scholar ]
• Wagle J, Farner L, Flekkoy K, et al. Early post-stroke cognition in stroke rehabilitation patients predicts functional outcome at 13 months. Dement Geriatr Cogn Disord. 2011; 31 :379–387. [ PubMed ] [ Google Scholar ]
• Wagner AD, Shannon BJ, Kahn I, et al. Parietal lobe contributions to episodic memory retrieval. Trends Cogn Sci. 2005; 9 :445–453. [ PubMed ] [ Google Scholar ]
• Wagner T, Valero-Cabre A, Pascual-Leone A. Noninvasive human brain stimulation. Annu Rev Biomed Eng. 2007; 9 :527–565. [ PubMed ] [ Google Scholar ]
• Wagner U, Gais S, Haider H, et al. Sleep inspires insight. Nature. 2004; 427 :352–355. [ PubMed ] [ Google Scholar ]
• Walker MP. A refined model of sleep and the time course of memory formation. Behav Brain Sci. 2005; 28 :51–64. [ PubMed ] [ Google Scholar ]
• Walker MP, Brakefield T, Morgan A, et al. Practice with sleep makes perfect: sleep-dependent motor skill learning. Neuron. 2002; 35 :205–211. [ PubMed ] [ Google Scholar ]
• Walsh V, Pascual-Leone A. Transcranial Magnetic Stimulation: A Neurochronometrics of Mind. MIT Press; Cambridge, MA: 2003. [ Google Scholar ]
• Watter S, Geffen GM, Geffen LB. The n-back as a dual-task: P300 morphology under divided attention. Psychophysiology. 2001; 38 :998–1003. [ PubMed ] [ Google Scholar ]
• Wheaton LA, Hallett M. Ideomotor apraxia: a review. J Neurol Sci. 2007; 260 :1–10. [ PubMed ] [ Google Scholar ]
• Wheeler ME, Petersen SE, Buckner RL. Memory’s echo: vivid remembering reactivates sensory-specific cortex. Proc Natl Acad Sci U S A. 2000; 97 :11125–11129. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Wittmann M. The inner experience of time. Philos Trans R Soc Lond B Biol Sci. 2009; 364 :1955–1967. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Yamanaka K, Yamagata B, Tomioka H, et al. Transcranial magnetic stimulation of the parietal cortex facilitates spatial working memory: near-infrared spectroscopy study. Cereb Cortex. 2010; 20 :1037–1045. [ PubMed ] [ Google Scholar ]
• Zaehle T, Sandmann P, Thorne JD, et al. Transcranial direct current stimulation of the prefrontal cortex modulates working memory performance: combined behavioural and electrophysiological evidence. BMC Neurosci. 2011; 12 :2. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Zanto TP, Rubens MT, Thangavel A, et al. Causal role of the prefrontal cortex in top-down modulation of visual processing and working memory. Nat Neurosci. 2011; 14 :656–661. [ PMC free article ] [ PubMed ] [ Google Scholar ]