CoChat

Decoding material-specific memory reprocessing during sleep in humans

Open access
Schönauer, M., Alizadeh, S., Jamalabadi, H., Abraham, A., Pawlizki, A., Gais, S. · 2017 · Nature Communications

The DOI matches its Crossref record; 67 of the 67 references it cites were found in a registry or an index.

Abstract

Neuronal learning activity is reactivated during sleep but the dynamics of this reactivation in humans are still poorly understood. Here we use multivariate pattern classification to decode electrical brain activity during sleep and determine what type of images participants had viewed in a preceding learning session. We find significant patterns of learning-related processing during rapid eye movement (REM) and non-REM (NREM) sleep, which are generalizable across subjects. This processing occurs in a cyclic fashion during time windows congruous to critical periods of synaptic plasticity. Its spatial distribution over the scalp and relevant frequencies differ between NREM and REM sleep. Moreover, only the strength of reprocessing in slow-wave sleep influenced later memory performance, speaking for at least two distinct underlying mechanisms between these states. We thus show that memory reprocessing occurs in both NREM and REM sleep in humans and that it pertains to different aspects of the consolidation process.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/ncomms15404

Cites

67 of 67 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Data preprocessing and MVPC analysis. (a) After artefact rejection, data from the remaining 4-s trials of 128-channel sleep EEG data was frequency transformed. To reduce the dimensionality of the data and to increase the signal-to-noise ratio, spectra were averaged over trials and neighbouring channels. Next, spectra of all channels were normalized separately to make them comparable, and a spectral sharpening filter was applied to remove the baseline spectrum and enhance differences between neighbouring frequency bins. (b) Training data were strictly separated from validation data in all MVPC analyses. Dimensionality of the data was further reduced in a two-step training procedure. Individual channel performance was determined using separate single-channel classifiers. An average of data from all channels weighted by their standalone performance was then used to train a classifier to distinguish between face and house stimulus conditions. Finally, classification was tested on independent validation data.
Figure 1 Data preprocessing and MVPC analysis. (a) After artefact rejection, data from the remaining 4-s trials of 128-channel sleep EEG data was frequency transformed. To reduce the dimensionality of the data and to increase the signal-to-noise ratio, spectra were averaged over trials and neighbouring channels. Next, spectra of all channels were normalized separately to make them comparable, and a spectral sharpening filter was applied to remove the baseline spectrum and enhance differences between neighbouring frequency bins. (b) Training data were strictly separated from validation data in all MVPC analyses. Dimensionality of the data was further reduced in a two-step training procedure. Individual channel performance was determined using separate single-channel classifiers. An average of data from all channels weighted by their standalone performance was then used to train a classifier to distinguish between face and house stimulus conditions. Finally, classification was tested on independent validation data.CC BY · PubMed Central
Figure 2 Classification results. (a) The content of a previous learning experience can be determined from sleep EEG during the second and fourth 90-min segment of the night. At these times, classification accuracy for all sleep stages is significant or approaches significance. The hatched area shows the 95% confidence interval. Classification accuracies for S4 sleep as well as REM sleep in the second sleep segment remain significant after Bonferroni–Holm correction considering all tests (S4: P=0.048, REM: P=0.014). (b) Significance was assessed using permutation tests to ensure that classification rates are higher than can be expected from data sets with random labelling of the data, that is, not containing any information. To estimate the displayed null-distribution from which exact significance levels of classification results can be determined, the MVPC analysis was repeated 1,001 times on the actual data with randomly shuffled condition labels. Dark grey areas show those randomizations during which classification accuracy on randomly labelled data exceeded accuracy obtained on correctly labelled data. (c) If classification accuracies are similar between the training and validation sets, generalizable information could be extracted and the classifier was not overfitted on the training data set. This was the case for all analyses that were significant, that is, for data from the second (circles) and fourth (stars) 90-min segments of the night. Here patterns detected in one set of subjects during classifier training can be generalized to data from a new set of subjects. Data from the first (triangles) and third (squares) 90-min segments show low training accuracy and low accuracy on validation data, indicating that the classifier could not extract information about previous learning content from these periods of the night. *P<0.05, **P<0.01, ***P<0.001.
Figure 2 Classification results. (a) The content of a previous learning experience can be determined from sleep EEG during the second and fourth 90-min segment of the night. At these times, classification accuracy for all sleep stages is significant or approaches significance. The hatched area shows the 95% confidence interval. Classification accuracies for S4 sleep as well as REM sleep in the second sleep segment remain significant after Bonferroni–Holm correction considering all tests (S4: P=0.048, REM: P=0.014). (b) Significance was assessed using permutation tests to ensure that classification rates are higher than can be expected from data sets with random labelling of the data, that is, not containing any information. To estimate the displayed null-distribution from which exact significance levels of classification results can be determined, the MVPC analysis was repeated 1,001 times on the actual data with randomly shuffled condition labels. Dark grey areas show those randomizations during which classification accuracy on randomly labelled data exceeded accuracy obtained on correctly labelled data. (c) If classification accuracies are similar between the training and validation sets, generalizable information could be extracted and the classifier was not overfitted on the training data set. This was the case for all analyses that were significant, that is, for data from the second (circles) and fourth (stars) 90-min segments of the night. Here patterns detected in one set of subjects during classifier training can be generalized to data from a new set of subjects. Data from the first (triangles) and third (squares) 90-min segments show low training accuracy and low accuracy on validation data, indicating that the classifier could not extract information about previous learning content from these periods of the night. *P<0.05, **P<0.01, ***P<0.001.CC BY · PubMed Central
Figure 3 Correlation between classification probability estimates and overnight memory consolidation during SWS sleep. The more confident the classifier was in placing each subject in the correct condition, the more positive the change in memory performance during later recall. Spearman's rho is reported. **P<0.01.
Figure 3 Correlation between classification probability estimates and overnight memory consolidation during SWS sleep. The more confident the classifier was in placing each subject in the correct condition, the more positive the change in memory performance during later recall. Spearman's rho is reported. **P<0.01.CC BY · PubMed Central
Figure 4 Time course of classification accuracy across the night. Separate analyses were performed for sleep stages S2, S4 and REM sleep. Classification performance follows an oscillatory pattern and peaks around 3 and 6 h after learning in all stages. Timing therefore is more relevant to when memory reprocessing occurs than sleep stage.
Figure 4 Time course of classification accuracy across the night. Separate analyses were performed for sleep stages S2, S4 and REM sleep. Classification performance follows an oscillatory pattern and peaks around 3 and 6 h after learning in all stages. Timing therefore is more relevant to when memory reprocessing occurs than sleep stage.CC BY · PubMed Central
Figure 5 Frequency contributions to memory reprocessing in NREM and REM sleep. (a) Discrimination weights show that in NREM sleep stages S2 and S4 spindle activity in the frequency range between 11 and 16 Hz is predictive for learning content. In REM sleep, theta, alpha and higher beta frequencies contributed more to correct classification. Slow frequencies <4 Hz were informative in all sleep stages. (b) The topography of predictive channels clearly differs between NREM and REM sleep. In NREM sleep stage S2, mainly delta and spindle frequencies contributed to correct classification. Similarly, frontal delta power and right parieto-temporal spindle activity were most informative for classification during NREM sleep stage S4, together with posterior higher frequency activity. REM sleep shows a more complex pattern. Here slow oscillations of central electrodes and frontal and temporal theta as well as occipital alpha contributed most to discrimination between learning conditions.
Figure 5 Frequency contributions to memory reprocessing in NREM and REM sleep. (a) Discrimination weights show that in NREM sleep stages S2 and S4 spindle activity in the frequency range between 11 and 16 Hz is predictive for learning content. In REM sleep, theta, alpha and higher beta frequencies contributed more to correct classification. Slow frequencies <4 Hz were informative in all sleep stages. (b) The topography of predictive channels clearly differs between NREM and REM sleep. In NREM sleep stage S2, mainly delta and spindle frequencies contributed to correct classification. Similarly, frontal delta power and right parieto-temporal spindle activity were most informative for classification during NREM sleep stage S4, together with posterior higher frequency activity. REM sleep shows a more complex pattern. Here slow oscillations of central electrodes and frontal and temporal theta as well as occipital alpha contributed most to discrimination between learning conditions.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 67 references

Cited by

Metadata

Authors
Schönauer, M., Alizadeh, S., Jamalabadi, H., Abraham, A., Pawlizki, A., & Gais, S.
Affiliations
  • Bernstein Center for Computational Neuroscience Munich (DE)
  • Ludwig-Maximilians-Universität München (DE)
Corresponding author
  • Schönauer, M.
  • Gais, S.
Venue
Nature Communications
Published
2017-05-17
Volume
8
Issue
1
Pages
15404-15404
ISSN
2041-1723 (ISSN-L)
OpenAlex
W2615649700
Cited by
139
References
67
Citations by year
2026: 10 · 2025: 9 · 2024: 21 · 2023: 17 · 2022: 17 · 2021: 11 · 2020: 19 · 2019: 13 · 2018: 18 · 2017: 4
FWCI
8.47
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Communications (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Open access LMU (Ludwid Maxmilian's Universitat Munchen) (submittedVersion)
Funders
Deutsche Forschungsgemeinschaft (GA730/3–1)
Topics
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Sleep and Wakefulness Research
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Memory and Neural Mechanisms
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Memory Processes and Influences
Keywords
sleep-dependent memory consolidation, REM sleep, NREM sleep, slow-wave sleep, multivariate pattern classification, neural decoding, synaptic plasticity
MeSH terms
Adult, Brain, Electroencephalography, Humans, Learning, Memory, Neuronal Plasticity, Neurons, Sleep, Young Adult, Healthy Volunteers

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.