Circuit mechanisms for the maintenance and manipulation of information in working memory

Nicolas Y. Masse, Guangyu R. Yang, H. Francis Song, Xiao-Jing Wang, David J. Freedman

Research output: Contribution to journalArticle

Abstract

Recently it has been proposed that information in working memory (WM) may not always be stored in persistent neuronal activity but can be maintained in ‘activity-silent’ hidden states, such as synaptic efficacies endowed with short-term synaptic plasticity. To test this idea computationally, we investigated recurrent neural network models trained to perform several WM-dependent tasks, in which WM representation emerges from learning and is not a priori assumed to depend on self-sustained persistent activity. We found that short-term synaptic plasticity can support the short-term maintenance of information, provided that the memory delay period is sufficiently short. However, in tasks that require actively manipulating information, persistent activity naturally emerges from learning, and the amount of persistent activity scales with the degree of manipulation required. These results shed insight into the current debate on WM encoding and suggest that persistent activity can vary markedly between short-term memory tasks with different cognitive demands.

Original languageEnglish (US)
Pages (from-to)1159-1167
Number of pages9
JournalNature Neuroscience
Volume22
Issue number7
DOIs
StatePublished - Jul 1 2019

Fingerprint

Short-Term Memory
Maintenance
Neuronal Plasticity
Learning
Neural Networks (Computer)

ASJC Scopus subject areas

  • Neuroscience(all)

Cite this

Circuit mechanisms for the maintenance and manipulation of information in working memory. / Masse, Nicolas Y.; Yang, Guangyu R.; Song, H. Francis; Wang, Xiao-Jing; Freedman, David J.

In: Nature Neuroscience, Vol. 22, No. 7, 01.07.2019, p. 1159-1167.

Research output: Contribution to journalArticle

Masse, Nicolas Y. ; Yang, Guangyu R. ; Song, H. Francis ; Wang, Xiao-Jing ; Freedman, David J. / Circuit mechanisms for the maintenance and manipulation of information in working memory. In: Nature Neuroscience. 2019 ; Vol. 22, No. 7. pp. 1159-1167.
@article{da8e54d30cd946d789f75a74ef75c4b6,
title = "Circuit mechanisms for the maintenance and manipulation of information in working memory",
abstract = "Recently it has been proposed that information in working memory (WM) may not always be stored in persistent neuronal activity but can be maintained in ‘activity-silent’ hidden states, such as synaptic efficacies endowed with short-term synaptic plasticity. To test this idea computationally, we investigated recurrent neural network models trained to perform several WM-dependent tasks, in which WM representation emerges from learning and is not a priori assumed to depend on self-sustained persistent activity. We found that short-term synaptic plasticity can support the short-term maintenance of information, provided that the memory delay period is sufficiently short. However, in tasks that require actively manipulating information, persistent activity naturally emerges from learning, and the amount of persistent activity scales with the degree of manipulation required. These results shed insight into the current debate on WM encoding and suggest that persistent activity can vary markedly between short-term memory tasks with different cognitive demands.",
author = "Masse, {Nicolas Y.} and Yang, {Guangyu R.} and Song, {H. Francis} and Xiao-Jing Wang and Freedman, {David J.}",
year = "2019",
month = "7",
day = "1",
doi = "10.1038/s41593-019-0414-3",
language = "English (US)",
volume = "22",
pages = "1159--1167",
journal = "Nature Neuroscience",
issn = "1097-6256",
publisher = "Nature Publishing Group",
number = "7",

}

TY - JOUR

T1 - Circuit mechanisms for the maintenance and manipulation of information in working memory

AU - Masse, Nicolas Y.

AU - Yang, Guangyu R.

AU - Song, H. Francis

AU - Wang, Xiao-Jing

AU - Freedman, David J.

PY - 2019/7/1

Y1 - 2019/7/1

N2 - Recently it has been proposed that information in working memory (WM) may not always be stored in persistent neuronal activity but can be maintained in ‘activity-silent’ hidden states, such as synaptic efficacies endowed with short-term synaptic plasticity. To test this idea computationally, we investigated recurrent neural network models trained to perform several WM-dependent tasks, in which WM representation emerges from learning and is not a priori assumed to depend on self-sustained persistent activity. We found that short-term synaptic plasticity can support the short-term maintenance of information, provided that the memory delay period is sufficiently short. However, in tasks that require actively manipulating information, persistent activity naturally emerges from learning, and the amount of persistent activity scales with the degree of manipulation required. These results shed insight into the current debate on WM encoding and suggest that persistent activity can vary markedly between short-term memory tasks with different cognitive demands.

AB - Recently it has been proposed that information in working memory (WM) may not always be stored in persistent neuronal activity but can be maintained in ‘activity-silent’ hidden states, such as synaptic efficacies endowed with short-term synaptic plasticity. To test this idea computationally, we investigated recurrent neural network models trained to perform several WM-dependent tasks, in which WM representation emerges from learning and is not a priori assumed to depend on self-sustained persistent activity. We found that short-term synaptic plasticity can support the short-term maintenance of information, provided that the memory delay period is sufficiently short. However, in tasks that require actively manipulating information, persistent activity naturally emerges from learning, and the amount of persistent activity scales with the degree of manipulation required. These results shed insight into the current debate on WM encoding and suggest that persistent activity can vary markedly between short-term memory tasks with different cognitive demands.

UR - http://www.scopus.com/inward/record.url?scp=85067314319&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067314319&partnerID=8YFLogxK

U2 - 10.1038/s41593-019-0414-3

DO - 10.1038/s41593-019-0414-3

M3 - Article

C2 - 31182866

AN - SCOPUS:85067314319

VL - 22

SP - 1159

EP - 1167

JO - Nature Neuroscience

JF - Nature Neuroscience

SN - 1097-6256

IS - 7

ER -