Reevaluating the Role of Persistent Neural Activity in Short-Term Memory

• Published in: Trends in cognitive sciences Vol. 24; no. 3; pp. 242 - 258
• Main Authors: Masse, Nicolas Y., Rosen, Matthew C., Freedman, David J.
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01.03.2020
• Subjects: Brain; hidden states; Humans; Memory, Short-Term; Neuronal Plasticity; oscillations; persistent activity; recurrent neural networks; short-term memory; short-term synaptic plasticity; working memory; recurrent neural networks; working memory; persistent activity; short-term synaptic plasticity; short-term memory; hidden states; oscillations
• ISSN: 1364-6613; 1879-307X
• DOI: 10.1016/j.tics.2019.12.014

A traditional view of short-term working memory (STM) is that task-relevant information is maintained ‘online’ in persistent spiking activity. However, recent experimental and modeling studies have begun to question this long-held belief. In this review, we discuss new evidence demonstrating that information can be ‘silently’ maintained via short-term synaptic plasticity (STSP) without the need for persistent activity. We discuss how the neural mechanisms underlying STM are inextricably linked with the cognitive demands of the task, such that the passive maintenance and the active manipulation of information are subserved differently in the brain. Together, these recent findings point towards a more nuanced view of STM in which multiple substrates work in concert to support our ability to temporarily maintain and manipulate information. It has been commonly believed that information in short-term memory (STM) is maintained in persistent delay-period spiking activity.Recent experiments have begun to question this assumption, as the strength of persistent activity appears greater for tasks that require active manipulation of the memoranda, as opposed to tasks that require only passive maintenance.New experiments have revealed that information in STM can be maintained in neural ‘hidden’ states, such as short-term synaptic plasticity.Machine-learning-based recurrent neural networks have been successfully trained to solve a diversity of working memory tasks and can be leveraged to understand putative neural substrates of STM.