Synaptic augmentation in a cortical circuit model reproduces serial dependence in visual working memory

Abstract

Recent work has established that visual working memory is subject to serial dependence: current information in memory blends with that from the recent past as a function of their similarity. This tuned temporal smoothing likely promotes the stability of memory in the face of noise and occlusion. Serial dependence accumulates over several seconds in memory and deteriorates with increased separation between trials. While this phenomenon has been extensively characterized in behavior, its neural mechanism is unknown. In the present study, we investigate the circuit-level origins of serial dependence in a biophysical model of cortex. We explore two distinct kinds of mechanisms: stable persistent activity during the memory delay period and dynamic “activity-silent” synaptic plasticity. We find that networks endowed with both strong reverberation to support persistent activity and dynamic synapses can closely reproduce behavioral serial dependence. Specifically, elevated activity drives synaptic augmentation, which biases activity on the subsequent trial, giving rise to a spatiotemporally tuned shift in the population response. Our hybrid neural model is a theoretical advance beyond abstract mathematical characterizations, offers testable hypotheses for physiological research, and demonstrates the power of biological insights to provide a quantitative explanation of human behavior.

Interplay between persistent activity and activity-silent dynamics in prefrontal cortex during working memory

“…However, dependencies of serial biases with delay and ITI durations [24][25][26] , which demonstrate their relationship with memory maintenance, are largely consistent with activity-silent, and not activity-based mechanisms 13,22,27 . Here, we sought to specify the interaction of activity-based and activity-silent prefrontal cortex (PFC) mechanisms in supporting serial biases while subjects performed a spatial working memory task that engages attractor dynamics in prefrontal cortex 6 .…”

“…4a, black line and Fig. 4b, blue) 22,27 . However, the synapses of neurons that had participated in memory maintenance in the previous delay were facilitated due to STP ( Fig.…”

“…Importantly, the reactivation strength correlated with the strength of serial biases in both monkeys and humans, as predicted by a computational model integrating activity-based and activity-silent mechanisms. Finally, single-pulse TMS applied to human prefrontal cortex prior to trial start enhanced serial biases, demonstrating the causal role of prefrontal reactivations in determining working memory behavior.1 Both attractor dynamics 24,28 and activity-silent 13,22,27 mechanisms have been proposed to carry stimulus-selective information from one trial to the next to effect serial biases.However, dependencies of serial biases with delay and ITI durations [24][25][26] , which demonstrate their relationship with memory maintenance, are largely consistent with activity-silent, and not activity-based mechanisms 13,22,27 . Here, we sought to specify the interaction of activity-based and activity-silent prefrontal cortex (PFC) mechanisms in supporting serial biases while subjects performed a spatial working memory task that engages attractor dynamics in prefrontal cortex 6 .…”

“…1 Both attractor dynamics 24,28 and activity-silent 13,22,27 mechanisms have been proposed to carry stimulus-selective information from one trial to the next to effect serial biases.…”

“…Serial biases in spatial working memory tasks denote small but systematic biases in reporting the location memorized in the current trial slightly attracted to nearby locations memorized in the previous trial [23][24][25][26] . That these apparently non-adaptive serial biases reflect the direct interaction of persistent activity and activity-silent mechanisms is a theoretically appealing hypothesis that still lacks experimental support 13,22,27 .…”