Neuronal timescales are functionally dynamic and shaped by cortical microarchitecture

Abstract

Complex cognitive functions such as working memory and decision-making require information maintenance over seconds to years, from transient sensory stimuli to long-term contextual cues. While theoretical accounts predict the emergence of a corresponding hierarchy of neuronal timescales, direct electrophysiological evidence across the human cortex is lacking. Here, we infer neuronal timescales from invasive intracranial recordings. Timescales increase along the principal sensorimotor-to-association axis across the entire human cortex, and scale with single-unit timescales within macaques. Cortex-wide transcriptomic analysis shows direct alignment between timescales and expression of excitation- and inhibition-related genes, as well as genes specific to voltage-gated transmembrane ion transporters. Finally, neuronal timescales are functionally dynamic: prefrontal cortex timescales expand during working memory maintenance and predict individual performance, while cortex-wide timescales compress with aging. Thus, neuronal timescales follow cytoarchitectonic gradients across the human cortex, and are relevant for cognition in both short- and long-terms, bridging microcircuit physiology with macroscale dynamics and behavior.

Intracortical recordings reveal Vision-to-Action cortical gradients driving human exogenous attention

“…Conversely, the Caudal TPJ-Prefrontal and the Rostral TPJ-Prefrontal clusters were located closer to core regions of the cortical gradient, and their cue and target neural responses at the short cue-target delay were integrated in a single peak of activity. This phenomenon is consistent with a key feature of the core-periphery gradient: a temporal hierarchy of receptive windows, analogous to the spatial hierarchy of receptive fields (26)(27)(28)(29)(30)(31)(32). Going from peripheral regions such as early visual cortex, to core regions such as the intraparietal sulcus, temporal receptive windows become longer, integrating over longer durations, and selectivity for coherent temporal structures increases (26,(28)(29)(30).…”

“…Over longer cue-target delays, which allowed temporal segregation of cue and target, IOR could result from temporal normalization, an operation that leads to decreased response to repeating stimuli and increased response to novel ones (30)(31)(32). Recent evidence (27) shows that neural timescales can change to serve cognitive functions. Specifically, prefrontal cortex timescales expand during working memory maintenance and predict individual performance (27).…”

“…The gradient spans between sensorimotor and transmodal areas, and is suggested to be a primary organizing axis of the human cerebral cortex (17,26). The position of a cortical region along this gradient reflects its microstructural and genetic features, connectivity profile, and functional role (26,27). The Visual cluster is situated at the peripheral end of the cortical gradient; its neural responses to the cue and target were segregated, even at the short cue-target delay.…”

“…Recent evidence (27) shows that neural timescales can change to serve cognitive functions. Specifically, prefrontal cortex timescales expand during working memory maintenance and predict individual performance (27).…”

Topographic gradients of intrinsic dynamics across neocortex

“…These micro-architectural properties – increasingly measured directly from histology or inferred from other measurements, such as microarray gene expression – provide a unique opportunity to relate circuit architecture to temporal dynamics and computation. Indeed, multiple studies have focused on how intrinsic timescales vary in relation to microscale and macroscale attributes ( Murray et al, 2014 ; Mahjoory et al, 2019 ; Shine et al, 2019 ; Gao et al, 2020 ; Ito et al, 2020 ; Raut et al, 2020 ). The primary functional consequence of this hierarchy of timescales is thought to be a hierarchy of temporal receptive windows: time windows in which a newly arriving stimulus will modify processing of previously presented (i.e.…”