Neuronal timescales are functionally dynamic and shaped by cortical microarchitecture

Complex cognitive functions such as working memory and decision-making require the maintenance of information over many timescales, from transient sensory stimuli to long-term contextual cues1. However, while theoretical accounts predict that a corresponding hierarchy of neuronal timescales likely emerges as a result of graded variations in recurrent synaptic excitation2-4, direct evidence in the human cortex is lacking. This limits our ability to study how other cytoarchitectural and cell-intrinsic features shape the temporal patterns of cortical activity5-7, and whether neuronal timescales are dynamic and relevant for human cognition. Here, we use a novel computational approach to infer neuronal timescales from intracranial recordings and construct a continuous gradient across the human cortex. We find that timescales increase along the principal sensorimotor-to-association axis7-9, where higher-order association areas have longer neuronal timescales. These measurements reflect transmembrane current fluctuations and scale with single-unit spiking timescales across the macaque cortex10. Cortex-wide transcriptomic analysis11-13 in humans confirms direct alignment between timescales and expression of excitation- and inhibition-related genes, but further identifies genes specifically related 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.