Fig. 1: Overview of neural timescales estimation and experimental design.
From: Neural timescales reflect behavioral demands in freely moving rhesus macaques
A Hierarchical organization of neural timescales at rest (τ) estimated from neuronal spiking data. Neural timescales were estimated in 14 cortical and 3 subcortical areas). Traditionally, neural timescales were estimated in the pre-trial period of various tasks (i.e., chaired electrophysiology) by fitting an exponential decay function to the autocorrelation function (i.e., time-lagged correlation). Each circle represents the population-level \(\tau\) for each cortical area and the stars represent population-level τ for each subcortical area reported in each study. B Depiction of the cage and foraging task. The subjects were allowed to freely explore and interact with reward stations in an open space - i.e., 2.45 × 2.45 × 2.75 m cage with barrels. C Our recording system and recording sites in the striatum, OFC, VLPFC, DLPFC, ACC, FEF, PM, and SMA. D Local field potential (LFP) power spectral densities (PSDs) from example channels in Subject W. E Top: Aperiodic component fit for the example PSDs. We applied spectral parameterization to infer timescales from the PSDs (Donoghue et al., 2020; Gao et al., 2020). The periodic oscillatory peaks were discarded and the ‘’knee frequency” (fk vertical dashed lines) was extracted from the fit of the aperiodic component. Bottom: Neural timescales (τ) were inferred from fk via the embedded equation. OFC orbitofrontal cortex, VLPFC ventrolateral prefrontal cortex, DLPFC dorsolateral prefrontal cortex, ACC anterior cingulate cortex, FEF frontal eye fields, PM premotor cortex, SMA supplementary motor area.
