Consistent hierarchies of single-neuron timescales in mice, macaques and humans

Abstract

The intrinsic timescales of single neurons are thought to be hierarchically organized across the cortex, but whether hierarchical variation in timescales is a general brain organizing principle across mammalian species remains unclear. Here, we took a cross-species approach and estimated neuronal timescales of thousands of single neurons recorded across frontal cortex, amygdala, and hippocampus in mice, monkeys, and humans of both sexes using a task-agnostic method. We identify largely consistent hierarchies of timescales in frontal and limbic regions across species: hippocampus had the shortest timescale whereas anterior cingulate cortex had the longest. Within this scheme, variability across species was found, most notably in amygdala and orbitofrontal cortex. We show that variation in timescales is not simply related to differences in spiking statistics nor the result of cytoarchitectonic features such as cortical granularity. Thus, hierarchically organized timescales are a consistent organizing principle across species and appear to be related to a combination of intrinsic and extrinsic factors.Significance StatementIntrinsic timescales are a measure of the temporal receptive field of a neuron or brain area. Timescales are thought to be hierarchically organized in the brain, but a cross-species evaluation of this has been lacking. Using recordings of the activity of over 17,000 isolated single neurons from cortical and subcortical structures, we report that there are largely similar hierarchies of timescales in mice, macaques, and humans. We also show that the cytoarchitecture of an area is not sufficient to explain cortical variation in timescales. Thus, timescale hierarchies appear to serve as an organizing principle of mammalian brains, likely emerging from the complex interaction of many anatomical and physiological factors.

Files

Additional details