Abstract
Many contemporary models of time perception are based on the notion that our brain houses an internal “clock”, specialized for tracking duration. Here we show that specialized mechanisms are unnecessary, and that human-like duration judgements can be reconstructed from neural responses during sensory processing. Healthy human participants watched naturalistic, silent videos and rated their duration while fMRI was acquired. We constructed a computational model that predicts video durations from salient events in participants’ visual cortex activation. This model reproduced biases in participants’ subjective reports, whereas control models trained on auditory or somatosensory activity did not. Our data reveal that subjective time is inferred from information arising during the perception of our dynamic sensory environment, providing a computational basis for an end-to-end account of time perception.
Footnotes
