Bayesian Prior Uncertainty and Surprisal Elicit Distinct Neural Patterns During Sound Localization in Dynamic Environments

Abstract

Estimating the location of a stimulus is a key function in sensory processing, and widely considered to result from the integration of prior information and sensory input according to Bayesian principles. A deviation of sensory input from the prior elicits surprisal, depending on the uncertainty of the prior. While this mechanism is increasingly understood in the visual domain, much less is known about its implementation in audition, especially regarding spatial localization. Here, we combined human EEG with computational modeling to study auditory spatial inference in a noisy, volatile environment and analyzed behavioral and neural patterns associated with prior uncertainty and surprisal. First, our results demonstrate that participants indeed used prior information during periods of stable environmental statistics, but showed evidence of surprisal and discarded prior information following environmental changes. Second, we observed distinct EEG activity patterns associated with prior uncertainty and surprisal in both the time- and time-frequency domain, which are in line with previous studies using visual tasks. Third, these EEG activity patterns were predictive of our participants sound localization error, response uncertainty, and prior bias on a trial-by-trial basis. In summary, our work provides novel behavioral and neural evidence for Bayesian inference during dynamic auditory localization.