PT  - JOURNAL ARTICLE
AU  - Tyler S. Manning
AU  - Benjamin N. Naecker
AU  - Iona R. McLean
AU  - Bas Rokers
AU  - Jonathan W. Pillow
AU  - Emily A. Cooper
TI  - A General Framework for Inferring Bayesian Ideal Observer Models from Psychophysical Data
AID  - 10.1523/ENEURO.0144-22.2022
DP  - 2023 Jan 01
TA  - eneuro
PG  - ENEURO.0144-22.2022
VI  - 10
IP  - 1
4099  - http://www.eneuro.org/content/10/1/ENEURO.0144-22.2022.short
4100  - http://www.eneuro.org/content/10/1/ENEURO.0144-22.2022.full
SO  - eNeuro2023 Jan 01; 10
AB  - A central question in neuroscience is how sensory inputs are transformed into percepts. At this point, it is clear that this process is strongly influenced by prior knowledge of the sensory environment. Bayesian ideal observer models provide a useful link between data and theory that can help researchers evaluate how prior knowledge is represented and integrated with incoming sensory information. However, the statistical prior employed by a Bayesian observer cannot be measured directly, and must instead be inferred from behavioral measurements. Here, we review the general problem of inferring priors from psychophysical data, and the simple solution that follows from assuming a prior that is a Gaussian probability distribution. As our understanding of sensory processing advances, however, there is an increasing need for methods to flexibly recover the shape of Bayesian priors that are not well approximated by elementary functions. To address this issue, we describe a novel approach that applies to arbitrary prior shapes, which we parameterize using mixtures of Gaussian distributions. After incorporating a simple approximation, this method produces an analytical solution for psychophysical quantities that can be numerically optimized to recover the shapes of Bayesian priors. This approach offers advantages in flexibility, while still providing an analytical framework for many scenarios. We provide a MATLAB toolbox implementing key computations described herein.