Computational protocol for hierarchical Bayesian modeling of perception and generalization in fear conditioning

Understanding human generalization behavior requires disentangling underlying cognitive and perceptual mechanisms. Here, we present a computational protocol to analyze individual differences in fear generalization by integrating a Bayesian state-space perceptual model with a hierarchical mixture generalization model. We describe steps for applying a state-space model to measure perceptual data and for calculating stimulus distance from these probabilistic representations. We then detail procedures for employing a hierarchical mixture generalization model to distinguish between perceptual and learning-based generalization processes. • Steps for applying a Bayesian state-space model to track perceptual representations • Procedures for computing stimulus similarity from perceptual distribution overlap • Instructions for fitting a hierarchical mixture model to identify generalization pathways • Guidance on comparing perceptual assumptions using Bayes factors in a super-model Publisher’s note: Undertaking any experimental protocol requires adherence to local institutional guidelines for laboratory safety and ethics. Understanding human generalization behavior requires disentangling underlying cognitive and perceptual mechanisms. Here, we present a computational protocol to analyze individual differences in fear generalization by integrating a Bayesian state-space perceptual model with a hierarchical mixture generalization model. We describe steps for applying a state-space model to measure perceptual data and for calculating stimulus distance from these probabilistic representations. We then detail procedures for employing a hierarchical mixture generalization model to distinguish between perceptual and learning-based generalization processes.