Operant & Classical Conditioning Lab

This simulator models a virtual agent learning a light-reward association in a 3D T-maze, the classic setup behind both Pavlovian and operant conditioning experiments. Each "Run trial" sends the agent from the start box to a choice point, where a light cue predicts a food reward at the end of one arm. After every trial, associative strength updates via the Rescorla-Wagner learning rule — the same delta-rule mechanism thought to underlie dopamine reward-prediction-error signaling in the brain. Tune the learning rate and reward magnitude, or flip into extinction to watch a learned association fade when reinforcement stops.