Inverse Rational Control

Xaq Pitkow describes a principles-based approach to explaining cognition. Topics include inverse rational control, strategies that are rational relative to an organism’s beliefs, probabilistic computation and graph-network inference. The conversation also considers attention, constraints on neural codes and studying cognition in ecological settings.

Inverse Rational Control (IRC) is a new theoretical framework to infer subjects' beliefs about the world from their actions. Here we apply IRC for the first time to experimental data and use it to interpret monkey behavior in a spatial navigation task. In this task, monkeys use a joystick to navigate to a transiently visible target (a 'firefly') using optic flow cues. In general, we define beliefs to be summary statistics for posteriors over partially observable latent world states. For our task, these beliefs represent the subject's estimated current position and uncertainty, assumed observat