Theoretical proposals have previously been put forward regarding the computational basis of interoception. Following on this, we recently reported using an active inference approach to 1) quantitatively simulate interoceptive computation, and 2) fit the model to behavior on a cardiac awareness task. In the present work, we attempted to replicate our previous results in an independent group of healthy participants. We provide evidence confirming our previous finding that healthy individuals adaptively adjust prior expectations and interoceptive sensory precision estimates based on task context. This offers further support for the utility of computational approaches to characterizing the dynamics of interoceptive processing.
bioRxiv Subject Collection: Neuroscience