Causal Inference in Bimanual Touch

Project Summary / Abstract Accurately perceiving touch on our hands is essential to daily life. Discerning the source of sensory stimuli within the environment is referred to as Causal Inference (CI), and is vital in sensory processing. While principles of CI are well known, it is less known how the inferred causal structure of sensory stimuli modulates perception. Bayesian CI provides a probabalistic framework using Bayesian statistics to model how the perceived relationship between stimuli conditions stimulus perception. I hypothesize bimanual tactile processing can be understood within a Bayesian CI framework. To evaluate this hypothesis, I have developed a novel tactile paradigm that allows for the manipulation of inferred causal structure of touch across the hands. Using this tactile paradigm, I will characterize how concurrent visual feedback (Aim 1a) and previous visual feedback (Aim 1b) affect bimanual tactile sensitivity to timing differences. Additionally, I predict that Bayesian CI judgements and sensory percepts are represented within the brain. I will use functional Magnetic Resonance Imaging to identify human brain regions associated with CI judgements and evaluate to what extent brain activation reflects Bayesian CI computations (Aim 2). Completion of these aims will allow for better understanding of how bimanual touch is integrated and can shed light on broader principles of sensory integration. Understanding how touch is integrated across the hands can inform novel therapies for sensory- related disorders and advance development of haptics and brain-computer interfaces.