Humans use sensory information from the world to take action and complete tasks. While there are many strategies for augmenting human perception, touch-based systems provide a large bandwidth and high-fidelity approach for improving human sensory capacity. However, this feedback often does not perfectly match what people are seeing -- and we have a limited understanding of how people integrate information during such sensory mismatch (e.g., seeing a mug while feeling a cylinder with different mechanical properties). We design new systems to explore haptic feedback that augments human performance in environments with sensory mismatch, running human user studies to test our hypotheses, and building generative models to understand how humans act when given designated sensory information.
Human Performance Augmentation