Embodied human intelligence: Characterizing Interactions Between Mental and Sensorimotor Workspaces

Unlike most AI systems, human intelligence is embodied. Cognitive processes such as memory and attention drive goal-directed body movements. This project investigates this interaction, focusing on how the systems involved in producing action connect with the systems responsible for higher order cognition, through cognitive/behavioral experiments linked to the development and testing of novel computational models of embodied human intelligence. In addition, the investigators examine how successfully people adapt when the relationship between perception and action changes. Understanding the ability to learn new actions provides novel insights into how our embodied, adaptive human intelligence operates. This project investigates how cognitive abilities and moment-to-moment mental effort shape our ability to learn new motor skills and builds new computational models of embodied human intelligence. It includes a series of carefully designed behavioral experiments, with pupillometry and computational modeling, to investigate how cognitive resources influence sensorimotor learning. The central hypothesis is that motor learning is not purely motoric, but depends on cognitive systems to maintain spatial information, to develop different kinds of learning strategies, and to regulate effort. Results should characterize how stable cognitive abilities and changing internal states shape individual learning trajectories, with implications for theories of cognitive control and sensorimotor learning. The research may help guide more personalized approaches to support motor skill learning, for example through adaptive human-AI training systems that can incorporate wearable technologies and artificial intelligence. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.