Specifying the Cognitive Mechanisms of Deliberate Error-based Motor Learning

Adaptive behavior is a defining feature of human intelligence. Whether learning to use a new tool or recovering function after injury, people must detect when their actions no longer achieve desired outcomes and then discover new solutions. This project investigates the cognitive mechanisms that allow humans to adapt their behavior flexibly in changing environments. Rather than viewing adaptation as a slow and automatic process driven solely by trial-and-error correction, the project examines whether people instead test structured hypotheses about environmental change, leading to abrupt “moments of insight” that lead to effective new strategies. Understanding these mechanisms has broad implications for education, rehabilitation, healthy aging, and the development of intelligent technologies capable of flexible adaptation. The project will also advance open science through the public release of data, code, and educational resources, while supporting training opportunities for students in cognitive science and computational neuroscience. This project combines careful behavioral experiments, computational modeling, and unsupervised machine learning to study the cognitive mechanisms underlying deliberate error-based motor learning. Participants will perform visuomotor learning tasks in which sensory feedback is systematically altered, allowing researchers to measure patterns of exploration and strategy formation. The project tests the theory that deliberate error-based motor learning is governed by a structured hypothesis-testing process, characterized by discrete exploration and abrupt “aha” moments, rather than by passive error minimization. The work will generate new theoretical models of human learning and problem solving, provide insight into how flexible behavior changes across the lifespan, and establish scalable experimental tools for studying human intelligence in both controlled laboratory and large-scale online settings. 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. NSF Award ID: 2545300 | Program: 01002627DB NSF RESEARCH & RELATED ACTIVIT | Principal Investigator: Jonathan Tsay | Institution: Carnegie Mellon University, PITTSBURGH, PA | Award Amount: $599,994 View on NSF Award Search: https://www.nsf.gov/awardsearch/show-award/?AWD_ID=2545300 View on Research.gov: https://www.research.gov/awardapi-service/v1/awards/2545300.html