Keynotes | K&Ts | GACs | Talks | Posters | Search

Poster F in Poster Session F: Thursday, August 6, 1:45 – 3:30 pm, Kimmel Center, Shorin & Rosenthal Rooms

Understanding explicit sensorimotor learning as a process of hypothesis testing

Anjuli Niyogi¹, Elizabeth Cisneros², Wei Ding³, Richard Ivry², Jonathan Tsay¹; ¹Carnegie Mellon University, ²University of California, Berkeley, ³Tsinghua University

Presenter: Anjuli Niyogi

Multiple learning processes contribute to successful goal-directed actions under changing physiological states and environmental demands. Amongst these, explicit strategies are essential, rapidly recalibrating our movements to achieve precise, accurate performance. Yet how explicit strategies are discovered during sensorimotor adaptation remain unknown. To answer this question, we developed a behavioral paradigm that isolates strategy discovery in response to a visuomotor rotation. Rather than exhibiting a smooth, gradual learning curve, behavior was characterized by extensive, multimodal exploration that culminated in an “aha” moment, as participants converged on the solution. To account for these findings, we propose a mechanism in which strategy discovery emerges through the generation and testing of discrete action–outcome hypotheses. This “Hypothesis Testing” model captures a broad range of unexplained phenomena in the literature and outperforms alternative computational accounts. Together, these results define a new computational mechanism by which humans discover effective strategies during sensorimotor learning.

Topic Area: Memory, Learning & Knowledge Structures