Contributed Talk Sessions | Poster Sessions | All Posters | Search Papers

Poster Session C: Friday, August 15, 2:00 – 5:00 pm, de Brug & E‑Hall

Dynamic Multi-level Learning in a Control-Learning Environment

Junwu Chen¹, Leslie Held, Senne Braem¹; ¹Universiteit Gent

Presenter: Junwu Chen

Humans are often confronted with multiple learning contingencies in real-life situations. Previous studies suggested that people tend to learn on different levels of abstraction sequentially rather than in parallel. In a reward learning environment where reward is only contingent on the task sequence (task switching/repetition), we employed a hierarchical reinforcement learning (HRL) model to investigate if individuals dynamically shifted the level at which they choose to learn over time. Our modelling analyses suggested that participants gradually shifted their priority from task-level learning to learning task sequences. Together, our findings are consistent with the sequential multilearning hypothesis.

Topic Area: Memory, Spatial Cognition & Skill Learning

Extended Abstract: Full Text PDF