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

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

Analogous neural representations underlying risky decision making in deep reinforcement learning agents and humans

T. Alexander Price¹, Andrew Liu, Rhiannon L Cowan¹, Tyler S Davis¹, Niloufar Shahdoust, Shervin Rahimpour, Ben Shofty, John D Rolston², Elliot H Smith¹, Alla Borisyuk¹; ¹University of Utah, ²Harvard University

Presenter: T. Alexander Price

We employed deep reinforcement learning to discover behavioral and neural strategies underlying a spectrum of performance on a risky decision- making task. Working backwards, we identified analogous behavior from a large cohort of neurosurgical patients from whom we recorded single neuron activity in decision making circuits. Examining low dimensional factors in neuron population activity uncovered temporal and trial factors differentiating task performance groups, with improved task performance being associated with more nonlinear neural representations of reward prediction.

Topic Area: Reward, Value & Social Decision Making

Extended Abstract: Full Text PDF