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

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

Drift Rate Reflects Value, Response Bias Reflects Habit: A DDM Analysis of the Reward Pairs Task

Viktor Timokhov¹, Hugo Fluhr¹, Philippe Tobler, Stephan Nebe; ¹University of Zurich

Presenter: Viktor Timokhov

Human behavior is guided by both goal-directed and habitual systems. While reinforcement learning (RL) models have captured these influences using choice data, integrating RL with sequential sampling models like the drift-diffusion model (DDM) enables modeling of both choices and response times (RTs). In this study, we tested whether value differences modulate the DDM drift rate and whether prior choice frequency affects response bias simultaneously. Using data of 213 participants in the Reward Pairs Task - an instrumental learning paradigm that independently manipulates stimulus value and choice frequency - we applied hierarchical DDM modeling with collapsing boundaries. Results showed that the best-fitting model captured both value-based and habit-based influences: drift rate scaled with value differences, and response bias reflected differences in choice frequencies. Posterior predictive checks confirmed alignment with observed behavior. These findings support a dual-process view of decision-making, showing that goal-directed and habitual factors influence choice and decision speed via distinct mechanisms.

Topic Area: Reward, Value & Social Decision Making

Extended Abstract: Full Text PDF