Technical Program

Paper Detail

Paper: PS-1B.22
Session: Poster Session 1B
Location: H Fläche 1.OG
Session Time: Saturday, September 14, 16:30 - 19:30
Presentation Time:Saturday, September 14, 16:30 - 19:30
Presentation: Poster
Publication: 2019 Conference on Cognitive Computational Neuroscience, 13-16 September 2019, Berlin, Germany
Paper Title: Dynamic Gaze Effects on Cost-Benefit Decisions: from Value Modulation to Additive Influences
Manuscript:  Click here to view manuscript
License: Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 Unported License.
Authors: Andrew Westbrook, Brown University, United States; Lieke Hofmans, Jessica Määttä, Danae Papadopetraki, Ruben van den Bosch, Roshan Cools, Donders Institute for Brain, Cognition and Behaviour, Netherlands; Michael Frank, Brown University, United States
Abstract: Gaze biases choice during value-based decision-making. The attention drift diffusion model describes this bias as a multiplicative interaction whereby gaze amplifies the value of the attended relative to the unattended option. Another account proposes that the direction of gaze effects on choice might be reversed, such that a latent decision to choose an option causes participants to dwell on that option, resulting in additive rather than interactive effects of value and gaze. Here, we tracked dynamic gaze patterns while participants chose between two options, the costs and benefits of which are spatially separated on the screen. Influences of gaze on choice evolved over time: Early gaze at benefits versus costs biases choices toward high-cost / high-benefit options, consistent with the attention drift diffusion model. Conversely, later gaze increasingly reflects the upcoming choice, so that gaze at both high benefits and high costs accompany choice of high-cost / high-benefit options. Formally, early gaze predicts drift rates via a multiplicative interaction with value, while late gaze is additive with value, consistent with a reversal of the direction of influence. Our results help reconcile competing models by applying drift diffusion modeling to the domain of multi-attribute decisions.