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

Poster Session B: Wednesday, August 13, 1:00 – 4:00 pm, de Brug & E‑Hall

Cross-Context Value Dynamics: The Impact of Contextually Irrelevant Values on Choice Behavior

Neele K. Elbersgerd¹, Nir Moneta¹, Nicolas W. Schuck¹; ¹Universität Hamburg

Presenter: Neele K. Elbersgerd

In everyday decisions, the same choice option can yield different rewards depending on the context. Two coats will, for instance, have different utility depending on whether the goal is to be shielded from rain or to dress for a dinner. We reanalyzed behavioral data from a previous study, where participants switched between contexts with different object features predicting rewards: color in one context and motion in another (Moneta et al., 2023). Participants were explicitly cued in each trial which features to focus on, and outcome depended only on the cued context. Our analysis focused on how competing contexts influence choice and learning trial by trial. We identified two potential learning signals from irrelevant features, carrying information about the value expectation for (1) the chosen object’s irrelevant feature (Object Prediction Error) and for (2) the best feature in the irrelevant context (Context Prediction Error). Combining reinforcement learning and general linear models, we found evidence for both learning signals influencing participants’ behavior, i.e. that value updates were also guided by the difference between context-irrelevant values and outcomes. The strongest support was found for Object Prediction Error updates. Through examining how outcome-irrelevant information influences subjective value assignment, we deepen our understanding of how competing goals are processed and shape future choice behavior.

Topic Area: Reward, Value & Social Decision Making

Extended Abstract: Full Text PDF