Evaluation Strategies Modulate the Distractor Effect in Multi-Attribute Decision Making

Poster Session B - Sunday, March 30, 2025, 8:00 – 10:00 am EDT, Back Bay Ballroom/Republic Ballroom

Kaicheng Yan^1,2 (kaicheng.yan@mail.mcgill.ca), Lesley Fellows^1,2; ¹McGill University, ²Montreal Neurological Institute-Hospital

Humans frequently make irrational choices in the presence of irrelevant distractor options. However, there is ongoing debate regarding whether a highly rewarding distractor facilitates or impairs value-based decision-making. To reconcile conflicting findings, we propose that evaluation strategy plays a critical role in shaping distractor effects, with elemental and configural strategies influencing decision outcomes in distinct ways. Under an elemental strategy, individuals evaluate each attribute of an option separately, whereas under a configural strategy, options are assessed holistically, rendering individual attributes non-informative in isolation. We hypothesize that a highly rewarding distractor impairs choice accuracy under an elemental strategy but has a weaker detrimental effect—or may even enhance accuracy—under a configural strategy. To test this, 32 participants first learned the values of multi-attribute pseudo-objects (fribbles) under each evaluation strategy and then made two-alternative forced choices while an unavailable distractor was present. Preliminary analysis using a generalized linear mixed-effects model revealed a significant negative effect of distractor value on choice accuracy across evaluation strategies. Critically, a significant interaction between distractor value and evaluation strategy emerged, indicating that the negative effect of a highly rewarding distractor on accuracy was pronounced under elemental evaluation but attenuated under configural evaluation. These findings suggest that the effect of distractors on value-based decision-making depends on how option values are constructed. Ongoing analyses will refine these findings by fitting a Drift Diffusion Model to estimate drift rate and decision thresholds, providing a mechanistic account of how distractor value influences option discriminability and cautiousness in decision-making.

Topic Area: THINKING: Decision making