A Common Mechanism Underlying Food Choice and Social Decisions.

Krajbich, Ian; Hare, Todd; Bartling, Björn; Morishima, Yosuke; Fehr, Ernst · 2015 · DOAJ

DOI: 10.1371/journal.pcbi.1004371

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study investigates whether a single computational mechanism underlies decision-making across distinct domains, specifically comparing choices involving primary rewards (food) with social decisions. The authors address a critical gap in neuroeconomics: while mathematical models often describe choice behavior within specific contexts, it remains unclear if these models generalize across different types of decisions. The research tests the hypothesis that the brain uses a consistent process for making decisions, regardless of whether the stakes involve personal consumption or social interactions. The researchers employed a Sequential Sampling Model (SSM), specifically an attentional Drift-Diffusion Model (aDDM), previously calibrated on data from a food-choice experiment. Instead of fitting new parameters to social decision data, they applied the existing parameters—most notably a fixation-bias parameter ($\theta = 0.3$)—to predict behavior in four independent social-decision tasks. These tasks included two variations of the Dictator game (involving binary choices between selfish and fair monetary allocations) and an Ultimatum game. The model was adapted to account for the lack of eye-tracking data in social tasks by assuming equal attention allocation and treating other players' payoffs as discounted inputs, analogous to unattended items in the food-choice model. The study utilized aggregate data from subjects in the U.S. and Switzerland, conducted in both behavioral and fMRI settings, to test the model's out-of-sample predictive power for both choice probabilities and reaction times (RTs). The results demonstrated that the aDDM, with parameters fixed from the food-choice domain, accurately predicted both choices and RTs in all four social-decision experiments. The model’s predictions showed consistent overlap with empirical data, with mean error magnitudes for choice probabilities ranging from 2.7% to 5.6% and for RTs ranging from 3.1% to 6.1%. Specifically, in the Dictator game, the model correctly predicted that the probability of choosing a selfish option increased with the dictator’s personal gain and decreased with the receiver’s loss. Furthermore, the model captured the "scope insensitivity" effect in a three-player Dictator game, predicting that redundant payoff information for a co-dictator would be ignored, a finding that contradicted traditional social preference models like the Fehr-Schmidt model. These findings provide robust behavioral evidence for a common decision-making mechanism across perceptual, primary reward, and social domains. The study suggests that value coding in the brain is subject to adaptive coding, allowing the same computational process to handle different input ranges and contexts. By demonstrating that a model fit to food choices can predict social decisions without re-fitting, the authors challenge the notion that separate models are required for different economic contexts. This supports the view that human decision-making relies on a unified, dynamical process rather than domain-specific static preferences.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• decision making risk perception