Deciding for others as a neutral party recruits risk-neutral perspective-taking: Model-based behavioral and fMRI experiments

Akitoshi Ogawa,Tatsuya Kameda,Keigo Inukai,Atsushi Ueshima

doi:10.1038/s41598-018-31308-6

Abstract

Risky decision making for others is ubiquitous in our societies. Whereas financial decision making for oneself induces strong concern about the worst outcome (maximin concern) as well as the expected value, behavioral and neural characteristics of decision making for others are less well understood. We conducted behavioral and functional magnetic resonance imaging (fMRI) experiments to examine the neurocognitive underpinnings of risky decisions for an anonymous other, using decisions for self as a benchmark. We show that, although the maximin concern affected both types of decisions equally strongly, decision making for others recruited a more risk-neutral computational mechanism than decision making for self. Specifically, participants exhibited more balanced information search when choosing a risky option for others. Activity of right temporoparietal junction (rTPJ, associated with cognitive perspective taking) was parametrically modulated by options’ expected values in decisions for others, and by the minimum amounts in decisions for self. Furthermore, individual differences in self-reported empathic concern modified these attentional and neural processes. Overall, these results indicate that the typical maximin concern is attenuated in a risk-neutral direction in decisions for others as compared to self. We conjecture that, given others’ diverse preferences, deciding as a neutral party may cognitively recruit such risk-neutrality.

Highlights

Humans often have to make decisions on behalf of others, in situations involving financial and social services
This study investigated multimodal processes underlying risky decision making for others
In line with previous studies[3,4], participants showed a strong maximin concern across the behavioral and functional magnetic resonance imaging (fMRI) experiments, weighting the worst possibility in risky decision making for others as well as for self (Figs 1b and 2b)

Summary

Introduction

Humans often have to make decisions on behalf of others, in situations involving financial and social services. It is well established that decision making for oneself deviates systematically from a simple maximization of expected value, due to the element of risk. People are concerned with how variable (variability concern1,2) or how bad (maximizing the minimum payoff, i.e., maximin concern3,4) choice outcomes can be, generally exhibiting risk-averse preferences. Hsee & Weber[6] showed that people tend to think that others are less risk averse than themselves, whether the choices are between options with positive or negative outcomes. According to such predictions, people may focus more strongly on the expected values of choice options when they decide for others than for self. The asterisks indicate that both were significantly different from zero (***P < 0.001). (c) Temporal changes in the average percentages of L, M, or H views during the decision time used by each participant in each trial, divided into quartiles. (d)

Methods

Results

Conclusion