A new study from researchers Gregory Stanley, Jun Zhang, and Rick Lewis shows that when we decide whom to trust, our brains continuously update beliefs about others' motivations using a sophisticated Bayesian process, rather than sorting people into simple categories like 'generous' or 'selfish.'
The research
The team, whose work appears on arXiv, developed a Utility Bayesian Model (UBM) that represents beliefs about another person's social preferences as a continuous probability distribution over the parameters of their utility function. In repeated binary dictator games with randomized payoffs, participants predicted the choices of preprogrammed agents and of one another. The UBM captured these predictions better than non-Bayesian alternatives and better than over 15,000 models that treat others as discrete types. This suggests people track motives as graded positions in a continuous space, not as categories.
Because participants alternated between choosing and predicting, the framework separately estimated preferences and beliefs about preferences. Choosers' own utilities, fit with a seven-parameter function selected from 526 candidate forms, showed that most people place reliably positive weight on a stranger's payoff but weight their own roughly nine times more. Interestingly, aversion to being ahead exceeded aversion to being behind, reversing the canonical ordering found in earlier studies. Sensitivity to others' payoffs and to inequalities grows more than proportionally with their size, and antisocial preferences were more common than prior estimates suggested. Predictors' initial beliefs captured the dominance of self-interest but expected aversion to being behind to outweigh aversion to being ahead.
Why it matters
Understanding how we infer others' motives has practical implications for cooperation, negotiation, and everyday social interactions. This model provides a common seven-dimensional space to locate any individual's social preferences and beliefs about others, enabling cross-study comparisons and meta-analyses. For your own cognition, it highlights that your brain is constantly updating a nuanced model of those around you—a process you can become more aware of and potentially improve.
What you can do
Practice active perspective-taking in daily interactions: after someone makes a choice, try to infer the value they placed on yours vs. their own outcomes. Keep a mental note of how your predictions change over time—this exercises your Bayesian social inference system.
Source: arXiv q-bio.NC
Curious about your own brain? Take our free adaptive IQ test or try 306 brain training levels.
