Computational mechanisms of moral inference

Thesis

Accurately inferring the moral character of others is crucial for avoiding social threats. Putatively “bad” agents command more attention and are identified more quickly and accurately than benign or friendly agents. Such vigilance is adaptive but can also be costly in environments where people sometimes make mistakes, because incorrectly attributing bad character to good people damages existing relationships and discourages forming new ones. Evolutionary models demonstrate that responding to wrongdoers with probabilistic forgiveness can facilitate the evolution of cooperation, but the cognitive mechanisms that enable the implementation of forgiving strategies are unknown. In this dissertation, I explore these mechanisms using novel methods derived from computational science, social cognition, and behavioral economics. Part I of the dissertation demonstrates that moral inference is described by an asymmetric Bayesian updating mechanism, where beliefs about the morality of bad agents are more uncertain (and therefore more amenable to updating) than beliefs about the morality of good agents. The model and data reveal a cognitive mechanism that rapidly discounts prior expectations to permit flexible updating of beliefs about potentially threatening others, a mechanism that could facilitate forgiveness when initial bad impressions turn out to be inaccurate. Part II of the dissertation considers the consequences that ensue when these mechanisms break down. Disruptions in learning and decision-making lie at the heart of many populations characterized by maladaptive social functioning. Consequently, I examine moral inference in two populations associated with interpersonal disturbances: individuals exposed to community violence and patients with Borderline Personality Disorder. The data reveal novel cognitive processes that may explain the emergence of maladaptive behavior related to chronic exposure to violence and Borderline Personality Disorder. Collectively, the results in this dissertation provide insights into the computational mechanisms of moral inference and their role in adaptive and maladaptive social functioning.

Authors: Siegel, J

• Alternative title: Cognitive mechanisms of moral inference and their role in adaptive and maladaptive social functioning
• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English
• Keywords: moral character; person perception; Bayesian inference; computational psychiatry; impression formation
• Subjects: Moral impression formation