Generalized Morality Culturally Evolves as an Adaptive Heuristic in Large Social Networks

Abstract

Why do people assume that a generous person should also be honest? Why can a single criminal conviction destroy someone’s moral reputation? And why do we even use words like “moral” and “immoral”? We explore these questions with a new model of how people perceive moral character. According to this model, people can vary in the extent that they perceive moral character as “localized” (varying across many contextually embedded dimensions) vs. “generalized” (varying along a single dimension from morally bad to morally good). This variation might be at least partly the product of cultural evolutionary adaptations to predicting cooperation in different kinds of social networks. As networks grow larger and more complex, perceptions of generalized morality are increasingly valuable for predicting cooperation during partner selection, especially in novel contexts. Our studies show that social network size correlates with perceptions of generalized morality in US and international samples (Study 1), and that East African hunter-gatherers with greater exposure outside their local region perceive morality as more generalized compared to those who have remained in their local region (Study 2). We support the adaptive value of generalized morality in large and unfamiliar social networks with an agent-based model (Study 3), and experimentally show that generalized morality outperforms localized morality when people predict cooperation in contexts where they have incomplete information about previous partner behavior (Study 4). Our final study shows that perceptions of morality have become more generalized over the last 200 years of English-language history, which suggests that it may be co-evolving with rising social complexity and anonymity in the English-speaking world (Study 5). We also present several supplemental studies which extend our findings. We close by discussing the implications of this theory for the cultural evolution of political systems, religion, and taxonomical theories of morality.

Morality in the Anthropocene: The Perversion of Compassion and Punishment in the Online World

“…Online, groups function more to signal belonging to a specific social identity such as political party. The superficiality of these connections to relatively unknown strangers can lead people to have black-and-white judgments of morality with little nuance (Jackson et al, 2023). This can lead to a massive campaign of retribution against a complete stranger.…”

A Standard Framework for Social Cognition: interoperable algorithms for inference and representation

“…As the group grows in size, this becomes less possible. One strategy might be to coarse-grain the characteristics considered (Jackson et al, 2023), and generalize across a large range of contexts.…”

“…As mentioned, in competitive scenarios, individuals have been formulated to hold a unitary representation of a group (Khalvati et al, 2021). How this unfolds in larger scenarios, such as when estimating the generalisation of attributes across members of a larger group with stereotypes (Stewart & Raihani, 2023), or when generalising an attribute of a group member to other qualities of their person (Jackson et al, 2023) is still unclear. There are three hypotheses to consider: 1) computational capacities of groups scale with their size to accommodate discrete member representation (David-Barrett & Dunbar, 2013), 2) increasingly computationally 'cheap' heuristics are employed to enable efficient generalisation of traits and states across a group at the cost of detail (Jackson et al, 2023), 3) there is a hybrid model averaging process that, analogously to hierarchical Bayesian estimation (Huys et al, 2012), allows an internal constraint within the self of individual member representations under some group distribution.…”

“…How this unfolds in larger scenarios, such as when estimating the generalisation of attributes across members of a larger group with stereotypes (Stewart & Raihani, 2023), or when generalising an attribute of a group member to other qualities of their person (Jackson et al, 2023) is still unclear. There are three hypotheses to consider: 1) computational capacities of groups scale with their size to accommodate discrete member representation (David-Barrett & Dunbar, 2013), 2) increasingly computationally 'cheap' heuristics are employed to enable efficient generalisation of traits and states across a group at the cost of detail (Jackson et al, 2023), 3) there is a hybrid model averaging process that, analogously to hierarchical Bayesian estimation (Huys et al, 2012), allows an internal constraint within the self of individual member representations under some group distribution. We suggest running experiments where participants are required to keep in mind the preferences or policies of several different group members, both coherent and incoherent in their decisions, and assessing the degree to which candidate heuristic models may outperform models with higher information fidelity or hierarchical group constraints.…”