Abstract: Conspicuously displaying one’s altruistic behavior—conspicuous altruism—can promote the spread of prosocial norms and encourage further acts of kindness. However, such behavior often triggers psychological tension due to fears of being perceived as self-serving, a conflict known as the “conspicuous altruism paradox.” While previous studies have primarily examined this dilemma in individual contexts, little attention has been given to the potential role of social settings. We proposed that engaging in altruism within a group may help individuals reinterpret their actions as morally motivated rather than ego-driven. Specifically, we hypothesized that individuals would be more willing to publicly share their altruistic behaviors when these behaviors are performed in a group setting, and that this group effect would be driven by both internal motives (altruistic communication) and external motives (impression management).
To test this hypothesis, we conducted ten studies using a range of methods, including laboratory-based tasks, online surveys, real-world behavioral data, and large language model (LLM) simulations. In total, 1, 938 human participants took part in the studies, and 756 independent responses were generated via LLM simulations. Human participants were recruited from online crowdsourcing platforms and real-world volunteer groups. Studies 1a and 1b employed imagined scenarios in which participants considered engaging in altruistic acts such as time or monetary donations, either alone or with others, and indicated their willingness to share these actions. Study 1c further controlled for display materials to rule out the alternative explanation of group-framed display. Studies 2-5 expanded on these designs using memory recall, open-ended narratives, and real-world volunteer surveys. Complementing these human studies, the AI-based simulations included two parts: Study 6a tested whether LLMs could reproduce the human-patterned group effect of conspicuous altruism; and Study 6b examined whether observing group-based altruism could promote prosocial diffusion among subsequent agents.
Results across all studies consistently supported the proposed group effect of conspicuous altruism. Participants were more inclined to share their altruistic acts after engaging in group-based behavior compared with acting alone. This effect appeared across a variety of altruistic domains, including blood donation, tutoring support, and financial giving, and was also evident in real volunteer settings. The dual-path mechanism was robust: group altruism enhanced both altruistic communication motives—driven by a desire to inspire others—and impression management motives—concerned with social image and recognition. LLM simulations replicated this group effect and further demonstrated its potential for promoting prosocial behavioral diffusion, suggesting that group-framed altruistic acts are more likely to be passed on or imitated by others. A mini meta-analysis of all studies confirmed the robustness of the group effect across diverse contexts and methods.
This research extends the theoretical understanding of conspicuous altruism by identifying group-based prosocial behavior as a key contextual factor that reduces reputational concerns and enhances the likelihood of public sharing. Group participation helps individuals resolve the moral tension of altruistic display by reframing it as socially responsible rather than self-promotional. These findings offer practical implications for the design of charitable campaigns and social media strategies: promoting group-based engagement in altruistic initiatives may significantly increase both participation and the visibility of prosocial behaviors.

Key words: conspicuous altruism, altruistic behavior, prosocial behavior, impression management