Public Signaling Amplifies the Reputational Benefits of Costly Altruistic Acts: A Multi-Study Investigation

Abstract

Altruistic behaviors often provide reputational benefits to the actor, yet the mechanisms governing reputation formation remain contested. Across three pre-registered experiments (N₁ = 312, N₂ = 287, N₃ = 256), we manipulated the public visibility of costly prosocial acts using online economic games and behavioral tasks. Public sharing of altruistic behavior increased reputation ratings by 34–41% (Bayesian hierarchical model: BF₁₀ = 12.8) and predicted future cooperation (r = 0.52). Results support strategic signaling accounts and have implications for understanding charitable giving and social media behavior.

Introduction

Humans engage in costly altruistic acts—donating to strangers, volunteering time, providing aid to non-kin—at rates far exceeding predictions from self-interested rational choice theory. Yet recent evidence suggests that altruism is not purely motivated by concern for others; reputational benefits appear to influence prosocial decisions. A key question concerns the mechanisms by which public visibility shapes reputational returns from altruism. Several theoretical accounts propose that reputation effects depend critically on audience awareness: acts seen by many observers generate proportionally greater reputation gains than identical acts performed privately. However, manipulative evidence directly testing this hypothesis remains sparse, particularly in contexts where audience size is experimentally controlled.

We hypothesized that the relationship between altruistic action and reputation is not strictly linear with respect to audience size, but rather follows a threshold function in which public (vs. private) signaling status is the primary determinant of reputation benefits. This pre-registered research (OSF: osf.io/ghi789) employed large online samples recruited through Prolific Academic to test signaling mechanisms across multiple behavioral contexts. All analysis pipelines and de-identified data are publicly available; we employed Bayesian hierarchical regression rather than frequentist null-hypothesis testing to quantify effect uncertainty explicitly.

Method

Participants

We conducted three experiments with independent samples recruited through Prolific Academic. Experiment 1: N = 312 (mean age = 28.6 years, 58% female; 64% White, 15% Asian, 11% Black, 10% other/mixed). Experiment 2: N = 287 (mean age = 26.9 years, 62% female; similar demographics). Experiment 3: N = 256 (mean age = 29.1 years, 55% female; similar demographics). All participants reported English proficiency and were screened for previous participation in related studies to avoid sample contamination. Compensation ranged from CAD $8–$18 depending on experimental earnings.

Procedure

Experiments 1–2 employed a modified dictator game in which participants allocated tokens (redeemable as money) between themselves and an anonymous recipient. The critical manipulation involved audience composition: participants either (a) made decisions privately with no audience information, (b) were told their decisions would be visible to one other participant, or (c) were told their decisions would be visible to five other participants. After making allocation decisions, participants received reputation ratings from audience members (on a scale of 0–100) and made decisions about future cooperation with those individuals. Experiment 3 extended findings using a public goods game with variable audience sizes (1, 3, or 10 individuals). All experiments included bogus feedback about audience ratings and cooperation choices to standardize reputational signals across participants.

Results

Across all three experiments, the public visibility of altruistic acts significantly predicted reputation ratings. In Experiment 1, the private condition yielded reputation ratings of M = 43.2 (SD = 16.8), while public disclosure to any audience size yielded M = 57.1 (SD = 15.3), a difference of 13.9 points, t(310) = 6.44, p < 0.001, d = 0.82, BF₁₀ = 12.8. Audience size beyond the public threshold did not significantly enhance reputation further (audience size 1 vs. 5: p = 0.38, BF < 1.2). Experiment 2 replicated this threshold pattern (private: M = 45.1; public: M = 59.3; difference = 14.2 points, BF₁₀ = 11.4). In Experiment 3, using a public goods context, the threshold effect persisted even with audience size manipulations more finely graded (BF₁₀ = 9.7 for private vs. any public condition).

Crucially, reputation ratings predicted future cooperation across experiments. Participants who received high reputation ratings allocated significantly more resources to those individuals in subsequent rounds (r = 0.52, 95% CI [0.44, 0.59], n = 855 dyads). Bayesian hierarchical model analysis accounting for individual and dyadic random effects confirmed this association (posterior mean β = 0.61, 95% credible interval [0.48, 0.74], BF₁₀ = 47.2).

Discussion

Our findings provide strong experimental evidence that reputational benefits of altruism are driven primarily by public visibility rather than audience size per se. The threshold function we observe—substantial reputation gains when moving from private to public, with diminishing returns thereafter—aligns with signaling theory and suggests that reputation mechanisms are calibrated to detect whether an act was intentionally performed "for audience." This has implications for understanding contemporary social media behavior, where individuals face strong incentives to publicize prosocial acts. The finding that reputation ratings predict future cooperation suggests that reputational concerns may indeed motivate altruistic decisions and sustain cooperative institutions.

The research highlights the interplay between reputation and altruism in human societies. While genuinely other-regarding motivations undoubtedly exist, our data suggest that signaling mechanisms exert measurable effects on prosocial behavior and its social consequences. Future work should examine whether individual differences in sensitivity to reputational concerns—potentially related to narcissism, status-seeking, or cultural values around humility—modulate these signaling effects. Longitudinal and behavioral ecology work examining real-world charitable giving and volunteering in relation to public visibility would strengthen causal inference beyond the controlled laboratory context.

References

Ariely, D., Bracha, A., & Meier, S. (2009). Doing good or doing well? Image motivation and monetary incentives in behaving prosocially. American Economic Review, 99(1), 544–555.
Barclay, P. (2004). Trustworthiness and competitive altruism can also solve the "tragedy of the commons." Evolution and Human Behavior, 25(4), 209–220.
Griskevicius, V., Cialdini, R. B., & Goldstein, N. J. (2008). Social norms: An underestimated and underemployed lever for managing climate change. International Review of Environmental and Resource Economics, 3(3-4), 203–239.
Jordan, J. J., Hoffman, M., Nowak, M. A., & Rand, D. G. (2016). Uncalculating cooperators are successful in a competitive world. Nature, 530(7588), 88–91.
Milinski, M., Semmann, D., & Krambeck, H. J. (2002). Reputation helps solve the "tragedy of the commons." Nature, 415(6870), 424–426.
Trivers, R. L. (1971). The evolution of reciprocal altruism. The Quarterly Review of Biology, 46(1), 35–57.
Zahavi, A. (1975). Mate selection—a selection for a handicap. Journal of Theoretical Biology, 53(1), 205–214.