The current investigation sought to examine whether people were more willing to endorse interventions when IH was borne by men than women. Our first two studies supported this premise. Importantly, however, our results showed that this asymmetry was driven primarily by women, but not men, being more likely to accept IH to men than to women across a variety of contexts (i.e., supporting Hypothesis 2). Study 3 tested a boundary condition to this gender bias in harm tolerance: stereotypically female caregiving contexts. When instrumental harm benefitted vulnerable individuals (e.g., infants, young children, sick, or the elderly), both men and women exhibited a bias in their willingness to accept IH to men versus women (i.e., supporting Hypothesis 1; not supporting Hypothesis 3). That is, contrary to what might be expected by historical gender roles (Eagly & Wood, 1999), people believed men ought to bear greater costs, even in traditionally female sacrificial domains.
Theoretical and Practical Implications
Our findings offer four contributions. First, we extended the literature on gender and harm endorsement, which has primarily emphasized high-conflict sacrificial dilemmas involving questions of life or death (e.g., FeldmanHall et al., 2016; Skulmowski et al., 2014). The current findings revealed this gender bias persists in highly consequential, yet understudied domains: assessments of beneficial interventions carrying negative externalities across a variety of contexts: medical, psychological, educational, sexual, and caregiving. Second, we demonstrated that when evaluating interventions, female participants were more likely than male participants to accept IH borne by men than women. This pattern lends further support to the well-documented finding that women have a stronger in-group bias than men (e.g., Glick et al., 2004; Rudman & Goodwin, 2004) and are more likely to perceive one another as victims than perpetrators (Reynolds et al., 2020). This disparity suggests women may prioritize one another’s welfare over men’s in the construction or approval of social, educational, medical, and occupational interventions. If so, female policymakers might be especially wary of advancing policies or initiatives risking harm to other women, but less so when they risk harming men.
Third, we tested a boundary condition to this gender bias by investigating contexts previously unstudied in sacrificial dilemmas: stereotypically female caregiving roles. Although consideration of gender stereotypes and role congruence (Eagly & Wood, 1999) might predict a greater tolerance for female sacrifice in such contexts, men and women alike were more tolerant of IH incurred by men (versus women). These patterns suggest that although women traditionally fill and sacrifice in these roles, people may not necessarily endorse that ought to be the case. Rather, our results align with emerging evidence documenting diminished concern for men’s suffering due to a greater tendency to stereotype men as perpetrators rather than victims (Reynolds et al., 2020).
Fourth, our findings identified individual-level factors that contribute to asymmetries in harm tolerance. Namely, Studies 2 and 3 revealed that individuals more strongly endorsing egalitarian, feminist, or liberal ideologies exhibited greater disparities in their acceptance of instrumental harm, such that they more readily tolerated instrumental harm borne by men. These patterns suggest those most concerned about rectify- ing historical injustices might most ardently oppose explora- tory interventions potentially providing long-term benefits to women.
Limitations, Emerging Questions, and Future Directions
Although the current investigation has its strengths (e.g., consistent results across varied contexts, within and between-person designs, diverse beneficiaries, pre-registrations), it is not without limitations. First, future investigations might profit, for example, from examining contexts that explicitly signal one’s willingness to sacrifice on behalf of others (e.g., voluntary military service or blood donation) to determine the generalizability of these patterns. Second, our conclusions are limited by our reliance on American MTurk and CloudResearch users. Thus, our results might not generalize to other contexts and cultures. Indeed, changes in stereotypes over time (Charlesworth & Banaji, 2022), and cultural differences in norms surrounding masculinity and femininity might shift beliefs about the value of IH incurred by men versus women (see Glick et al., 2004 for a cross-cultural comparison of attitudes toward men and women). Examining whether the reluctance to expose women to instrumental harm emerges across cultures remains an open avenue for future work. Moreover, our data were collected during the earlier days of COVID-19, which could have influenced the composition or motivations of our samples (Arechar & Rand, 2021). Thus, replication is warranted before strong conclusions can be inferred.
Fourth, although the results of Studies 1 and 2 consistently revealed women’s gender bias in instrumental harm acceptance, their methods could not disentangle whether the bias more strongly emerged from an aversion toward harming women or a desire to benefit women. That is, because both studies pit harm to one sex against the benefit to the other, it is unclear which more strongly contributed to these findings. That Study 3’s female participants (along with male) more readily tolerated men’s (versus women’s) suffering in contexts benefitting vulnerable individuals (rather than women) suggests the possibility Studies 1 and 2’s results reflected women’s greater aversion to harming fellow women, rather than a motivation to benefit them per se. Nonetheless, future research might examine interventions whereby only one sex is benefitted or harmed to adjudicate the relative contribution of these two factors.
Altogether, our findings point to potentially consequential implications for laypeople’s perceptions of exploratory interventions and programs. The asymmetry we documented may place disparate pressures on researchers and policymakers to intervene experimentally on men’s versus women’s afflictions in ways that minimize instrumental harm to women. The biases uncovered here suggest the possibility that women were excluded historically from exploratory research due to an aversion toward inflicting instrumental harm onto women, such as in medicine (Holdcroft, 2007). This ultimately proved costly to women, as men’s overrepresentation in medical research yielded treatments more effective among men than women (Holdcroft, 2007). Thus, although such an aversion may have benefitted women in the short term because women were spared incidental harm imposed by risky experiments, in the long run, experimentation on men unearthed medical and safety advancements better suited for male bodies. Experimental examinations and interventions carry both costs and benefits. If, as our results suggest, people are less willing to accept instrumental harm befalling women, women might lose out on the long-term benefits of such experimental endeavors.
Throughout history, countless male lives have been sacrificed on the battlefield, ostensibly to promote the greater good (Baumeister, 2010). Our findings suggest that these sentiments persist beyond the field of combat. For many people, accepting instrumental harm to men is perceived as worth the cost to advance other social aims. We invite researchers to further investigate how individuals appraise the value of suffering and whether those appraisals differ across target characteristics. A deeper understanding of the biases embedded in such calculations may minimize the unforeseen and unintended consequences of those preferences, thereby reducing harm to men and women alike.
Posts
