The male choice hypothesis begins with an observation about the Darwinian puzzle of same-sex sexual behavior. Same-sex attraction and behavior appear in virtually every human culture, yet they seem to offer no obvious direct reproductive or survival advantage, particularly in people who identify as heterosexual. Apostolou’s proposed solution was that heterosexual men evolved a preference for female partners who exhibit some degree of same-sex attraction. According to the theory, if a woman engages in same-sex contact instead of opposite-sex infidelity, her partner’s paternity is protected because no conception can result. In addition, a female partner attracted to women might facilitate the man’s access to other women—a “participation effect” that would boost his reproductive output without requiring a new committed relationship. The hypothesis also proposed that women with same-sex attractions would enjoy better relationship quality because men particularly value the trait, granting such women extra resources and leniency. Underlying all of this was a genetic model in which mutations create alleles that shift individuals along a spectrum from exclusive heterosexuality toward exclusive homosexuality, with the alleles maintained in the population partly because ancestral parental control forced same-sex attracted women to marry men and reproduce.

Wilczkowski’s critique identifies multiple places where this framework conflicts with core principles of evolutionary psychology and sexual selection. The “participation effect” requires a third-party woman to consent to sex with the male partner despite the risks of unplanned pregnancy and lack of male investment—something that would undermine the very logic of why women supposedly seek resources through mating. If the woman’s same-sex attraction evolved to help her acquire more resources, it makes little sense that it would also cause her to share her partner’s support with another woman. Furthermore, the hypothesis assumes men can gain fitness benefits by mating widely without investment, but in a species with extremely dependent offspring who typically require biparental care, such a strategy would reduce offspring survival and cancel out the proposed advantage. And if a third-party woman conceived and then sought resources from another man, the supposed cuckoldry protection would simply be exported to someone else, creating a cuckoldry liability elsewhere rather than eliminating it.

The critique also highlights contradictions with well-established empirical findings on female sexuality. The male choice hypothesis claims that only “mostly heterosexual” women possess the capacity for same-sex activity, yet decades of research document sexual fluidity—the capacity for attractions and behavior to shift with context—in women across the entire spectrum of identity, including those who identify as exclusively heterosexual. Large-scale surveys show many exclusively heterosexual women have considered or engaged in same-sex contact. The hypothesis likewise dismisses the robust phenomenon of gender-nonspecific physiological arousal in women, documented across studies of genital response, pupil dilation, neural activity, and viewing time, in which heterosexual women show measurable arousal to stimuli featuring both sexes. Apostolou has argued that such women are simply untruthful about their responses, a stance Wilczkowski calls perplexing given the overwhelming methodologically diverse evidence. The hypothesis further asserts a rigid genetic trade-off in which same-sex attraction necessarily reduces interest in men, yet bisexual women report more male partners than heterosexual women and have similar reproductive output, and homosexual women freely engage in sex with men.

The genetic assumptions fare no better against modern data. The landmark genome-wide association study of nearly 478,000 people published in Science in 2019 identified only five genetic markers associated with same-sex sexual behavior, together explaining less than one percent of the variation in the trait; overall genetic influences accounted for an estimated 8 to 25 percent, with the remainder attributed to environmental factors. Crucially, the genetics of ever engaging in same-sex behavior differed from the genetics of the proportion of same-sex partners, and no genetic connection was found between people with exclusively same-sex partners and those with only mild bisexuality—undermining the smooth, single-dimensional genetic spectrum the male choice hypothesis requires. The critique also challenges the hypothesis’s reliance on strong ancestral parental control over daughters’ sexuality, noting instead that female promiscuity has deep evolutionary roots reflected in male anatomy: relatively large testes suggesting sperm competition, penis morphology potentially adapted to displace rivals’ semen, and male sexual jealousy and mate guarding evolved precisely to counter female infidelity. Overt suppression of female sexuality, by this account, is more likely an evolutionarily recent phenomenon than the ancient selective force the hypothesis needs.

Perhaps the most consequential part of the new paper is its methodological reanalysis. In the studies testing the male choice hypothesis, participants rated agreement or likelihood on five-point Likert-type scales whose endpoints were labeled (for example, “strongly disagree” to “strongly agree”) but whose intermediate points—including the midpoint at 3—were left unlabeled. According to standard survey methodology, an unlabeled midpoint typically represents neutrality or uncertainty. Yet in the original publications, responses of 3, 4, and 5 were all classified as “affirmative,” effectively treating every neutral response as moderate agreement. Because Likert distributions often pile up at the midpoint, this coding choice substantially inflates apparent endorsement of key items—statements about women finding same-sex contact exciting, considering it, or being willing to do it, and men tolerating or being excited by a partner’s same-sex infidelity.

Wilczkowski reanalyzed all relevant published percentage distributions, excluding the unlabeled midpoint as analytically indeterminate and coding only responses of 4 and 5 as affirmative. Statistical comparisons used two-sample proportion t-tests, with effect sizes expressed as Cohen’s h and the Benjamini–Hochberg procedure applied to control the false discovery rate across multiple comparisons. The corrected results were dramatic. Where the original data suggested, for example, that 31.3 percent of women in long-term relationships would find partner-sanctioned same-sex contact sexually exciting, the unskewed figure was 14.0 percent, with 64.7 percent explicitly not finding it exciting. Willingness to actually engage in such contact dropped from 16.7 percent to 5.8 percent. Similar reversals appeared across short-term relationship scenarios, across scenarios involving the partner’s participation, and among the subgroup of women reporting same-sex attractions. Negative reactions—anger, sadness, feeling uncared for, considering a breakup—remained common even after correction, particularly in short-term contexts.

The reanalysis of men’s responses was equally damaging to the hypothesis. Originally reported figures suggested that around three-quarters of men would consider a partner’s same-sex contact to be cheating and that roughly half to sixty percent would find it sexually exciting. After correction, sexual excitement fell to 25.4 percent in long-term contexts and 32.6 percent in short-term contexts—majorities explicitly did not find such scenarios exciting—and substantial numbers of men still considered the behavior infidelity worth breaking up over. When women were asked whether they would cheat with a woman or a man, the corrected data showed overwhelming preference for cheating with a man in every condition, directly contradicting the cuckoldry-protection mechanism. Notably, propensity to cheat with a man relative to a woman was actually greater in short-term than long-term contexts, the opposite of what the hypothesis would predict if same-sex contact served as a substitute for risky heterosexual infidelity.

Taken together, Wilczkowski argues, none of the three proposed mechanisms survives scrutiny. The cuckoldry-protection account fails because very few women express interest in same-sex contact and because men do not preferentially select such partners the way they prefer chastity and fidelity cues. The participation effect fails because women show little inclination toward the required behavior and most men do not desire it, even in short-term contexts where the hypothesis predicts maximal enthusiasm. The relationship-quality account fails because women with same-sex attractions are no more open to partner-suggested same-sex encounters than other women. The author notes that the conclusions are robust to alternative treatments of the midpoint: even if neutral responses were included, the overall pattern would be unlikely to change, though a definitive test would require newly collected raw datasets, since the reanalysis necessarily relied on published percentage tables.

The paper does not leave the Darwinian puzzle unanswered. Wilczkowski points to two alternative evolutionary accounts that better fit the evidence: the alloparenting hypothesis, which frames female sexual fluidity as a conditional strategy for recruiting allomothering help when paternal investment was unreliable, and the polygyny hypothesis, which proposes that fluidity evolved to reduce conflict among co-wives and facilitate cooperative child-rearing in polygynous ancestral households. Both accounts explain sexual fluidity and women’s broader erotic plasticity, and the polygyny hypothesis additionally assigns functional relevance to men’s widespread cross-cultural arousal to female–female sexuality, treating it as a mechanism promoting co-wife harmony rather than an adaptation for cuckoldry protection. Whether either of these frameworks ultimately withstands the same rigorous scrutiny that has now undone the male choice hypothesis remains an open empirical question—but the burden of proof for the male-benefit account, this analysis makes clear, has become considerably heavier.