We are grateful to Zietsch and Sidari (1) for their Letter, which has encouraged us to conduct analyses that address the important issue they raise. Zietsch and Sidari use our data (2) to argue that the Kinsey scale does not measure a unidimensional trait. Lower Kinsey scores represent greater attraction to women, and higher Kinsey scores represent greater attraction to men, implying a negative correlation between attraction to men and attraction to women. Because of the close conceptual link between sexual attraction and arousal, an analogous negative correlation is expected for sexual arousal. But Zietsch and Sidari show that, after statistically adjusting for genital arousal to neutral stimuli, genital arousal to male stimuli is essentially uncorrelated with genital arousal to female stimuli. They conclude that male sexual orientation is better understood as two-dimensional than as one-dimensional, the two dimensions being attraction/arousal to men and attraction/arousal to women. Although their conclusion seems reasonable given their result, we provide further analyses that contradict it.
Associations between raw genital arousal scores to different stimuli, such as those analyzed by Zietsch and Sidari [and by us (2)], are inflated by various factors extraneous to sexual orientation, including penis size, responsivity, and calibration of instruments. Zietsch and Sidari’s approach adjusts for penis size, but not for the other known factors we mentioned, nor additional unknown factors. This is why most researchers using genital arousal methodology to study sexual orientation standardize their data within subjects (3).
Zietsch and Sidari might still reasonably ask, Is there any evidence that sexual arousal to male stimuli is negatively correlated with sexual arousal to female stimuli? Answering this question requires data that do not share similar sources of extraneous variance as genital arousal. Fortunately, we routinely asked participants to rate their subjective arousal to all stimuli. Genital and subjective arousal measures are associated primarily through the intended latent variable of actual sexual arousal. Their association does not reflect the kinds of methodological issues we have mentioned. Thus, we examined how genital arousal to stimuli featuring actors of one sex is associated with subjective arousal to stimuli featuring the other sex.
If sexual arousal to men and sexual arousal to women are uncorrelated, then genital arousal to one sex should not provide any information about subjective arousal to the other sex. For both sexes, however, higher genital arousal to one sex was associated with reduced subjective arousal to the other. The partial correlation between genital arousal to male stimuli and subjective arousal to female stimuli, statistically adjusting for both genital and subjective arousal to neutral stimuli, was −0.36. The partial correlation between genital arousal to female stimuli and subjective arousal to male stimuli, statistically adjusting for the same covariates, was also −0.36. (For both correlations, P < 0.0001.) Almost certainly, both correlations are attenuated due to measurement error. We believe the only plausible explanation of this finding is that sexual arousal to women and sexual arousal to men are at least moderately negatively associated.
Footnotes
The authors declare no competing interest.
References
- 1.Zietsch B. P., Sidari M. J., The Kinsey scale is ill-suited to most sexuality research because it does not measure a single construct. Proc. Natl. Acad. Sci. U.S.A. 117, 27080 (2020).33144520 [Google Scholar]
- 2.Jabbour J.et al., Robust evidence for bisexual orientation among men. Proc. Natl. Acad. Sci. U.S.A. 117, 18369–18377 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Blanchard R., Klassen P., Dickey R., Kuban M. E., Blak T., Sensitivity and specificity of the phallometric test for pedophilia in nonadmitting sex offenders. Psychol. Assess. 13, 118–126 (2001). [DOI] [PubMed] [Google Scholar]