Deep neural networks are more accurate than humans at detecting sexual orientation from facial images - PubMed (original) (raw)
. 2018 Feb;114(2):246-257.
doi: 10.1037/pspa0000098.
Affiliations
- PMID: 29389215
- DOI: 10.1037/pspa0000098
Deep neural networks are more accurate than humans at detecting sexual orientation from facial images
Yilun Wang et al. J Pers Soc Psychol. 2018 Feb.
Abstract
We show that faces contain much more information about sexual orientation than can be perceived or interpreted by the human brain. We used deep neural networks to extract features from 35,326 facial images. These features were entered into a logistic regression aimed at classifying sexual orientation. Given a single facial image, a classifier could correctly distinguish between gay and heterosexual men in 81% of cases, and in 71% of cases for women. Human judges achieved much lower accuracy: 61% for men and 54% for women. The accuracy of the algorithm increased to 91% and 83%, respectively, given five facial images per person. Facial features employed by the classifier included both fixed (e.g., nose shape) and transient facial features (e.g., grooming style). Consistent with the prenatal hormone theory of sexual orientation, gay men and women tended to have gender-atypical facial morphology, expression, and grooming styles. Prediction models aimed at gender alone allowed for detecting gay males with 57% accuracy and gay females with 58% accuracy. Those findings advance our understanding of the origins of sexual orientation and the limits of human perception. Additionally, given that companies and governments are increasingly using computer vision algorithms to detect people's intimate traits, our findings expose a threat to the privacy and safety of gay men and women. (PsycINFO Database Record
(c) 2018 APA, all rights reserved).
Comment in
- Cambridge Analytica controversy must spur researchers to update data ethics.
[No authors listed] [No authors listed] Nature. 2018 Mar 29;555(7698):559-560. doi: 10.1038/d41586-018-03856-4. Nature. 2018. PMID: 29595795 No abstract available.
Similar articles
- Presentation in self-posted facial images can expose sexual orientation: Implications for research and privacy.
Wang D. Wang D. J Pers Soc Psychol. 2022 May;122(5):806-824. doi: 10.1037/pspa0000294. J Pers Soc Psychol. 2022. PMID: 35404640 - Facial Structure Predicts Sexual Orientation in Both Men and Women.
Skorska MN, Geniole SN, Vrysen BM, McCormick CM, Bogaert AF. Skorska MN, et al. Arch Sex Behav. 2015 Jul;44(5):1377-94. doi: 10.1007/s10508-014-0454-4. Epub 2014 Dec 31. Arch Sex Behav. 2015. PMID: 25550146 - Shape differences between the faces of homosexual and heterosexual men.
Valentova JV, Kleisner K, Havlíček J, Neustupa J. Valentova JV, et al. Arch Sex Behav. 2014 Feb;43(2):353-61. doi: 10.1007/s10508-013-0194-x. Epub 2013 Oct 17. Arch Sex Behav. 2014. PMID: 24132775 - Discrimination versus specialization: a survey of economic studies on sexual orientation, gender and earnings in the United States.
Schmitt ED. Schmitt ED. J Lesbian Stud. 2008;12(1):17-30. doi: 10.1300/10894160802174250. J Lesbian Stud. 2008. PMID: 19042290 Review. - Real-Time Gender Recognition for Juvenile and Adult Faces.
Gupta SK, Yesuf SH, Nain N. Gupta SK, et al. Comput Intell Neurosci. 2022 Mar 17;2022:1503188. doi: 10.1155/2022/1503188. eCollection 2022. Comput Intell Neurosci. 2022. PMID: 35341170 Free PMC article. Review.
Cited by
- Smile variation leaks personality and increases the accuracy of interpersonal judgments.
Witkower Z, Tian L, Tracy J, Rule NO. Witkower Z, et al. PNAS Nexus. 2024 Aug 22;3(9):pgae343. doi: 10.1093/pnasnexus/pgae343. eCollection 2024 Sep. PNAS Nexus. 2024. PMID: 39246668 Free PMC article. - The effect of sexual orientation on voice acoustic properties.
Holmes L, Rieger G, Paulmann S. Holmes L, et al. Front Psychol. 2024 Aug 7;15:1412372. doi: 10.3389/fpsyg.2024.1412372. eCollection 2024. Front Psychol. 2024. PMID: 39171236 Free PMC article. - Representativeness and face-ism: Gender bias in image search.
Ulloa R, Richter AC, Makhortykh M, Urman A, Kacperski CS. Ulloa R, et al. New Media Soc. 2024 Jun;26(6):3541-3567. doi: 10.1177/14614448221100699. Epub 2022 Jun 19. New Media Soc. 2024. PMID: 38774557 Free PMC article. - Large language models know how the personality of public figures is perceived by the general public.
Cao X, Kosinski M. Cao X, et al. Sci Rep. 2024 Mar 20;14(1):6735. doi: 10.1038/s41598-024-57271-z. Sci Rep. 2024. PMID: 38509191 Free PMC article. - Visual Data and the Law.
Brayne S, Levy K, Newell BC. Brayne S, et al. Law Soc Inq. 2018 Fall;43(4):1149-1163. doi: 10.1111/lsi.12373. Epub 2018 Dec 27. Law Soc Inq. 2018. PMID: 38343725 Free PMC article.
MeSH terms
LinkOut - more resources
Full Text Sources
Other Literature Sources