Reviews and recommendations are unbiased and products are independently selected. Postmedia may earn an affiliate commission from purchases made through links on this page.
Facial image capture and analysis have become hot topics around the world. In China, cameras are used to spot and shame jaywalkers. A U.S. company called Clearview AI, Inc. sells access to photos that it scrapes, without permission, from social media sites like Facebook. Police forces around the world buy their service, using it to identify looters and other criminals. A 2021 investigation by Canada’s federal and provincial privacy commissioners found it contravened Canada’s privacy law, so it closed shop in this country.
Now, it turns out that one of the most sophisticated facial analysis engines is located inside the male brain. Researchers have discovered that men can accurately detect where women are on their monthly cycles. Guess what? We find them most attractive around the time they are most likely to get pregnant.
Researchers studied 182 heterosexual men aged 18 to 52. They showed them photographs of the same women at various points in their menstrual cycles. The women were instructed to adopt a neutral expression and not wear jewelry or makeup.
When asked to rate the physical attractiveness of the ladies, the men overwhelmingly found those who were in the late follicular phase most attractive. This correlates with the dates when a woman is most likely to conceive. Interestingly, the men could not explain how they did this. In fact, when they were explicitly asked to “select which photograph you think depicts the woman who has a greater chance at successfully getting pregnant?” their answers were no better than chance.
The authors, led by psychologist Lisa L.M. Welling of Oakland University in Rochester, Minn., also evaluated the men on something called the sociosexual scale. This is a measure of how willing a man is to engage in uncommitted sexual activity. To gauge this factor, participants were asked questions like “With how many partners have you had sex in the past 12 months?” and “In everyday life, how often do you have spontaneous fantasies about having sex with someone you just met?”
This factor produced a result that surprised the investigators, who wrote, “Men with lower sociosexuality preferred high fertility faces more than men with higher sociosexuality.” They suggested that “it is possible that it is more important for men who are more interested in long-term mating to select high quality, fertile mates as compared to men who are looking only for short-term mates.”
What else can our faces reveal about us? In 2017, Stanford University professor Michal Kosinski created a furor with a paper that claimed an algorithm could do a better job than a human at figuring out sexual orientation. He scraped (again without explicit permission) images from gay and straight dating sites and claimed 81 per cent accuracy at picking the gay man from a pair of photos, compared to 61 per cent for human beings.
I was critical of this research at the time, noting that even in the sample photos provided in the article, there were clear differences that could skew the results. Homosexual men were more likely to have facial hair, and the heterosexual women used more makeup. Still, Kosinski went on to make claims about other things that could be deduced from our faces, including political orientation.
To be fair, he did acknowledge in an interview that there could be spurious factors that may influence the algorithm. “It might be that Republicans tend to take pictures [of their faces] outdoors,” he told the Financial Times, “and Democrats indoors, and there would be a difference on the brightness level.”
The shakiness of making inferences from facial images has not prevented many countries from trying to use cameras to identify criminals, sometimes even before they commit a crime. This evokes the concept of “precrime,” coined by science fiction author Phillip K Dick and featured in the film Minority Report. An Israeli company sells a facial analysis system that claims it can spot terrorists. A Chinese firm called Hikvision claims its product can predict protests by setting tracking alarms on certain people. Even in North America, researchers have demonstrated racial bias in systems that are trained mainly on white faces.
The discovery that men can unconsciously ferret out the most fertile women raises the chilling possibility of a human/computer system that could combine the power of machine vision and AI with the inexplicable magic of human intuition. For now, you’ll have to be content with looking over your Facebook or Tinder feed and guessing who took their profile photo at the point of peak fertility.
Dr. Tom Keenan is an award-winning journalist, public speaker, professor in the School of Architecture, Planning and Landscape at the University of Calgary, and author of the best-selling book Technocreep: The Surrender of Privacy and the Capitalization of Intimacy.