For centuries, researchers have explored the idea that “average” faces are beautiful.
According to the Face Research Lab at the University of Glasgow, the idea dates back to at least the 1800s when Sir Francis Galton was seeking to prove that certain types of people—i.e. criminals or soldiers—had similar facial characteristics.
He never proved that, but did find that when he layered numerous images of faces onto one piece of photographic paper to create one composite “average” face, the result was surprisingly attractive.
Since then, computer-imaging techniques have improved the process. You can try making your own composite faces at faceresearch.org.
Ben Jones, a professor at the Institute of Neuroscience and Psychology who also runs the Face Research Lab, answered some questions about his work for Metro.
Q: What do you think accounts for the continued interest in this topic?
Among researchers, I think the continued interest is because the idea of average faces being attractive ties in really nicely to many of the models of how the brain might represent and process faces. Lots of these models suggest that average faces are coded in ways that make them particularly easy for the brain to process, which might cause people to ‘like’ them more. I think lots of different people are quite drawn to the paradoxical nature of average-ness too; the idea that something average might also be exceptional can really make you think.
Q: When measuring facial attractiveness, have any researchers taken into account the diverse tastes of the subjects who judge the faces? What has been learned about faces that are highly attractive to some, yet off-putting to others? (Interesting research from dating website OK Cupid here)
Early work on attractiveness really focused on trying to understand what people generally thought was a good-looking face and tended to ignore personal taste. More recently, however, there has been a lot of work on the possible reasons why some people are attracted to certain types of faces, while other people are attracted to different types of faces. A lot of this work has focused on women’s preferences for masculine and feminine men. Some women show very strong preferences for masculine men, but other women show very strong preferences for feminine guys. Research suggests that lots of different factors contribute to these differences of opinion, including women’s own attractiveness and age, whether or not they are in a relationship, the types of relationships they enjoy, and maybe even their hormone levels. Trying to understand personal taste is complicated, but extremely interesting.
Q: If attractiveness is not merely average-ness, but the average composite faces are still found attractive, what is it that’s attractive about the composite faces?
Although there is probably a lot more to attractiveness than average-ness, and many studies have shown that some non-average traits can also be extremely attractive, it’s also true that average faces are often more attractive than many of the individual faces they were made from. Additionally, while it was once thought that the appeal of average faces might simply be down to the fact that they tend to be almost perfectly symmetric and have clear complexions, subsequent work suggests those factors don’t fully explain the appeal of average faces. One possibility is that average faces are attractive because they are easy to process (see above). Another possibility is that people with more average faces are healthier. For example, a series of studies by an Australian researcher named Gill Rhodes has presented some really nice evidence for links between average-ness and different measures of health. Of course, these two possibilities aren’t mutually exclusive. Both could be true.
Q: What non-average attractive traits did the hyper-attractive composite face used in Perrett’s study have?
In 1994, Perret et. al took 60 photos of women’s faces and asked volunteers to rate them in attractiveness from one to seven. The researchers then made a composite face from all 60 faces and a composite face from only the 15 most attractive faces which, of the two, was found to be more attractive. Then, the researchers made a “hyper-attractive” face, by digitally exaggerating the difference between the 60-face average and the more attractive 15-face average. New study participants found the digitally enhanced face to be the most attractive, even though it was mathematically the least average.
The hyper-attractive female faces had more feminine characteristics, like larger eyes and a smaller nose and jaw. Lots of subsequent work also suggests that women’s faces with characteristics that are more feminine than average are highly attractive.
Q: How do culture and race factor in? Do people prefer the average in faces of other cultures and races?
There is pretty good evidence that preferences for average-ness show up in many cultures, at least when the averages are made from faces of peoples’ own race. However, there is also some really nice evidence that you need to have some experience with a race in order to show preferences for average faces of that race, perhaps because it takes experience for you to ‘build up’ an average for them.
Q: Over the years of study on average-ness and attractiveness, has the average changed? Has the view of what is considered attractive changed?
It’s tough to say really. Things like health and diet are visible in faces, so big population-level changes in those would probably cause a difference in what was average; if the population gets heavier, for example, the average would probably look heavier, too. As for changes in what is considered attractive, it’s also hard to say. As it stands, we don’t really have enough data to answer the question properly. Anecdotally, you see changes in how attractiveness is represented in art, sculpture and literature, but it’s difficult to say whether these changes simply reflect the personal tastes of various artists, sculptors and writers. We know that experimentally manipulating people’s concerns about things like illness and financial security in the lab can alter their face preferences, so I’d imagine that changes in environmental factors over time could well drive changes in the types of people we consider attractive.