The Face Facts team is a multidisciplinary group that combines Psychology with Biology, Computer Science and Anthropology.
Participate in demonstrations of our experimental studies and learn more about how we use computer-graphics to understand how humans perceive faces.
Click on the face of the baby who looks cuter
Biological theories of sexual selection suggest that our perceptions of attractiveness function to choose healthy mates. Body weight is one potential cue of health. Here, we've manipulated shape cues to body weight to see if lower body weight reliably changes your perceptions of health.
Sexual dimorphism contributes to our perceptions of aggressiveness and dominance. We can manipulate sexual dimorphism by calculating the shape differences between average male and female faces. Then we can manipulate the shape of an individual face using those differences to masculinise or feminise it.
Baby baby baby baby.
Different expressions take different amounts of time to see.
In this experiment, you will be asked to find specific expressions. On each trial, you will see 4 faces. After a few seconds, they will start to move and you need to tap on the target expression as fast as you can.
Our brains are specially equipped to see facial expressions of fear and disgust quickly. Being able to see these facial expressions quickly helps us avoid different types of danger in our environment, like fast moving poisonous animals (e.g., snakes), which cause people to make a fear face, or slower moving germs (e.g., dog poo), which cause people to make disgust faces.
Your fastest times
Who is in this image?
What is the facial expression of this image?
What is the ethnic background of the person in this image? (choose all that apply)
What is the age of the person in this image?
Tags: put anything here, like your first name, job, your school, etc. This will be used to create averages of categories like teachers, rugby fans, or Glaswegians. Separate tags with commas.
Login to see saved images or save the images so you can see them later or from another device.
You are logged in as . Logout to change accounts.
Click on the questions below to learn more about the research we do using computer-graphic faces.
We use computer graphic methods like the ones on the Transform Faces page to manipulate faces along a dimension, like weight or masculinity. Examining how those manipulations shape judgments like attractiveness or health can reveal what dimensions influence particular judgments. We can then compare those effects between different groups of people, like men and women or people from different cultures, to investigate how different groups of people respond to those dimensions.
We can make composite faces with the average shape, colour and texture information for faces that are perceived to be particularly high on a trait or physical characteristic and faces that are perceived to be particularly low on that trait or characteristic. For example, we can create composites representing the average facial appearance of people with particularly good or poor medical health. By asking people to say which of the two faces looks healthiest, more attractive or more trustworthy, we can establish whether the faces of people with good medical health look healthier, more attractive or more trustworthy than those of people with poor medical health.
Many different types of facial cues influence social judgments. For example, increasing the masculinity of face shapes makes people look more aggressive, and giving faces a golden glow shows that facial colour information influences health judgments. Dynamic information, such as eye movements and facial expressions, can also be very important for social judgments of faces. People prefer faces that are making eye contact with them, particularly when they are smiling.
Top-down analyses of faces are ones where a researcher takes an idea from scientific theories, uses that idea to generate a hypothesis about how a particular facial cue might influence social judgments, and then designs an experiment to test that hypothesis. Work linking masculine facial traits to perceptions of aggressiveness is a good example of the utility of that approach; researchers took ideas from work on non-human animals suggesting males in many species displaying more masculine characteristics had higher social rank and designed experiments to test for similar effects in human social judgments.
Bottom-up analyses are different because they are less reliant on theories to generate the initial ideas. Bottom-up analyses construct models of how people respond to randomly selected combinations of traits and use those models to reveal the different combinations of traits that underpin different social judgments. Research demonstrating cultural differences in the information people use to judge emotions is a good example of this approach.
Both top-down and bottom-up analyses are now commonly used in face research, helping researchers to test key predictions from existing theories and generate new theories for the next generation of scientists.
Yes. If people are shown faces manipulated to possess a particular trait, they generally begin to think that that trait is more normal and, as a consequence, often find it more attractive. This is often referred to as visual adaptation. For faces, the effects are typically specific to the category of faces you were shown. For example, experience with men’s faces that are manipulated to have wide-spaced eyes makes other men’s faces with wide-spaced eyes look more normal and attractive, but does not really alter how people perceive female faces with that trait. The effects of experience on responses to facial cues have provided insight into how cultural differences in face perception might emerge.
People in western cultures scan the whole face when judging emotional expressions. By contrast, people in eastern cultures tend to focus primarily on the eye regions. Focusing on the eye regions in this way can make it harder for them to distinguish between emotional expressions that share similar eye information but differ in other ways. For this reason, people in eastern cultures find it harder to distinguish between the emotions disgust and fear than people in western cultures do. Disgust and fear are very similar in the eye regions, but are very different in other face regions.
Although attractiveness judgments of faces and voices rely on different senses, there is some reason that they can be remarkably similar indeed. For example, women who show particularly strong preferences for masculine characteristics in men’s faces, like strong jaws and heavy brows, also tend to show particularly strong preferences for masculine characteristics in men’s voices, like low pitch. These results suggest that variation among women in their preferences for masculine characteristics is not random but reflects systematic differences in the extent to which women value the characteristics that masculine traits in faces and voices are associated with.
Do we like people who have face that looks like our own? The answer is more complicated than you might expect. Our research has shown that people perceive faces with shape characteristics similar to those in their own face to be particularly trustworthy and, when judging own-sex faces, more attractive. However, people tend to find opposite-sex faces with shape characteristics similar to those in their own face to be relatively unattractive, especially when judging the attractiveness of the faces for a sexual relationship. We seem to have evolved to cooperate with individuals displaying cues of kinship, but find other-sex individuals displaying those cues unappealing as mates.
The images below show “heat maps” of which facial muscles are active during emotional expressions.