Revealing the information hidden in faces

The Face Facts team is a multidisciplinary group that combines Psychology with Biology, Computer Science and Anthropology.

Institute of Neuroscience & Psychology
University of Glasgow



New Face Skin Spin   + Holo


How do humans recognise kin?

The Kinship project is a 5-year European Research Council funded project investigating the question “How do humans recognise kin?”

This project is led by Dr Lisa DeBruine, one of the founders of Face Facts. Dr Iris Holzleitner and Kieran O'Shea are postdoctoral researchers employed on the project and Vanessa Fasolt is a PhD student involved with the project. Dr Anthony Lee was employed as a postdoc on the project from 2015-2016 and is currently completing a 2-year ERC Marie Curie Fellowship in the lab.

Participating in the Project

If you had your 3D Family Photos taken at the Glasgow Science Centre or at Hunter Halls for the 2016 Glasgow Science Festival, you can access your 3D interactive face at My Faces by entering your photo code.

You can email kinship@facelab.org to set up an appointment to have your photos taken in our lab on the University of Glasgow Hillhead campus. We are keen to photograph families from a diverse range of ethnic backgrounds to combat the bias of most research on faces being done with only white or East Asian faces.

KINSHIP in the Media

  • Your Face Says it All (2016) Channel 4

    Dr DeBruine spoke on face perception generally and facial resemblance among kin.
  • Finding My Twin Stranger (2016) Channel 4

    We assessed facial resemblance between doppelgängers using 2D and 3D image assessments developed for the KINSHIP project. Dr Holzleitner was featured in the documentary running the 3D camera, while Dr DeBruine was featured in the documentary explaining the science behind facial similarity metrics to the participants.
  • Cousins: A Wedding In The Family (production title) BBC3

    We created self-resembling and father-resembling images for the subject of the documentary, an 18-year-old trying to decide whether or not to pursue a cousin marriage. Dr DeBruine also spoke to her about the research on mate choice and family resemblance.


  • DeBruine, L. M., Jones, B. C. & Little, A. C. (2017). Positive Sexual Imprinting For Human Eye Color. BioRχiv (preprint). doi: 10.1101/135244
  • DeBruine, L. M., Hahn, A. C. & Jones, B. C. (2016). Perceiving infant faces. Current Opinion in Psychology, 7: 87-91. doi: 10.1016/j.copsyc.2015.08.010
  • Wolffhechel, K., Hahn, A. C., Jarmer, H., Fisher, C., Jones, B. C. & DeBruine, L. M. (2015). Testing the utility of a data-driven approach for assessing BMI from face images. PLoS One, 10. doi: 10.1371/journal.pone.0140347
  • Dal Martello, M. F., DeBruine, L. M. & Maloney, L. T. (2015). Allocentric kin recognition is not affected by facial inversion. Journal of Vision, 15(13): 5. doi: 10.1167/15.13.5
Iris, Robbie, Ant and Amanda at the Glasgow Science Centre.

This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (Grant No 647910)

Institute of Neuroscience & Psychology University of Glasgow


Combining Psychology with Biology,

Computer Science and Anthropology

Combining Psychology with Biology, Computer Science and Anthropology
Select tags below and click a face to see the average
Institute of Neuroscience & Psychology
University of Glasgow



Participate in demonstrations of our experimental studies and learn more about how we use computer-graphics to understand how humans perceive faces.

Institute of Neuroscience & Psychology
University of Glasgow


Who looks healthier?

Click on the face of the baby who looks cuter

More Experiments The Science


Find the Expression

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.

Institute of Neuroscience & Psychology University of Glasgow


Take My Picture


Image Info

Who is in this image?

Someone I know
they know I'm using this image
Someone I know
they don't know I'm using this image
A public figure


Image Info

What is the sex/gender of the person in this image?



Image Info

What is the facial expression of this image?



Image Info

What is the ethnic background of the person in this image? (choose all that apply)

Black / African / Caribbean
East Asian / Pacific Islander
West Asian / Indian
White / European
Native American
Latin@ / Hispanic
Other Ethnicity
Not Human


Image Info

What is the age of the person in this image?

under 10
over 100


Image Info

Where is the person in this image from? (select all that apply)


Image Info

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.

Popular Tags


Average Faces - Click on the faces below to select them

Select categories below
My Faces
Harry Potter
Sci Fi
Click on the faces above to select them, then click the average button


My Faces

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.

My Exhibit 3D Faces (click to open in the 3D viewer)
My Exhibit Images
My Uploaded Images

If you are not logged in, uploaded images will not be accessible after you leave this site.

Continue Without Login Login
Login Register Reset Password

Are you sure you want to delete these images?

Delete Cancel


Transform Faces

Select a face, then choose a transform below
Select a face above, then choose a transformation



Institute of Neuroscience & Psychology University of Glasgow


About Us

The Organisers

The Scientific Team

The Support Team

Institute of Neuroscience & Psychology University of Glasgow



Click on the questions below to learn more about the research we do using computer-graphic faces.

How do we use computer graphic techniques for manipulating face images?

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.

A male face transformed from more feminine than average to more masculine than average

How do we use average faces to study social judgments?

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.

Average healthy-looking face
Average unhealthy-looking face

What different types of facial cues matter for social judgments of faces?

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.

What are top-down and bottom-up analyses?

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.

An aggressive mandrill
An aggressive man

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.

Does experience shape how people respond to facial cues?

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.

How does culture influence responses to facial cues?

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.

Eastern Disgust
Eastern Fear
Western Disgust
Western Fear

How similar are face and voice preferences?

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.

How do we respond to people whose faces look like ours?

Do we like people who have faces 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.

Two women (left) with their virtual sisters (middle) and virtual brothers (right)

How do we visualise emotions?

The images below show “heat maps” of which facial muscles are active during emotional expressions.


Institute of Neuroscience & Psychology University of Glasgow