What comes first: a face or a name?

As you get older, your face may change to resemble stereotypes associated with your name. Scientists have found that adults are more comfortable with their names than children are.

By William Harley

Parents have many decisions to make when raising a child. They have to decide how to discipline their child, how to feed them, how to keep them safe. The list is endless. Every decision impacts the child, but there is one decision that can have a uniquely significant impact on the child's future: the child's name.

A name can affect a child's future in many ways. It can hint at family history or bring ridicule from peers. However, one consequence that parents probably haven't thought about is that the name could affect how the child looks as an adult.

Scientists at Reichman University and the Hebrew University of Jerusalem have shown that adults with the same name look more alike than children with the same name. This finding suggests that a person's appearance can change to resemble stereotypes associated with a name. It provides concrete support for a phenomenon known as the “face-name matching effect,” in which the nickname a parent gives their child could predict what face they will have as an adult.

The face-name matching effect

The face-name matching effect assumes that your face can show your name or vice versa.

But how does someone get a name that seems to fit their face? There are two theories: the self-fulfilling prophecy and the birth-matching one.

The self-fulfilling prophecy assumes that appearance can be influenced by behavior, societal expectations and personality. A name conveys stereotypes and expectations conveyed by society. Individuals internalize these and integrate them into their identity. This can influence the appearance of the face through stylistic choices such as hair and makeup or facial expressions.

The theory that a baby is fit from birth arose from anecdotes in which parents renamed their baby from what they had originally planned to something they thought was “fit” after seeing their newborn. The implication here is that a baby can possess characteristics associated with a particular name.

RELATED TOPICS: Women in STEM: You don't look like a scientist

A matching and pattern game

The researchers tested these theories in three different ways: 1) with people matching names, 2) with machine learning identifying similarities between faces, and 3) by repeating the same two experiments with digitally aged faces, free of social pressure.

In the first experiment, both children and adults were given a picture of a child or adult and asked to match it to a name from a list. They repeated this for 36 trials. Both children and adults were able to match the adults' faces to the correct names with a much higher probability than if it had been purely by chance. They were unable to do this with the children's faces.

In the second experiment, a machine learning model was provided with a large selection of faces and their names to train its understanding of whether there is a connection between the face and the name. The model was then provided with new faces and looked for patterns it had learned during training. It was able to detect significantly more similarities between faces with the same name than by chance, confirming the results of the experiment with human participants.

RELATED TOPICS: Facial recognition: How many faces do you know?

Both results contradict the fit-from-birth theory, which assumes that children fit their names even if they are given them based on their appearance. Instead, these two experiments support the self-fulfilling prophecy theory. To further investigate the validity of this conclusion, another test was conducted.

The third experiment was based on the same methodology as the first two, using human subjects and machine learning, but with one key difference. Faces of real adults were replaced with digitally altered photos to simulate aging. With this sample, adults and children were unable to match faces with greater accuracy than if they had done so by chance. Furthermore, the machine learning model was unable to detect greater similarities between artificially aged faces than if it had guessed.

This result refutes the idea that the face-name matching effect is just a consequence of normal aging. The self-fulfilling prophecy prevails.

I am looking forward to

The researchers have convincingly demonstrated that the appearance of a face can change to match a person's name, but there are a number of other interesting findings and future work that emerge from this study.

The observation that children were able to match names to faces suggests that stereotypes associated with faces are learned at an early age—before the child's face itself can represent a stereotype. This calls for an investigation into how and when stereotypes are reinforced with faces.

In addition, a future study could examine how parents detect subtle cues when naming their infants and compare individuals named before and after birth.

This study was published in the peer-reviewed journal. Proceedings of the National Academy of Sciences.

reference

Zwebner, Y., Miller, M., Grobgeld, N., Goldenberg, J., & Mayo, R. (2024). Can names influence facial appearance? Proceedings of the National Academy of Sciences, 121(30), e2405334121.

Featured image “Hello, my name is anonymous” by quinn.anya is licensed under CC BY-SA 2.0.

About the author

William Harley earned degrees in biochemistry, molecular biology, and chemical engineering from Oregon State University. Although he is an engineer by profession, he enjoys making music, trail running, and baking.