The photos of faces do not make it possible to predict someone’s willingness to cooperate.

Quick prints from photos of faces are unreliable according to recent study
Quick prints from photos of faces are unreliable according to recent study

A study published in "Scientific Reports", in which the Autonomous University of Madrid (UAM) participates, shows that our quick impressions based on pictures of faces barely overcome chance when assessing a person’s willingness to cooperate. Although intuition seems to play a role in identifying cooperators, the results suggest that the cues visible in static images are not sufficient to accurately predict cooperative behaviors.

In our daily interactions we tend to form quick impressions based on physical appearance, especially the face, often from static images. This happens, for example, when browsing social networks, reviewing a CV or using dating apps, where we make decisions about someone’s trustworthiness in a matter of seconds, based almost exclusively on their photographs.

A new multidisciplinary study, led by researchers from the Autonomous University of Madrid (UAM), Leuphana University of Lüneburg (Germany) and King’s College London (UK), set out to explore whether it is possible to infer a person’s willingness to cooperate by looking at photos of their face alone.

To do this, the researchers used an economic game known as the ’prisoner’s dilemma’, which measures the tendency to cooperate, and asked a group of 300 participants to evaluate this tendency by looking only at photos of the players’ faces.

The results, published recently in Scientific Reports, revealed that the accuracy in assessing cooperation was only slightly higher than expected by chance, indicating that face photos do not provide enough cues to reliably predict whether someone will be cooperative. However, significant differences were observed under certain conditions.

"For example, participants who had a limited time of maximum 5 seconds to make their evaluations were more accurate in identifying cooperative subjects," says co-author Enrique Turiegano of UAM’s Department of Biology.

"This," the researcher adds, "suggests that intuition, understood as quick and poorly thought-out judgment, may have a limited role in detecting cooperative individuals.

Unconscious biases

The researchers observed that participants tended to be more accurate in identifying cooperators if they expected others to be cooperative, and were also more effective at identifying noncooperators if they expected them to be cooperative. However, the overall ability to detect cooperators from photos remained limited. This is consistent with previous research suggesting that cooperative cues are often subtle and difficult to pick up in static images.

In addition, the study identified certain biases, such as the tendency to misperceive people with certain characteristics (women, older people, women with very feminine traits) as cooperative, which opens questions about the role of unconscious biases in our perceptions.

"In sum," concludes Turiegano, "although headshots influence our impressions of willingness to cooperate, they are not reliable indicators of this behavior. This leads us to reconsider how we judge others and underscores the importance of evaluating additional factors before deciding to trust someone, especially in sensitive or compromising situations."

Bibliographic reference:

Lohse, J., Sanchez-Pages, S., & Turiegano, E. (2024). T he role of facial cues in signalling cooperativeness is limited and nuanced. Scientific Reports, 14(1), 22009.

More scientific culture in UAM Gazette