Recent faces can shift snap judgments of trustworthiness

People often form a quick impression of a stranger’s trustworthiness from their face. Researchers have studied which facial features tend to lead to these impressions. This study asked a different question: can the face a person has just seen change their impression of the next face?

The researchers used a task in which people rated facial trustworthiness. They examined whether three basic features of the viewing experience changed these ratings: how long a previous face was shown, how similar the previous and current faces were, and how far apart the faces appeared on the screen.

The study found two opposite kinds of shifts. After a previous face was shown only briefly, ratings of the current face tended to move toward the earlier face’s rating. Researchers call this an attractive bias, or serial dependence. In everyday terms, the earlier face seemed to pull the impression of the next one closer to it.

After a previous face was viewed for longer, the pattern changed. Ratings of the current face tended to shift away from the earlier face’s rating. This is known as a negative aftereffect, or repulsive bias.

How alike the faces looked also mattered. Greater similarity in identity between the earlier and current faces strengthened the attractive, toward-the-earlier-face effect. When the faces had different identities, the researchers found a shift in the opposite direction, away from the earlier face.

Screen position played a more specific role. The relative distance between the earlier and current face affected the strength of the attractive bias. It did not affect the negative aftereffect in the same way.

Together, the results suggest that trustworthiness impressions are not fixed responses to one face alone. They can be influenced by what a viewer has just seen, and by timing, visual similarity, and spatial arrangement.

The authors say these patterns cannot be fully explained by higher-level processes such as remembering an earlier response or making a decision based on it. The role of basic visual features points to an effect of recent visual experience itself.

This was a controlled rating task, so it cannot show that a person’s real-life choices will change in the same way. It also does not show that faces judged as more or less trustworthy truly are more or less trustworthy. The supplied study information does not report the number or characteristics of participants, the size of the effects, or how long any influence lasted.

Further research would need to test whether these patterns appear in more realistic settings, such as reviewing profiles or making other rapid social judgments. It would also need to clarify how large and reliable the effects are across different people and face images.