Trustworthiness judgments without the halo effect: A data-driven computational modeling approach.

Trustworthy-looking faces are also perceived as more attractive, but are there other meaningful cues that contribute to perceived trustworthiness? Using data-driven models, we identify these cues after removing attractiveness cues. In Experiment 1, we show that both judgments of trustworthiness and attractiveness of faces manipulated by a model of perceived trustworthiness change in the same direction. To control for the effect of attractiveness, we build two new models of perceived trustworthiness: a subtraction model, which forces the perceived attractiveness and trustworthiness to be negatively correlated (Experiment 2), and an orthogonal model, which reduces their correlation (Experiment 3). In both experiments, faces manipulated to appear more trustworthy were indeed perceived to be more trustworthy, but not more attractive. Importantly, in both experiments, these faces were also perceived as more approachable and with more positive expressions, as indicated by both judgments and machine learning algorithms. The current studies show that the visual cues used for trustworthiness and attractiveness judgments can be separated, and that apparent approachability and facial emotion are driving trustworthiness judgments and possibly general valence evaluation.

Learning the affective value of people: More than affect-based mechanisms.

People's ability to learn about the affective value of others is impressive. However, it is unclear whether this learning solely reflects general affect-based processes or a mixture of affect-based and person-attribution processes. Consistent with the former possibility, people's ability to learn the affective value of people and places have been shown to be comparable (Falvello, Vinson, Ferrari, & Todorov, 2015). To investigate whether general affect-based processes are sufficient to account for this kind of learning, we presented participants with images paired with valenced statements that were either relevant (e.g., a person statement with a person image) or irrelevant (e.g., a person statement with a non-person image). After this presentation, participants evaluated the goodness or badness of the images. In Experiment 1, we found that the learning effects for faces and places were comparable and occurred only when the statements were relevant. However, when we presented the images with multiple statements of the same valence (Experiments 2-4), we found that places acquired affective value from both relevant and irrelevant statements. In contrast, faces were less likely to acquire affective value from irrelevant statements. Our findings suggest that although general affect-based processes might be sufficient to account for affective learning of places, affective learning of faces might involve both affect-based and person-attribution processes.