Forming impressions

Although people readily form impressions based solely on facial appearance, they are also quite good at updating these impressions once additional information, such as information about a person’s past actions, becomes available.  In my behavioral work, I have found that people readily retrieve the affective knowledge associated with familiar faces, even under circumstances that prevent careful reflection (Verosky et al., 2018).  A recent project in the lab explores how people communicate facial impressions with each other.  On the one hand, facial appearance has a pervasive influence on a range of outcomes, suggesting that people might communicate their facial impressions directly.  On the other hand, as expressed by the adage “Don’t judge a book by its cover”, there is stigma against judging others based on facial appearance.  A current and future direction in the lab is examining how exposure to multiple images of a person influences the formation and updating of impressions.

Neural Responses to Faces

In my work examining the neural basis of face processing, I switched from using fMRI to using EEG when I came to Oberlin because it is more cost effective and because I am able to collect the data on campus and train students to participate in the research.  Recent studies in our lab have used a technique called fast periodic visual stimulation (FPVS), where the brain is repetitively stimulated at a constant frequency, resulting in a periodic response at the rate of stimulation and its harmonics that can be recorded using EEG.  People are better at recognizing familiar faces across a variety of changes in visual properties, and our lab has found that a minimal learning manipulation is enough to increase the face individuation response as measured via FPVS (Verosky et al., 2020a).  We have also demonstrated that it is possible to detect implicit responses to perceived face trustworthiness using FPVS (Verosky et al., 2020b).  A current direction brings these previous studies together by investigating whether it is possible to detect implicit responses to the trustworthiness of a person’s past actions using FPVS (Verosky et al., 2024).