Two people instantly judge each other based on their appearance. These first impressions can have a huge impact on our relationships, and even influence criminal sentencing. In collaboration with Princeton University, Stevens Institute of Technology has developed an AI algorithm that can accurately predict how people will perceive them based on their first impressions. This algorithm was created by researchers at Stevens Institute of Technology. This work is published in the Proceedings of the National Academy of Sciences.
Jordan W. Suchow is a cognitive scientist at Stevens and an AI expert. He said that there’s a lot of research on modeling people’s physical appearances. “There’s a wide body of research that focuses on modeling the physical appearance of people’s faces. We’re bringing that together with human judgments and using machine learning to study people’s biased first impressions of one another.”
Suchow and his team at Chicago Booth, which included Joshua Peterson and Thomas Griffiths at Princeton and Stefan Uddenberg at Chicago Booth, asked thousands to rate their impressions of more than 1,000 computer-generated photos. The results were ranked according to how intelligent, electable and trustworthy the subject of each photograph was. These responses were used to train a neural system to make similar snap judgments of people based on only photographs of their faces.
Suchow explained that if you take a picture of your face, you can use the algorithm to predict how people will first perceive you and what stereotypes they will project onto your face.
Many of the algorithm’s results align with cultural assumptions or common intuitions. People who smile are more trustworthy than people who wear glasses. It’s sometimes a bit more difficult to see why an algorithm attributes a certain trait to someone.
Suchow stated that the algorithm does not provide any targeted feedback or explain why an image evokes a particular judgement. It can still help us understand how we are seen. For example, we could rank photos according to which makes you most trustworthy. This would allow you to make decisions about how to present yourself.
Although the algorithm was originally created to aid psychological researchers in creating face images for experiments on perception and social cognition. However, it could be used in real life. According to Suchow, people carefully curate their public image, sharing only those photos that they feel make them most attractive or intelligent. He said that there is already a social norm for presenting yourself positively, which helps to avoid some ethical concerns.
The algorithm can also be used for manipulating photos to make a subject look a certain way. This could include making a candidate seem more trustworthy or making their opponent appear unintelligent and suspicious. The new algorithm could subtly alter images to influence the viewers’ opinions about the subjects.
Suchow stated that it was possible to modify a photograph to create an impression. We need to be cautious about how we use this technology, for obvious reasons.
The research team obtained a patent to protect their technology and is creating a startup in order to license the algorithm for ethical purposes. Suchow stated that they are taking every precaution to make sure this technology doesn’t cause harm.
The current algorithm only predicts how people will react to one person’s face. However, Suchow hopes to create an algorithm that can predict how individuals will respond to each other’s faces. This could provide richer insight into how snap judgments influence social interactions and help people recognize and see beyond first impressions when making crucial decisions.
Suchow stated that “It’s important to remember that the judgments we’re modeling don’t reveal anything about a person’s actual personality or competencies. What we’re doing here is studying people’s stereotypes, and that’s something we should all strive to understand better.”