What makes some faces more attractive than others? news.sciencemag.org/brain-behavior/2014/07/video-what-makes-some-faces-more-attractive-others Each frame of the video above is a different face with the features slightly tweaked. If youre like the U.K. university students who took part in a new study, the faces should strike you as increasingly dominant personalities, peaking in the middle of the video and then becoming meek again by the end. The statistical model that underlies those cartoon faces was derived from a study of 1000 photographs of Caucasian faces that the students looked at and scored for various traits. Although there is plenty of noise in these data, people tended to judge faces similarly. But what is it exactly about the faces that was being judged? To find out, researchers measured 393 image properties, from the size of the eyes and cheeks to the shape of the chin. By mapping these properties to peoples judgments of the faces, the researchers found that first impressions are surprisingly predictable—58% of the variation in judgments could be explained just by the relative size and position of a few facial traits, the team reports online today in the Proceedings of the National Academy of Sciences. For example, people with mouths that have a natural smile shape are seen as far more approachable, while larger eyes are the strongest factor in attractiveness. To make the face-morphing videos, the researchers then did the reverse, generating cartoon faces from scratch based on their statistical model to elicit predictable snap judgments. Wondering about the possible applications? By slightly tweaking a photograph of someones face—a politician, for example—it may be possible to custom-design first impressions. - Facial features are the key to first impressions A new study by researchers in the Department of Psychology at the University of York shows that it is possible to accurately predict first impressions using measurements of physical features in everyday images of faces, such as those found on social media. york.ac.uk/news-and-events/news/2014/research/first-impressions/ When we look at a picture of a face we rapidly form judgements about a person’s character, for example whether they are friendly, trustworthy or competent. Even though it is not clear how accurate they are, these first impressions can influence our subsequent behaviour (for example, judgements of competence based on facial images can predict election results). The impressions we create through images of our faces (“avatars” or “selfies”) are becoming more and more important in a world where we increasingly get to know one another online rather than in the flesh. Previous research has shown that many different judgements can be boiled down to three distinct “dimensions”: approachability (do they want to help or harm me?), dominance (can they help or harm me?) and youthful-attractiveness (perhaps representing whether they’d be a good romantic partner - or a rival!). To investigate the basis for these judgements the research team took ordinary photographs from the web and analyzed physical features of the faces to develop a model that could accurately predict first impressions. Each of 1,000 faces was described in terms of 65 different features such as “eye height”, “eyebrow width” and so on. By combining these measures the model could explain more than half of the variation in human raters’ social judgements of the same faces. Reversing the process it was also possible to create new cartoon-like faces that produced predictable first impressions in a new set of judges. These images also illustrate the features that are associated with particular social judgements. The study, published today in Proceedings of the National Academy of Science (PNAS), shows how important faces and specific images of faces can be in creating a favourable or unfavourable first impression. It provides a scientific insight into the processes that underlie these judgements and perhaps into the instinctive expertise of those (such as casting directors, portrait photographers, picture editors and animators) who create and manipulate these impressions professionally. Richard Vernon, a PhD student who was part of the research team, said: “Showing that even supposedly arbitrary features in a face can influence peoples perceptions suggests that careful choice of a photo could make (or break) others’ first impressions of you.” Fellow PhD student, Clare Sutherland, said: “We make first impressions of others so intuitively that it seems effortless - I think its fascinating that we can pin this down with scientific models. Im now looking at how these first impressions might change depending on different cultural or gender groups of perceivers or faces.” Professor Andy Young, of the Department of Psychology at York, said: “Showing how these first impressions can be captured from very variable images of faces offers insight into how our brains achieve this seemingly remarkable perceptual feat.” Dr Tom Hartley, who led the research with Professor Young, added: “In everyday life I am not conscious of the way faces and pictures of faces are influencing the way I interact with people. Whether in “real life” or online; it feels as if a person’s character is something I can just sense. These results show how heavily these impressions are influenced by visual features of the face - it’s quite an eye opener!” Reference Modeling first impressions from highly variable facial images PNAS 2014, doi: 10.1073/pnas.1409860111 pnas.org/content/early/2014/07/23/1409860111 Significance Understanding how first impressions are formed to faces is a topic of major theoretical and practical interest that has been given added importance through the widespread use of images of faces in social media. We create a quantitative model that can predict first impressions of previously unseen ambient images of faces (photographs reflecting the variability encountered in everyday life) from a linear combination of facial attributes, explaining 58% of the variance in raters’ impressions despite the considerable variability of the photographs. Reversing this process, we then demonstrate that face-like images can be generated that yield predictable social trait impressions in naive raters because they capture key aspects of the systematic variation in the relevant physical features of real faces. Abstract First impressions of social traits, such as trustworthiness or dominance, are reliably perceived in faces, and despite their questionable validity they can have considerable real-world consequences. We sought to uncover the information driving such judgments, using an attribute-based approach. Attributes (physical facial features) were objectively measured from feature positions and colors in a database of highly variable “ambient” face photographs, and then used as input for a neural network to model factor dimensions (approachability, youthful-attractiveness, and dominance) thought to underlie social attributions. A linear model based on this approach was able to account for 58% of the variance in raters’ impressions of previously unseen faces, and factor-attribute correlations could be used to rank attributes by their importance to each factor. Reversing this process, neural networks were then used to predict facial attributes and corresponding image properties from specific combinations of factor scores. In this way, the factors driving social trait impressions could be visualized as a series of computer-generated cartoon face-like images, depicting how attributes change along each dimension. This study shows that despite enormous variation in ambient images of faces, a substantial proportion of the variance in first impressions can be accounted for through linear changes in objectively defined features. Supporting Information pnas.org/content/suppl/2014/07/23/1409860111.DCSupplemental/pnas.201409860SI.pdf Download Movie_S01 (AVI) pnas.org/content/suppl/2014/07/23/1409860111.DCSupplemental/pnas.1409860111.sm01.avi Movie S1. Animation showing changes in facial features associated with the Approachability dimension. Face-like images generated using the approach illustrated in Fig. 1 C and D with target factor scores varying from (−1, 0, 0) to (1, 0, 0) (and back again). Animation showing changes in facial features associated with the Approachability dimension. Does this person want to harm me, or will they help me? The video starts with the unapproachable-looking face and moves toward an approachable-looking face and then back again. It can be looped. Download Movie_S02 (AVI) pnas.org/content/suppl/2014/07/23/1409860111.DCSupplemental/pnas.1409860111.sm02.avi Movie S2. Animation showing changes in facial features associated with the Youthful-Attractiveness dimension. Face-like images generated using the approach illustrated in Fig. 1 C and D with target factor scores varying from (0, −1, 0) to (0, 1, 0) (and back again). Animation showing changes in facial features associated with the Youthful-Attractiveness dimension. Might this person be a suitable romantic partner (or love rival)? The video starts with the older, less attractive-looking face and moves toward a younger, more attractive-looking face and then back again. It can be looped. Download Movie_S03 (AVI) pnas.org/content/suppl/2014/07/23/1409860111.DCSupplemental/pnas.1409860111.sm03.avi Movie S3. Animation showing changes in facial features associated with the Dominance dimension. Face-like images generated using the approach illustrated in Fig. 1 C and D with target factor scores varying from (0, 0, −1) to (0, 0, 1) (and back again). Animation showing changes in facial features associated with the Dominance dimension. Does this person have the capacity to carry out their intentions toward me? The video starts with least dominant appearance and moves toward the most dominant-looking face and then back again. It can be looped.
Posted on: Tue, 05 Aug 2014 23:03:37 +0000
Trending Topics
Recently Viewed Topics
© 2015