Most of my current research involves explaining our perception of faces in terms of physical differences between different kinds of faces. For example, the physical anthropology community has shown that male skin is darker than female skin (e.g., the work of Peter Frost). I have shown that while male skin is darker than female skin, male eyes and lips are not much darker than female eyes and lips. The result is greater contrast in female faces between the eyes, mouth, and the rest of the face.
People use this sex difference in facial contrast to decide the sex of a face and how masculine or feminine it is. Manipulating this facial contrast has opposite effects on male and female attractiveness. Interestingly, cosmetics exaggerate this sex difference, which suggests that cosmetics are used to manipulate sex differences to make the female face more feminine, and hence attractive. Recent work in my lab has shown that aspects of facial contrast also decrease with age.
Prosopagnosia (also called face blindness) is an impairment in the recognition of faces. More information about prosopagnosia can be found at Faceblind.org. Some people are the opposite of prosopagnosic, and have excellent face recognition ability. I have named these people "super-recognizers."
Although there is significant agreement about the relative attractiveness of different people, the agreement is far from complete. I am also very interested in individual differences in attractiveness preferences, and how they can be affected by relationships, experiences, and learning.
It is commonly assumed that shape is the dominant feature for face recognition, as is the case for most object classes. However, research that I have conducted with Ken Nakayama at Harvard University, Pawan Sinha at MIT and Irving Biederman at the University of Southern California, and work by others (especially Alice O'Toole at the University of Texas at Dallas), has shown that surface reflectance and shape properties are about equally important for face recognition.