It’s uncontroversial to point out that people have preferences and these preferences can vary considerably from one person to the next. My niece loves the color yellow; my grandmother hates the color yellow. But, it’s more interesting to point at how consistent preferences can be. Almost everyone likes certain shades of blue. In the first part of this dissertation, I examine people’s preferences in two domains.

First, I examine how well people’s preferences can be predicted for a certain kind of images known as visual textures (Chapter 1). I find that people’s preferences can be explained well by an ecological model according to which people like visual textures to the degree that they like the objects most-associated with those textures.

Next, I examine people’s preferences for faces (Chapter 2). One of the most robust findings in face research is that people rate faces with average configurations as highly attractive. Despite the consistency of this finding, we still don’t know why this should be true. In Chapter 2, I use a statistical mediation model to investigate why averaged faces are so attractive. I find that the result is not explained by any of the mediators tested and argue that averageness per se is attractive.

Finally, I pivot in the third chapter: Instead of examining the surprising consistency of high-level experience (preference ratings), I examine the potential idiosyncrasy of low-level experience. Low-level experiences – of things like color and lightness – are thought to be relatively consistent across people. But, recently, researchers have claimed that low-level perception can be influenced by idiosyncratic cognitive factors like beliefs or desires. In the third chapter, I investigate whether the perceived race of a face (a high-level cognitive construct) can influence how light its skin looks (low-level lightness perception), regardless of how light its skin actually is. Contrary to previous claims, I do not find support for this assertion.

Overall I find that, on the one hand, people are surprisingly consistent in their preferences, both for visual textures and for faces. On the other hand, I find no evidence that idiosyncratic cognitive factors can affect low-level perception.