Dissertation Defense: Hassan Aleem
Candidate Name: Hassan Aleem
Major: Neuroscience
Advisor: Norberto Grzywacz, Ph.D.
Title: The Learning and Persistence of Aesthetic Values
How do our aesthetic preferences form? Each of us starts from a relatively similar state at birth, yet we end up with vastly different sets of aesthetic preferences. These preferences eventually define us both as individuals and as members of our cultures. Therefore, it is important to understand how aesthetic preferences develop and persist over our lifetimes. Philosophical accounts suggest two contrasting perspectives to this question. The objective perspective states that aesthetic preferences are located in the qualities of the world, whereas the subjective perspective states that aesthetic preferences are entirely confined to the observer. Here, I test these hypotheses in a series of studies through a variety of methods including statistical image analyses, computational modeling, and behavioral experiments. If aesthetic preferences are objective qualities, then artists throughout history should exhibit a bias towards them. I test this hypothesis by analyzing images of 153 portrait paintings from 26 master painters from the Early Renaissance. The results of this study show painters do not show this bias. Instead, each painter optimizes different combinations of attributes, displaying individuality. Therefore, suggesting a subjective component. How may subjectivity in aesthetic preferences arise? While this question has been addressed by many conceptual models, a computationally valid framework has been missing. Therefore, I present a computational model for the learning of aesthetic values. Based on neurobiological evidence, I propose that aesthetic values are formed through reward-based learning. This model builds on canonical reinforcement learning circuitry with crucial extensions related to internal motivations and societal contingencies on reward. Combined, these factors help explain the emergence of both culturality and individuality in aesthetic preferences. Finally, I test one of the predictions of the computational framework; the variability of aesthetic values over time. To do this, I measure the aesthetic preferences of 85 individuals over the course of a month. The results of this study confirm that aesthetic preferences are unstable. However, that instability is dynamic and varies greatly across individuals. Overall, the studies presented here show that aesthetic preferences formation relies on a combination of objective and subjective mechanisms, ultimately leading to individuality.