Describing aesthetics experiences with scientific models is a complex task, but there seems to be an emerging consensus that three main areas contribute to psychological experiences of aesthetic stimuli: 1) elements of the aesthetic object itself, 2) characteristics of the person, and 3) situational as well as contextual information. Taking music as an example domain we will demonstrate rigorous approaches that allow the construction of empirical models within each of the three areas. 1) We suggest that the comprehensive computational analysis of stimulus features of aesthetic objects can be employed to describe relationships with psychological responses (Jakubowski, Müllensiefen & Stewart, 2016; van Balen et al., 2015). In addition we will discuss new strategies going beyond mere feature analysis to gain a deeper understanding of how elements of the aesthetic object are causally linked to perceptual and cognitive responses. 2) For assessing characteristics of the individual we advocate the use of modern psychometric techniques such as item response theory, automatic item generation and adaptive testing (e.g. Harrison, Collins & Müllensiefen, 2017). Through examples we’ll show how ambiguous empirical results are often caused by fundamental measurement problems and by ignoring individual differences. 3) In addition, we’ll present experimental approaches (Anglada-Tort & Müllensiefen, 2017) inspired by paradigms from behavioral economics (Kahneman, 2011) that aim to assess the degree to which biases and heuristics introduced through the situational context affect human judgements of aesthetic stimuli.
The unifying conceptual bracket of these approaches is a thorough understanding of causality (Pearl, 2011) and its implementation in experimental and observational research. Finally, we’ll make sure the talk will be more fun than this abstract is able to convey!