The Constructivistly-Organised Dimensional-Appraisal (CODA) Model and Evidence for the Role of Goal-directed Processes in Emotional Episodes Induced by Music

Abstract

The study of affective responses to music is a flourishing field. Advancements in the study of this phenomena have been complemented by the introduction of several music-specific models of emotion, with two of the most well-cited ones being the BRECVEMA and the Multifactorial Process Model. These two models have undoubtedly contributed to the field. However, contemporary developments in the wider affective sciences (broadly described as the ‘rise of affectivism’) have yet to be incorporated into the music emotion literature. These developments in the affective sciences may aid in addressing remaining gaps in the music literature, in particular for acknowledging individual and contextual differences.

The first aim of this thesis was to outline contemporary theories from the wider affective sciences and subsequently critique current popular models of musical emotions through the lens of these advancements. The second aim was to propose a new model based on this critique: the Constructivistly-Organised Dimensional-Appraisal (CODA) model. This CODA model draws together multiple competing models into a single framework centralised around goal-directed appraisal mechanisms which are key to the wider affective sciences but are a less commonly acknowledged component of musical affect. The third aim was to empirically test some of the core hypotheses of the CODA model. In particular, examining goal-directed mechanisms, their validity in a musical context, and their ability to address individual and contextual differences in musically induced affect. Across four experiments which include exploratory and lab-based designs through to real- world applications, the results are supportive of the role of goal-directed mechanisms in musically induced emotional episodes. Experiment one presents a first test battery of multiple appraisal dimensions developed for music. The results show that several of the hypothesised appraisal dimensions are valid dimensions is a musical context. Moreover, these mechanisms cluster into goal-directed latent variables. Experiment two develops a new set of stimuli annotations relating to musical goals, showing that music can be more or less appropriate for different musical goals (functions). Experiment three, using the new stimuli set from experiment two, tests the effects of different goals with more or less appropriate music on musically induced affect. These results show that goal-directed mechanisms can change induced core-affect (valence and arousal) and intensity, even for the same piece of music. Experiment four extends the study of goal-directed mechanisms into a real-world context through an interdisciplinary and cross-cultural design. The final experiment demonstrates how goal-directed mechanisms can be manipulated through different algorithms to induce negative affect in a Colombian population.

The main conclusions of this thesis are that the CODA model, more specifically goal-directed mechanisms, provide a valuable, non-reductive, and more efficient approach to addressing individual and contextual differences for musically induced emotional episodes in the new era of affectivism.