Computational and psycho-physiological investigations of musical emotions

The ability of music to stir human emotions is a well known fact (Gabrielsson & Lindstrom. 2001). However, the manner in which music contributes to those experiences remains obscured. One of the main reasons is the large number of syndromes that characterise emotional experiences. Another is their subjective nature: musical emotions can be affected by memories, individual preferences and attitudes, among other factors (Scherer & Zentner, 2001). But can the same music induce similar affective experiences in all listeners, somehow independently of acculturation or personal bias? A considerable corpus of literature has consistently reported that listeners agree rather strongly about what type of emotion is expressed in a particular piece or even in particular moments or sections (Juslin & Sloboda, 2001). Those studies suggest that music features encode important characteristics of affective experiences, by suggesting the influence of various structural factors of music on emotional expression. Unfortunately, the nature of these relationships is complex, and it is common to find rather vague and contradictory descriptions. This thesis presents a novel methodology to analyse the dynamics of emotional responses to music. It consists of a computational investigation, based on spatiotemporal neural networks sensitive to structural aspects of music, which "mimic" human affective responses to music and permit to predict new ones. The dynamics of emotional responses to music are investigated as computational representations of perceptual processes (psychoacoustic features) and self-perception of physiological activation (peripheral feedback). Modelling and experimental results provide evidence suggesting that spatiotemporal patterns of sound resonate with affective features underlying judgements of subjective feelings. A significant part of the listener's affective response is predicted from the a set of six psychoacoustic features of sound - tempo, loudness, multiplicity (texture), power spectrum centroid (mean pitch), sharpness (timbre) and mean STFT flux (pitch variation) - and one physiological variable - heart rate. This work contributes to new evidence and insights to the study of musical emotions, with particular relevance to the music perception and emotion research communities.EThOS - Electronic Theses Online ServiceGBUnited Kingdo