What Emotions and Free Associations Characterize Different Musical Styles?

Abstract

Bisesi, Parncutt, & Friberg (2011) extended Director Musices (DM) - a software package for automatic rendering of expressive performance (Friberg et al., 2006) that relates the expressive features of a performance not only to global or intermediate structural properties, but also to local events (grouping, metrical, melodic and harmonic accents). Friberg & Bisesi (2013) applied the new rule system to model stylistic variations corresponding to different historical periods and styles in piano music. In this project, we extend pDM - a pure data system for real-time expressive control of music performance (Friberg, 2006) - by adjusting algorithmic formulations inspired by a theory of musical accents and by introducing new gestural interfaces based on qualitative data. First, different presets corresponding to different stylistic conventions and interpretations are implemented in DM, and corresponding sound files are generated and recorded. Second, we ask 20 participants (10 musicians and 10 amateurs) to rate the character of the performances on rating scales depending on a previous categorization involving emotions and free associations (Bisesi, Bodinger, & Parncutt, in progress). Terms involve basic and complex emotions, and free associations such as static/dynamic and living/non-living things. Results will be used to map music expression into multidimensional spaces of emotions and free associations, and to develop new gestural interfaces in pDM.