Research in expressing continuous emotional response to music as a function of its psychoacoustic parameters: Current and future directions

Continuous response of emotion in music is undergoing a boom. One application of this kind of research is to gain further insight into the relationship between musical/psychoacoustic parameters and associated emotional responses. Continuous response methodology has potential to produce major new insights into these relationships, however there are pitfalls and challenges associated with it. The present paper discusses some of the methodological issues that may be hindering progress. Many of the problems s tem f rom t h e complex, interdisciplinary nature of this kind of research. In the past I have, for example, commented on some of the statistical issues and concerns. In this paper I will discuss problems the measurement of psychoacoustic/musical parameters, and in coding and relating them back to the emotional perception. There are various issues that require consideration regarding the coding of loudness, pitch, articulation, spectral centroid, harmony and so forth, some of which appear to be unresolved in the literature concerned with emotional response: What kind of loudness algorithm should be used, how can melody, harmony, articulation and so forth be coded and sampled, how should centroid be measured, and what is the relationship between the psychoacoustic and perceptual (nonemotional) parameters? What kinds of visual aids can be used to assist with analysis? In addressing some of these issues, I conclude that researchers need to incorporate new methodologies and be cognizant of potential problems.