Disparity in performance on tone-scramble tasks: generalizability and relevance to music

Music has a remarkable power to arouse the feelings of those who listen to it. What features of music imbue it with such emotional resonance? A prevailing notion in music theory is that musical scale has a central role in giving meaning to music. Indeed, many studies find that, according to listeners' average ratings, music of the major scale sounds happy, and music of the minor scale sounds sad. However, recent discoveries involving “tone-scramble” stimuli complicate our understanding of these results and suggest that sensitivity to scale is not universal. This thesis details a series of experiments designed to investigate (1) the generalizability of the findings of laboratory-based tone-scramble experiments and (2) the musical nature of a latent cognitive resource that is theorized to underlie performance in tone-scramble tasks. Chapter 1 reports that the same bimodal distribution in performance repeatedly observed in laboratory-based tone-scramble experiments is observed in a large, linguistically diverse web-based sample. Chapter 2 considers whether low-performing listeners are limited by some of the non-musical qualities inherent to tone-scrambles; data are provided to show that changes in presentation rate, frequency height, and timbre that yield tone-scrambles akin to actual music do not provide low-performing listeners any advantage over ordinary tone-scrambles. In Chapter 3, the latent cognitive resource theorized to underlie performance on tone-scramble tasks is shown to operate on musical scale and not individual frequencies, drawing a clear relationship between performance in tone-scramble tasks and sensitivity to musical scale