Guiding music guiding sleep: an investigation of subjective and objective properties of music that facilitates sleep induction

This thesis presents three papers advancing a systematic approach to understanding what music helps with sleep. The first reveals the distinctiveness of sleep music by examining playlists from the music streaming service Spotify, using their vast public data source to study the features of music in sleep playlists. The second paper adds to this assessment of musical properties by considering listener perspectives, revealing the relative importance of subjective and objective facets in predicting what makes music sleep inducing. The final paper tests the efficacy of music listening at night-time, finding that listening to music led to a significant improvement in sleep behaviours that was not directly associated with music type, nor the participants’ preferred choice of music. Together, these studies provide specific clarifications and recommendations to the field. Our results support previous descriptions of sleep music as acoustic, instrumental, and having little rhythmic activity or dynamic variation, and bring new emphasis to the importance of timbre. Subjectively, a sense of comfort and familiarity were significant factors that were repeatedly revealed by participants’ feedback, however the latter had opposing connotations between individuals. Notions of attention or distraction also appear to be important, but require better contextualisation. Future studies on the use of music to promote sleep should consider the subtle variabilities that can have a significant impact on its efficacy, and probe into other factors that contribute to its effect for an individual. Overall, we have shown that music can help with sleep, and suggest a path to optimising its potential.