EFFECTS OF TIMBRE DISTORTION AND AROUSAL ON MEMORY FOR VOCAL AND INSTRUMENTAL MELODIES

Background: Previous studies have shown that melodies presented in vocal timbre induce higher levels of arousal and better recognition performance than those presented in instrumental timbre. For instrumental memory representations, the surface feature of timbre has also been shown to be more integrated than the surface features of tempo or key. The purpose of this study was to examine whether the memorial representation of vocal melodies differs from that of instrumental melodies in surface feature binding of timbre, and to assess the degree to which individuals’ arousal and orienting efficiency affect memory recognition performance for the instrumental and vocal melodies. Methods: 24 healthy and young adults listened to and categorized 24 melodies (12 piano and 12 vocal timbres) as either “happy” or “sad”. The melodies were unfamiliar British and Irish folk melodies that varied in mode, meter and key. During a 10-minute delay, participants completed an attentional orienting task that provided measures of spatial orienting and phasic arousal. Participants then completed a memory recognition task in which they rated 48 melodies (24 new, 24 old) as “old” or “new” on a 7-point confidence scale. Half of the “old” melodies were distorted by timbre and half were left unchanged. Participants were told to indicate “old” if the melody was the same, regardless of any timbre change. Results: Melodies in major keys and vocal timbres were categorized as significantly more “happy” than those in minor keys and instrumental timbres. This difference between major and minor keys was greater for piano timbre. Overall, participants made significantly more “new” judgments to new melodies than to old melodies (distorted or unchanged). However, sensitivity to timbre distortions differed across vocal and piano melodies. There were significantly more “new” judgments for distorted than unchanged piano melodies, but no difference for vocal melodies. Additionally, phasic arousal (but not spatial orienting efficiency) levels were correlated to level of memory recognition performance for vocal melodies, but not piano melodies. Conclusions: These results indicate that the surface feature of timbre is more integrated into memorial representations for instrumental melodies than for vocal melodies, and arousal levels may modulate memory recognition for vocal, but not instrumental melodies. Taken together, these findings suggest distinctive auditory processing mechanisms for vocal and instrumental music, which may ultimately have implications for clinical music therapy practices.