Melodic contour identification by cochlear implant listeners.
Level 4 - case-series / case-control
Small pre-post interventional case series without a control group.
PubMed 17485980 · doi:10.1097/01.aud.0000261689.35445.20
What was done
Eleven cochlear implant users completed a closed-set melodic contour identification (MCI) task testing 135 musical contours (nine 5-note patterns across 1- to 5-semitone intervals and root notes A3, A4, and A5). Familiar melody identification (FMI) was evaluated across 12 melodies with and without rhythm cues, and speech/phoneme recognition was tested. Six subjects subsequently completed custom software-based MCI auditory training in a different frequency range to assess performance gains and generalization.
What was found
MCI recognition performance ranged from 14% to 91% correct across subjects, improving as semitone spacing increased and performing slightly worse for root note A3. Mean FMI performance was 58% correct with rhythm cues and 29% correct without rhythm cues. MCI performance correlated significantly with vowel recognition but not with FMI performance; FMI did not correlate with phoneme recognition. MCI training improved MCI performance in all 6 trained subjects and generalized to improved FMI performance (exact post-training scores were not reported in the abstract).
Why it matters
This study shows that closed-set melodic contour identification effectively measures music perception deficits in cochlear implant users and that targeted pitch training can transfer to familiar melody recognition.
Limits
The study is limited by a very small sample size (n=11, with only n=6 trained), lack of an untrained randomized control group to rule out test-retest familiarity, and omitted numeric post-training scores in the abstract.
Cited by
- supports On average, individuals with cochlear implants can perceive rhythm but cannot appreciate tonality or music, though musically trained patients tend to perform better.