Auditory imagery underlies many musical tasks, including good pitch imitation, and can be measured by self-report and in objective tasks. Pfordresher and Halpern (2013) showed that poor pitch imitators reported lower auditory imagery vividness on a self-report scale. A subsequent study (Greenspon, Pfordresher & Halpern, 2017) found that the poor pitch imitators showed degraded ability to form auditory images of novel short tunes that they were then told to manipulate. This perceptual-motor mapping was further explored in two studies.
One study extended the auditory imagery task to familiar melodies. As the task was difficult with novel tunes, we asked if having a prior memory representation would increase performance, particularly in poor imitators. A second study used EMG recorded from the face and neck to see if better pitch imitators differed from poor imitators in activation of the motor system when imagining tunes, given that (sub)cortical motor system activation is sometimes seen in the brain during imagery tasks.
Familiar/Unfamiliar Tune experiment: participants compared two melodies (Familiar or matched Unfamiliar melodies) in a same/different (one changed note) task: the target was either repeated or was presented in reverse order, transposed, or shifted serially. Pitch imitation accuracy, imagery vividness and control, and forward and backward digit span were predictors.
EMG experiment: we measured laryngeal and orofacial muscle activity during preparation for singing and also a control task of visual imagery. In the auditory task, participants listened to a four-note melody, imagined the melody, and then sang it aloud. In the visual task, participants studied an array of objects, imagined the array, and then were probed about a single object’s location.
In the Familiar/Unfamiliar Tune experiment, using familiar tunes did improve performance. Singing accuracy and pitch discrimination predicted performance in all conditions. Only self-reported vividness and control of imagery predicted performance for repetitions of familiar tunes. Only STM (digits forward) predicted recognition of unfamiliar tunes in all conditions.
In the EMG study, when controlling for baseline activity, subtle motor activity near the larynx and lip was larger in auditory compared to visual imagery. Worse pitch imitators showed larger laryngeal responses than better pitch imitators. Imagery vividness, assessed on each trial, correlated positively with both imagery tasks.
These experiments extend our understanding of perceptual-motor mapping in auditory imagery. The first experiment suggests a dissociation. Whereas imagery formation may be the critical stage in representing familiar tunes, unfamiliar tunes may need to draw on the “computing power” of STM as is needed in imagery manipulation. The second study links auditory imagery to supportive motor behavior, as the increased activity in poorer pitch imitators implies an attempt at compensation. Auditory imagery is clearly not one entity but has many components, which selectively support musically relevant behaviors.
Greenspon, E. B., Pfordresher, P. Q., & Halpern, A. R. (2017). Mental transformations of melodies. Music Perception, 34, 585- 604.
Pfordresher, P. Q. & Halpern, A. R. (2013). Auditory imagery and the poor-pitch singer. Psychonomic Bulletin and Review, 20, 747-753.