Hello Dear Reader,
I have been super busy of late – the world of earworms keeps me occupied but I have also been refreshing my knowledge of music memory research for a number of projects including a chapter in my book, a TEDMED talk that I will give in April, and a chapter for another volume that I am writing with one of the lovely PhD students in our lab, Georgina.
Thanks to my music memory research alerts I came across a recent paper published in PLoS ONE regarding musical imagery. Imagery and memory are very closely related in psychology research terms, since imagery is essential for many memory processes.
Research into musical imagery is relatively scarce – this situation can mostly be blamed on the difficulty of exploring imagery itself. How can we get an accurate idea of what is going on inside someone’s head?
The answer to this question has often been to rely on indirect indications of mental activity. So, for example, a priming task.
One study that used priming asked people to imagine tones at a soft or loud volume and then presented them with a second real tone that either matched the imagined tone for loudness or not. If people could imagine loudness then their responses to the second real tone should be faster if the imagined tone, the prime, matched the loudness of the real tone.
This early study found no evidence that people’s responses were primed by imagery, suggesting that we may not reliably be able to represent loudness in imagery. But a valid criticism of the study was that it used only single tones – maybe imagining loudness on a single sound is an odd thing to do? Maybe we are better at imaging how loudness changes, dynamically, over real pieces of music.
This hypothesis was the basis of the new paper by Laura Bishop and colleagues, and published a few weeks ago in PLoS ONE (hurray for free access!!) – click on this link to see the paper
The authors employed a different method to measure imagery, one previously reported by Professor Andrea Halpern. In the “continuous response method” participants imagine music and at the same time give dynamic, second by second feedback on how their imagery is changing, using a manual response known as a ‘slider’.
In Professor Halpern’s study people used a slider to track the emotions in imagined music. Halpern then compared the imagined profiles people produced to profiles that they created when listening to the same music for real. She found remarkable consistency between the two.
This evidence supports the idea that people are able to imagine music well enough to sense changing emotions, and that what they imagine in this sense is pretty similar to hearing real music.
In the new study, Bishop and her colleagues used the slider technique to indicate changes in loudness in both imagined and real music. They tested nonmusicians, novices and expert musicians. The researchers reported that it was difficult to find pieces that people knew well enough and in the end they only had reliable data for two that they trialled; Habanera (Bizet) and Blue Danube (Strauss).
Each participant listened to the two tracks twice a day, every day, for a week in advance of the study. When they came to the lab they were asked to first imagine the piece from start to finish and to use the manual slider to indicate when changes occurred in the loudness of their image.
If the imagery became louder then they moved the slider away from them and if it became quieter then they edged it closer to themselves. They then lapped loudness while listening to each piece.
RESULTS: The authors report a high degree of similarity in the loudness profiles people created when imagining music as compared to profiles that they created when they listened to the same music for real.
The musical experts, and to a lesser extent the novices, produced more accurately matched profiles (imagined and listening) compared to non-musicians, a finding which indicates that the ability to imagine changes in loudness may improve with increasing musical expertise.
One point to note is that there was no independent measure of music listening habits taken in the present experiment (this factor was included alongside the one total measure of musical training and other formal music skills) so we do not yet understand clearly how everyday musical behaviours impact on imagery of this kind.
The task of imagining these two classical pieces was very hard for people. The authors had to exclude 57% of the data from the Blue Danube trials and 53% for Habanera. No doubt imagining music over several minutes is very tricky and perhaps in the future studies could benefit from using shorter pieces – or people’s favourite pieces of music, thereby reducing the impact on memory demands.
The present study does, however, neatly demonstrate how the slider technique can be used to track the dynamic characteristics of musical imagery and, in this case, shows that aspects of changing loudness are likely to be maintained when we imagine music in our minds.
Paper: Bishop L, Bailes F, & Dean RT (2013) Musical Expertise and the Ability to Imagine Loudness. PLoS ONE 8(2): e56052. doi:10.1371/journal.pone.0056052