Sharing my research with the world

Today I’m writing from grey and wet London (where this summer’s day is strangely like the grey and wet Sydney I left 48 hours ago). I’ve learnt a lot in the last few days about both travelling and Londoner’s (eg. Although they drive on the left hand side of the road, when walking on footpaths, up stairs etc, they keep to the right hand side unless overtaking – learnt this the hard way by almost causing a traffic jam up the escalator at a tube station with my suitcase) but perhaps I’ll keep these lessons for a post on academic travelling…

With all this extra time and reduced responsibilities (I only have me to look after myself – what a luxury!) it seemed a good a time as any to write a blog about my research, and what I’ll be doing on this trip as I’ve tried to cram in as much as I can into these 3 weeks.

Main Question of my PhD

My PhD is seeking to identify what is going on in the brain, in terms of brain oscillations, whilst people are imagining music. To do this we’ve first had to design a task in which people were actually imagining what we wanted them to imagine (ie. it’s not good enough to say “Imagine this note” and then scan their brains – how do you know they aren’t really just thinking about what’s for lunch?)

Inspired by the early visual imagery experiments by Stephen Kosslyn which required 3 dimensional shapes to be mentally rotated, we set out to design a task that would require manipulation of sound images, that had objective measures of performance (accuracy and reaction time). We questioned the participants on the strategy they used to complete the task (rather just assuming they used musical imagery), and we made it adaptive so that it got progressively harder. Designing a task that isn’t too easy or too hard provides the flexibility to test both musicians and non-musicians.

Pitch Imagery Arrow Task

So we came up with a design:


It was a staircase design so with correct responses it got progressively more difficult. This example is for Level 1. The Level number corresponded to the number of imagined tones per trial. Within each level there were 3 stages:

Stage Major Key Signature Number of tones / arrows in set-up Starting Tone
1 C 3 – 4 Tonic
2 C, C#, D, Eb, E 3 – 5 Tonic
3 C, C#, D, Eb, E 3 – 6 Dominant

We included 2 control conditions: Perception (where trials identical but no continuation component so matching probe against last heard arrow) and Mental Arithmetic. We tested 24 Musicians and 16 Non-musicians and we wanted to answer two questions in this study:

  1. Does this task induce Musical Imagery?
  2. What role does musical experience, auditory vividness and mental Control play in musical imagery performance? We measured Vividness and Control from a questionnaire that participants also completed: the Bucknell Auditory Imagery Scale (BAIS; Halpern, 2015)

To see three trials (Musical Imagery, Listen and Mental Arithmetic) in random order all based on Level 3 click here.


So what did we find?

Q1: Does the Pitch Imagery Arrow Task induce Musical Imagery?

Yes – at the higher levels.

Here is a summary of the main results. Larger the dot the more musical experience the participant had (percentage of life years participating in learning / playing).


  • Using a Musical Imagery Strategy (that is, “I imagined the notes in my head”, “I sang it in my head”) outperformed those who used alternative strategies (eg. “counted arrows”, “intuition”, “imagined visual staircase/steps”).
  • Using an alternative strategy was not effective in successfully completing trials with 4 or 5 tones to be imagined. Hence to successfully complete the PIAT at higher levels, you need to use Musical Imagery
  • Some non-musicians outperformed musicians – particularly if the non-musicians self-reported higher vividness and control scores.

Q2: What role does musical experience, auditory vividness and mental Control play in musical imagery performance?


Regression Model for Maximum Level Reached on the PIAT, as a measure of performance accuracy

  • Vividness and Control were more important than Musical experience, but using a Musical Imagery Strategy was the most important predictor of performance on the task.
  • Musicians performed better at the task, but only because they were more likely to use a Musical Imagery strategy. When strategy was controlled for there was no difference in performance (eg. Someone with no musical training outperformed a professional jazz musician)

If you’d like to read about this study in more detail, see our paper.

KIT-MEG System at Macquarie University

We’ve since conducted a follow up study which has involved 19 participants (who all scored about 70% on Level 3 of the PIAT, and reported using a Musical Imagery Strategy) completing an adapted version of the PIAT whilst inside the magnetoencephalogram (MEG).

I’ve previously presented some of these MEG results in a poster here looking at differences occurring during the retention period (1 sec before the probe) in the Imagery trials (where the tone you retain was self-generated) compared with Listen trials (where the tone you retain was just heard).

However more recent analysis has looked at other aspects of the data and I’ll be sharing these results of this at a number of places over the coming weeks:

So just tweet me (@RebeccaGelding) if you are around London, Munster, or attending either of these conferences and want to discuss! I want to make the most of my time away.


One thought on “Sharing my research with the world

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s