Thompson & Williamon 2003
The reason I am attempting to research ACT for music is that so many musicians, especially in the classical music world, struggle to perform at their true capacity under pressure. I strongly believe that ACT processes can be helpful, but as I consider designing my own study to show how ACT might be useful, I have turned my mind to ask what do I need to measure, and how do I measure it?
It is sort of obvious that most music performance anxiety (MPA) studies would like to see a reduction in levels of anxiety during performance across time and to this end, these studies measure general anxiety, music performance anxiety and maybe take a measure of self-efficacy or confidence. The radical thing about an ACT interventions however, is that although a reduction in anxiety symptoms is a welcome outcome, ACT does not consider symptom reduction as the goal, but rather encourages moving toward what you care about in the presence of anxiety. Regardless, there are now standard, highly researched and reliable ways of measuring how anxious someone is, so for me tracking levels of anxiety should be a straightforward task.
The more crucial question for me, and I think most musicians, is ‘can ACT improve the musician’s performance?’ Of course, anxiety can be an extremely strong and uncomfortable emotion and who wouldn’t want less of it when you are playing, but the more important phenomenon I want to be able to measure is whether ACT can help you perform better under pressure, and this is where things get tricky. To measure whether ACT can do this, my idea was that would be some sort of reputable, standardised measure that had been 'validated' by the music research community, but after a beginning to look at this, there are a large number of different processes and none of them can be relied upon to 'prove' that one performance is any better than any other and this paper gave me cause for concern when it states:
researchers who wish to employ performance assessment as a dependent measure in experimental studies may have to accept that musical performances are simply not open to reliable and consistent scrutiny of the type they might wish.
This is something I had really should have considered already, as I have sat on audition panels for professional orchestras, and have seen fierce arguments about which candidate performed better in the audition. I have always found it fascinating to see such variations in opinions and always felt that it could be a more empirical and objective process. Little did I know that even respected experts in this field, after years of concerted effort have not produced a reliable way of quantifying differences in different performances, either between two performances by the same person or different people.
Musical performance assessment (MPA) is actually going to be the tricky thing and this paper highlights a few of the issues:
§ it is important that the people doing the listening and the assessment (judges) have had some training in how to do the assessments, and are more consistent if they are subject to regular scrutiny. It is a skill that can be learned and improved with practice
§ there a two types of assessment – holistic and segmented
o holistic is the most reliable and consistent, usually just one mark based on the judges overall impression
o segmented does not seem to be internally consistent even between the same judge, but does give more detailed feedback to performer
o sometimes the mark for overall impression within segmented marking system is not consistent
§ marks are influence by cultural expectations such as performers appearance, race, ethnicity and how the performer follows performance protocols like playing from memory.
§ judges seem to be more reliable if they play the instrument that is being evaluated
Wish me luck.