Hello Dear Reader,
One the biggest conferences in music psychology is going on right now in the beautiful city of San Francisco, ICMPC 14. This meeting always boasts a wealth of fascinating talks and posters from the cutting edge of research in music behaviour, neuroscience, education, development, perception, cognition…and many more!
I am not there. This makes me very sad, Dear Reader. I simply did not have the funds to attend; times are tough in UK research money right now and I had to prioritise student stipends and lab equipment.
Hey ho, there is no use complaining. I am enjoying the social media feeds from the conference (follow #ICMPC14 on Twitter) and I am sure I will spend many enjoyable hours diving around in the conference proceedings. So much to learn!
In the meantime, here I sit in my little office thinking of you, Dear Reader, and what we might chat about today. I decided to talk about music performance – do you think that you could spot a professional performance just from an audio recording?
This question is linked to some of my latest research. Last week I spent time back in my second home in Switzerland at the Hochscule Luzern Musik. As usual I enjoyed the stimulating company (and endless hospitality) of my Swiss colleagues. For the most part we focused on the latest data from our study of classical music critics. My Music and Wellbeing colleague Dr. Elena Alessandri has been running our interview study, asking professional critics about their work and their opinions as to the role of critique in the modern classical market. I have little doubt that these individuals, with the finely tuned responses and many years of experience, would do a good job of spotting a professional performance in a group of recordings. But what about the rest of us?
A new study by Carolyn Kroger and Elizabeth Hellmuth Margulis has the intriguing title “But they told me it was a professional“. Published this year in the journal Psychology of Music, the paper examines whether factors other than the quality of the performance itself might lead people to view it more favourably. These other influences are called the extrinsic factors, and include what you are told about the performance and, interestingly, the order in which you hear a performance (first or second). Do these extrinsic factors really influence our preferences?
Our evaluative judgements are complex and in reality it comes as no surprise that extrinsic context, the world outside the music and its performance, influences what we think is more or less valuable/ good. Our judgements are based on many things including, but not always, our emotional reaction, personality, familiarity with the work, mood, and even the price of the item or ticket.
Performance evaluation is a touchy subject for many. In a world where music exams and competitions mean so much and come down to human judgement, it can be hard to accept that in some studies inter-rater agreement level can reach a staggeringly low .09 to .16 (Fiske, 1978).
Fortunately for music students and artists,this level of variance goes down significantly when the judges are highly trained in the repertoire and instrument/voice being assessed – hence my belief that the music critics from my Swiss study really know their stuff when it comes to piano recordings.
However, there is the fascinating question of what kind of context can really influence the judgements of you and me, the ordinary music lover with no particular training or expertise in regards most of the music we like. Are we at the mercy of what we are told about a performance or the way it is presented to us?
The new study
The new paper examines how listeners with a good amount of general musical training and expertise are impacted by the presentation of extrinsic information before listening to a performance.
In the first experiment people were asked to listen to two performances of the same piece and to rate their enjoyment and the perceived quality of each. Crucially they were asked to select which in the pair was performed by a “world-renowned professional” and which by a “conservatory student of piano”.
In a follow-up experiment the study was repeated except that half the people were intentionally told incorrect information, that the performer was the student and vice versa.
If people are able to distinguish the performances based on the intrinsic qualities of the performance then they should prefer the professional regardless of where the performance appears in the order of presentation (first or second), and regardless of how they are primed (i.e. what they are told about the person).
The stimuli were 90-120 second segments of 8 piano pieces from the common practice period. These included piano pieces by Beethoven (Sonata in E, Op. 109, II), Mozart (Sonata in B Flat, K.570, II) and Brahms (Intermezzo in E, Op. 116, No. 4). The professional performances were recorded by the likes of Alfred Brendel, Vladimir Ashkenazy and Nikolai Demidenko. Student performances were drawn from user-uploaded recordings on the Internet as well as email submissions for the study.
The presentations were different for four groups of people. Some were given a mix of professional and student pieces to listen to (SP, PS). Half were given the same performance twice (SS, PP), but were told they were two different performances. I like to think I might spot a duplication like this, but apparently people don’t!
The ratings from the 40 participants (80 in total, two experiments) were done on 7 point scales. Participants were asked about their enjoyment and performer skill level in relation to tempo, dynamics, and expression. The middle of the scale was always “just right” with the extremes being “much too…” or “far too…” (slow (fast) / soft (loud) / (in)expressive/ (un) conventional).
Overall, people were more likely to say that a piece was performed by a professional when it really was – phew! Some relief for our ears there.
However, there was an additional effect of order. People rated the second performance as the professional on 62.8% of occasions, when random assignment should have led to a 50% rate if there were no bias.
Another matter influenced by order was the expressivity ratings. Expressivity was more likely to be rated higher in the second performance of a pair, including on specific measures of how good were the expressions of tempo and dynamics. Remember – in some cases these ratings were applied to exactly the same piece of music, just heard for the second time.
When people rated a performance as highly enjoyable they were much more likely to rate the performer as skilled. Interestingly however, skill and quality did not differ signficantly as a function of performer type or order.
In Experiment 2 we see the effects of prior information – basically of lying to participants about who is the professional. Overall, people preferred the performance that they were told was performed by the “world-renowned professional” on 65.3% of occasions, 15.3% higher than predicted by chance.
Again, there was an effect of order. When the first piece was primed as “professional” it was preferred on 56.9% of occasions; when the “professional” label was given for the second piece it was preferred 73.8% of the time.
This study shows that intrinsic factors, the professional ability of a performer, do still matter when we listen to a recording. People were able to pick out the professional more than would be expected by chance (Exp.1) and enjoyed their performance on more occasions than those given by the still developing students.
The other side to the coin is that extrinsic factors are not to be treated lightly; they have an effect on our reactions to a music performance. As in many things, we often prefer a performance that is familiar, when we hear it for the second time. Also, we are vulnerable to what we are told – responses can be shifted signifcantly when we are given information about the professional standing of the performer (even when that information is a lie).
We should bear in mind that the pieces in this paper that were played by “students” were all of conservatory quality so would be jolly impressive. In these terms it is actually quite impressive that the average music lover, someone with no particular expertise in the instrument or repetoire, can spot the professionals based soley on the accoustic quality of a recording. However, this research adds to a body of evidence that extrinsic factors like concert notes (Margulis, 2015) and familiarity with a piece (in this case immediate repitition) can shape our opinion of a music performance.