Monthly Archives: September 2012

Do You Hear What I Hear?

Note: This post is cross-published on my “Live…In Concert” blog on the Psychology Today website.

In many ways, the label “self-taught musician” is a misnomer. Even those who learn from YouTube videos avail themselves of the musical models of others. Most people who become skilled musicians only do so with the involvement of other people such as teachers, mentors, and musical peers. One of the most important functions they serve is giving feedback. Whether in the formal contexts of school classes or private lessons, or in more informal settings like a friend’s garage or basement, budding performers learn a lot about their musicianship from other people’s evaluation and advice.

At some point in their development, young musicians become less dependent on the appraisals of others and engage more in self-evaluation. Accurately hearing one’s own performance, however, is not always a straightforward process. In fact, self-awareness may not be a strong suit for some performers. I was recently asked about this by someone in audio production and engineering whose experiences had led him to believe that many musicians are poor judges of their own work. He wondered if there’s a psychological phenomenon by which performers mentally replace what is actually sounded with a version from their “mind’s ear.”

I think he’s onto something. In my own research, I’ve had numerous experiences in which musicians I’m working with are unaware of aspects of their own performance. In some ways, this is hardly noteworthy. Many of the expressive features that musicians put in their performances are done fairly automatically, presumably the result of much musical enculturation and practice. In one experiment (Woody, 2003), I asked advanced pianists to play a “deadpan” version of a melody—one with no expressive variations in tempo or loudness—and found that they were consistently unable to do it. But what’s more interesting are times when performers believe they have added certain expressive features in their music, when in fact they have not. It would seem that their musical intentions interfere with their ability to accurately hear their performance.

To understand this phenomenon better, I offer up a model of cognitive skills used in music performance (I believe it also can be applied to other types of skilled performance such as sports, but I’ll stick to music here). When people make music, there are three kinds of cognition going on:

  • Goal imaging is the ability to generate a clear idea of what the music should sound like. Because music deals with sound, this “image” is primarily aural, but it may also include some visual or conceptual aspects (e.g., focusing on a “high point” in a phrase). While most everyone holds goal images of music in memory—it’s how people can decide whether a particular rendition of a familiar song sounds good or bad—skilled musicians create precise images to guide their own performance.
  • Motor production is the ability to carry out the physical movements and responses needed to sing or play a particular instrument. At first thought, this ability might not seem cognitive in nature, but remember that “muscle memory” resides not in the muscles but in the memory (i.e., the mind). Learning a physical skill involves remembering the “feel” of it. New motor skills start off requiring much conscious effort, but with adequate repetition, they can reach a level of automaticity.
  • Self-monitoring is the ability to accurately hear one’s own performance. This is not always easy to do, which is why musicians can be surprised when they hear recordings of themselves (“Did I really sound like that?”). The importance of self-monitoring is seen in the research showing that musicians’ motor learning is significantly impaired when they cannot hear their performance (e.g., Brown & Palmer, 2012).

It is the interaction between these three skills that accounts for improvement made in musicians’ practice sessions. With a clear idea in mind of what they’re trying to sound like (goal image), musicians can compare it to how they do sound (self-monitoring). Identifying discrepancies between the two should guide them in adjusting their technique (motor production).

Especially when performing unfamiliar music, carrying out these cognitive skills can demand all of a musician’s attention. It may, in fact, demand more. Performers can cognitively “max out” just from concentrating on what they’re trying to sound like and on executing the physical skills required. They may not have attentional resources available to accurately hear the results of their efforts (see also Keller, 2001). This is why younger musicians can be so dependent on the feedback of others to know whether they performed something well. They simply cannot encode into memory both the goal and the outcome of performance. More experienced musicians may become so focused on the expressive intentions of their performances that they forget—or simply do not want to be bothered—to listen objectively to the sounds they’re producing. Instead of comparing the products of goal imaging and self-monitoring to guide their performance, they linger over the goal.

This model can be useful to musicians as they attempt to diagnose their performance problems. It can also be useful to teachers—perhaps even music producers—when they are trying to get performers to change something about their playing or singing. The key comes in pinpointing where a breakdown is happening. Does the musician not have a good idea of what he or she is supposed to sound like? If so, goal imaging can be built through additional listening, both to recordings and expert models (certainly school band directors would have an easier time in rehearsals if their students actually listened to band music). Alternatively, some musicians’ practice sessions may be ineffective because they are unaware of what they truly sound like. Perhaps they are practicing music that so challenges them technically that all their cognitive resources are devoted to merely producing the desired sounds on an instrument. In this case, I’d recommend they audio record their practicing and listen carefully upon playback.

By proposing this model, I’m not suggesting that making music should be experienced as an analytical process. Although I do believe these cognitive skills underlie performance, I’d suggest that musicians only try to bring them to the surface while practicing. Successful performance is marked by more fluid or “natural” music making. I suspect that through initial consciousness and the careful repetition of practice, these cognitive skills become automatized in expert musicians. It’s likely they come to use extramusical ideas—emotions, mental imagery, expressive metaphors—to more efficiently encode into memory the sounds and “feel” of performance that they acquire through experience.


Brown, R. M., & Palmer, C. (2012). Auditory-motor learning influences auditory memory for music. Memory & Cognition, 40, 567-578.

Keller, P. E. (2001). Attentional resource allocation in musical ensemble performance. Psychology of Music, 29, 20-38.

Woody, R. H. (2003). Explaining expressive performance: Component cognitive skills in an aural modeling task. Journal of Research in Music Education, 51, 51-63.

Copyright 2012 Robert H. Woody

Source of image: christophheinrich on Flickr Creative Commons.