News Release

Memory for music: Musicians don't have to hear themselves perform after they learn a song

Peer-Reviewed Publication

Ohio State University

COLUMBUS, Ohio -- Musicians who hear the music they are performing while learning a new piece have a better memory for the music later, a new study suggests.

But after they learn a song, actually hearing the music as they play does not improve the accuracy of their performance.

These results shed new light on how memory works and on theories about how people learn, said Caroline Palmer, co-author of the study and professor of psychology at Ohio State University.

Specifically, Palmer said the findings cast doubt on the universality of matching theories – theories that state memory works best when conditions are similar during learning and during recall of the information.

In this research, matching theory would suggest musicians who heard the music during learning would remember it better if they also heard their performance when they replayed it. Those who didn't hear their performance during learning would do better to not hear it when they replayed it later. But that wasn't the case.

"We found that musicians remembered music better when they heard it during learning, regardless of whether they heard the music when they replayed it later," Palmer said, "It wasn't important whether conditions matched during learning and recall."

Palmer conducted the study with Steven Finney, a former postdoctoral fellow at Ohio State now at Spirent Communications. Their results were published in a recent issue of the journal Memory and Cognition.

The researchers conducted three related experiments. In one, 16 adult pianists (with an average of about 10 years of piano instruction) performed a 23-note musical phrase from written notation. They were told they would perform it later without notation to see how well they remembered it. They played on an electronic keyboard and wore headphones. In some conditions they heard the music as they learned it and sometimes they heard nothing. They played the piece 10 times with notation and then played it four times from memory. The keyboards were hooked to a computer that analyzed how accurately they played the piece.

Results showed that during the learning trials, players who did not hear their music performed about as well as those who heard the music.

However 15 of the 16 participants played the piece more accurately during the memory test when sound was present during their learning trial. That was true whether or not they heard music when they were tested.

In other words, Palmer said, it did not matter if conditions matched during the learning phase and the test phase of the experiment.

"What matters is if they heard themselves play during learning," she said. "The pianists were not relying very much on the auditory feedback once they learned the piece, but the sound was very important during learning."

The second experiment was much like the first, except that the researchers made it harder for the participants to memorize the music after they learned it. After learning the music, but before playing it back, the participants had to perform an "interference" task: they either listened to unrelated music, played a separate piece of music with or without sound, or did an arithmetic task (counting backwards aloud).

Results showed the participants still did best when they heard the music while learning it. The music-related interference tasks caused worse recall than did the arithmetic task. "This suggests that the memories are truly musical," she said. "Music interference caused worse recall than arithmetic interference."

A third experiment tested whether hearing the music during performance affected the recall of more complex, well-learned music. In this experiment, 11 skilled adult pianists selected a musical piece they already knew from memory and played it – sometimes hearing it on headphones, and sometimes not.

Once again, Palmer said, error rates were very low and similar when participants heard themselves play and when they did not. "Some theories would suggest that because they are used to always hearing themselves play, they are dependent on this auditory feedback to play," Palmer said.

"But that theory wasn't supported in our study."

But why was auditory feedback – hearing the music – so important during learning? Palmer said there are two likely explanations. One is simply that memory improves when you have multiple sources of sensory information that add to the memory – in this case, adding the sound of the music to the motor movements of the hands over the keys and the visual stimuli of watching your hand movements.

Also, researchers believe that people are much more likely to remembers things that are meaningful. "Music is all about the sound," she said. "Music isn't about the movement of the hands, so the auditory feedback gives the music meaning, which makes it more memorable."

While there are some differences between memory for music and for other tasks – such as speech – this study showed many similarities, Palmer said.

For example, the errors people make during speech don't rely on auditory feedback either, she said. When speakers make errors they tend to be related to what they are thinking about saying, rather than what they are hearing. A speaker, for instance, may say "fork" when she means "spoon," because they are related objects. In this study, Palmer said the subjects also made these kinds of "smart" errors – playing notes that were wrong, but were related to the note they were supposed to play.

"We think there are very general memory and perceptual phenomena that drive how you learn music and how you learn speech," she said.

"Learning music reflects the same memory processes we use in everyday life."

This study was supported by a grant from the National Institutes of Mental Health.

###

Contact: Caroline Palmer, 614-292-7718; Palmer.1@osu.edu or Caroline.Palmer@mcgill.ca

Written by Jeff Grabmeier, 614-292-8457; Grabmeier.1@osu.edu


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.