Last night in the U.S. many televisions were tuned to one of the biggest spectacles of the year: the American Idol finale, where America would learn which singer had been chosen as “America’s favorite” (or, more cynically, who inspired the most teenagers to repeatedly dial toll-free numbers until all hours of the night). Greta and I are suckers for this sort of thing, so we watched along with the rest of the nation.
What impressed me about the show wasn’t so much the prodigious vocal talents of the two finalists, but how everything was put together so hastily: there had been only six days from the previous week’s episode (where the two finalists were revealed), and during this time each finalist learned at least three or four songs. The musicians who played along with them had no score to follow; they had to commit the songs to memory. Everything went off without a hitch, because these professional musicians routinely hold an astonishing variety of music in their memories.

If you’ve ever seen a symphonic concerto, you probably noticed that the soloist usually performs the entire piece — lasting 20 minutes or more — from memory: thousands of notes, all played with perfect pitch and intonation. Clearly many musicians have exceptional memories for the songs they play. So does this ability to remember hundreds of songs transfer into other types of memory? While there’s been some research into musicians’ memory, the results have been mixed. Most studies show that musicians have better memory for words than non-musicians, but there’s less evidence that musicians can remember spatial information better. In one study, musicians couldn’t recall locations on a map any better than non-musicians.
So a team led by Lorna Jakobsen tested 36 college students, 15 of whom had an average of 11.5 years of formal piano instruction and had passed a rigorous performance examination, while the rest had less than a year of musical training.
Two tests were given. In the first, the students listened to a list of 16 words chosen from one of four different semantic categories (e.g. birds, furniture, etc.). The list was presented five times, after each of which they tried to remember as many words as possible. Next they were presented a different list of 16 words as a filler interference task. Then they were asked to try to recall the original 16 words again, and accuracy was tracked. Then everyone took the separate visual test, and after completing that (about 20 minutes later), they were again asked to remember the original 16 words again. How did they do? Here are the results:

The students were actually tested on the words in two different ways: free recall, and cued recall, which was a multiple-choice test. The musicians performed significantly better than non-musicians in every test except the short-delay free recall. So this confirmed the previous research suggesting that musicians are better than non-musicians at remembering words.
The interference task was similar, but it focused on visuals. The students saw 15 simple drawings, presented one at a time. The drawings all consisted of a single line and a dot, arranged in different orientations. As with the word task, the students saw the pictures five times, trying to draw by hand as many of the pictures as they could recall each time. After they had done the word memory test, they returned about 15 minutes later to see how many of the drawings they could remember. The results were similar: musicians remembered significantly more of the drawings than non-musicians.
The researchers also paid careful attention to how the students were recalling both the words and pictures. Musicians, they found, used different strategies. With words, they were more likely than non-musicians to group the words into similar categories (like “eagle” and “robin”). With the pictures, non-musicians tried to verbalize the pictures, while musicians did not.
In both cases, the musicians’ strategies proved to be more effective, and they also seem to duplicate strategies experts use to remember large quantities of information: they chunk it into related groups so that it’s easier to recall later on. When they remember songs, musicians rely on the fact that the notes are related to each other (and in fact are no better than non-musicians at remembering random sequences of notes). They appear to apply a similar strategy to other memory tasks.
Jakobson, L., Lewycky, S., Kilgour, A., & Stoesz, B. (2008). Memory for Verbal and Visual Material in Highly Trained Musicians Music Perception, 26 (1), 41-55 DOI: 10.1525/mp.2008.26.1.41