top of page


The perception, comprehension, and memory for music suggest some intriguing questions. What does an expert musician “hear” when he/she listens to a series of tones in Classical Western tonal music?  Given that expert musicians must often draw on hundreds of skills to play even a single note, what have expert musicians learned that built this large memory bank of general musical knowledge that is so easily accessible to them?

All humans have the psychological resources to represent such structures.  Let us examine language.  Take a quick look at Example 1 and then cover it with your hand.


Example 1.




Can you spell that word in your mind correctly?  Now try the same process with Example 2.


Example 2.



VERNALT-- Have you ever seen this word?  The obvious answer is no.  Vernalt is not a word.   But because you are expert spellers, you were probably able to spell Vernalt correctly, as it is a pseudo word that has the same frequency distribution of letters as English words. It is this abstract coding system you possess for English in your brain that enables you give structure to the word, Vernalt, but not the other word.


Most musical material has as much patterning and structure as language. The expert can make use of this patterning when building up a mental representation of a piece of music.  In order to investigate the question concerning how music is specifically organized and represented by expert and novice musicians, I performed an experimental study with the viola sections of the Chicago Symphony and a Michigan high school (Saline High School).


First, I constructed a test to assess the subject's ability to replicate music patterns in a listen-imitate format. The music patterns I used for my test ranged from random music approximations (see Example 3, a sample of the music patterns used in Subtest 1: zero-order) to music patterns generated from systematically increased degrees of contextual constraint simulating classical Western tonal music (see Example 4, a sample of the music patterns used in Subtest 2:  2nd-order and Example 5 for Subtest 3: 3rd-order).  The subject was asked to listen to each motive, given the starting note and asked to play back the motive by ear on his/her instrument.


The 8-note patterns were generated by a computer using a program called Travesty (Kenner and O’Rourke,1984) that I adapted for music. The computer program memorized the statistical probabilities between all the melodic notes in Teleman's Concerto in G Major For Two Violins.  The results are detailed on the bar chart in Example 6, which suggest obvious differences between groups on the Subtests 2 and 3 were significant, favoring the expert.  However, when the melodic patterns were randomly generated there were no longer any differences between the novice and expert subjects' performance. Both the expert violists and the novice violists performed equally and poorly.  This suggests the experts did not have a particularly better aural memory (even though many were “world class” musicians).  The experts were superior only when the music material to be remembered made some sense and could be grouped into strategic chunks.



Example 6 - Between-groups comparison of mean Percentage scores for Subtests 1, 2, and 3

Perception, Comprehension, and Memory for Music 

It appears that the novice group had been learning music more by rote, with the separate notes floating around in their brains without relating to any other information. The experts had been learning a general coding system based on the probabilities that characterize Western tonal music. Simply stated, we can spend hours listening or performing music but the potential experts will also be unconsciously processing musical information in a rich assortment of indexed categories.


This investigation has provided evidence for the view that the expert musician is engaged in building a mental representation of the music he/she hears, and that when the listener performs music he/she is translating his mental representation of the music patterns into action.  It is proposed these mental representations of music patterns are coded and retained as chunks, along with their memory banks of associated pitches relating to Classical Western tonal music.


High school and university music programs that are committed to the preparation of students for instrumental performance and teaching careers are often dedicated to training students in the vast array of specific expert musical skills required of conductors in today’s orchestras (i.e., skills in rhythm reading, intonation, spiccato . . .).  The expert musician, however, is more than someone who simply possesses these separate skills.  In my investigation, only the musician/subject’s response to aural melodic patterns was observed.  What appeared to distinguish the experts from the novices was the extent and availability of their memory banks of general musical knowledge as they attempted to replicate the aural melodic patterns on their major performing instruments (viola).  The results of the investigation suggest that musicians achieve their expertise through using skills in chunking, motor programming (knowledge of the fingerboard), and monitoring the melodic patterns simultaneously, while using their performance of music to build a memory bank of general musical knowledge.





Kenner, Hugh and Joseph O'Rourke. “A Travesty Generator for Microes,”   BYTE, 11   (1984):  449-469.


Knecht, Melissa G. “Music Expertise and Memory:  The Relationship Between Music  Expertise and Memory of Music Patterns, Within Various Degrees of Contextual Constraint.” Music Education Research 5:3 (November 2003): 227-242.


Larkin, Jill,  John McDermott,  Dorothea Simon and Herbert A. Simon.  “Expert and Novice Performance in Solving Physics Problems.”  Science 208 (20 June 1980): 1335-42.


Miller, George A.,  Jerome S. Bruner and Leo Postman.  “Familiarity of Letter Sequences and Tachistosopic Identification.”  The Journal of General Psychology 50 (1954): 134-138.


Miller, George A. “The Magical Number Seven Plus or Minus Two:  Some Limitations on Our Capacity for Processing Information.”  Psychological Review 63 (1956): 81-96.


Slobado, John A. “The Effect of Item Position on the Likelihood of Identification by Inference in Prose Reading and Music Reading.”  Canadian Journal of Psychology 30 (1976): 228-236.


Slobado, John A.  The Musical Mind:  The Cognitive Psychology of Music. Oxford:  Claredon Press, 1990.




bottom of page