The Cognitive Neuroscience of Music

36     

than to novel passages (as observed in Ref. 17 research on infant linguistic long-term mem-

ory). The test consisted of two familiar and two novel passages, all 20 s in duration. The

familiar test passages were drawn from the middle of each of the two familiarized sonata

movements. The novel test stimuli were drawn from two other Mozart sonatas (taken from

the same CD recording and also composed during late 1774-early 1775: KV 280 in F-major,

Adagio, and KV 283 in G-major, Andante). To ensure that arbitrary properties of the test

stimuli could not account for the infants’ performance, we also tested a control group of

8-month-old infants who received no prior exposure to any of these pieces.

Mean listening times for the familiar and novel musical passages were calculated for each

infant (see Figure 3.1). Consistent with the hypothesis that infants maintain musical experi-

ences in long-term memory, there was a significant difference in listening times for the

familiar and novel passages for infants in the experimental group. However, the direction

of the preference was contrary to our hypothesis; the infants preferred the novel passages

over the familiar passages. To ensure that this pattern of results was not due to biased test

materials, we also examined the performance of the infants in the control group. These

infants did not show a significant preference for either type of test passage, suggesting that

the preferences which emerged for the experimental group reflected long-term memory for

the familiar passages.

Why did the infants prefer the novel passages in the first experiment? This direction of

preference is the opposite of the results from Jusczyk and Hohne’s linguistic study,^17 in

which infants preferred familiar stimuli after a two week delay. There are (at least) two pos-

sible explanations: (1) Mastery: If infants mastered the familiarized music, they might now

Familiar Novel

Listening time (s)

Experimental Control

0

2

4

6

8

10

12

Figure 3.1 Experiment 1: listening times for familiar and novel test passages for infants from the experimental
(prior exposure) and control (no prior exposure) groups.^30