Recognition of transposed melodies: a key-distance effect in developmental perspective

Four experiments examined the possibility of a key-distance effect in a transposition detection task. Subjects heard standard melodies followed by comparison melodies presented in the same key, a musically near key or a musically far key. The task was to recognize comparisons that were exact transpositions of the standards, rejecting nontranspositions. Results suggested a largely invariant key-distance effect with nontransposition comparisons (lures); same- and near-key lures evoked more false alarms than far-key lures. The variables of musical experience, age of subject, and familiarity of melody affected the level of transposition-recognition performance but did not consistently affect the size of the key-distance effect. The results support the psychological reality of key distance and are consistent with both musical and nonmusical-auditory theories of its effects. The key-distance effect was not found with transposition comparisons (targets), a result with implications for the separability of key and interval information in short-term memory for melodies.