The emergence of tonal scale relationships among the child’s cognitive struc-
tures has implications for the conduct of research. Using atonal materials with
infants has little impact on the results ,because babies do not respond to tonal
scale structures as such (Trainor & Trehub ,1992). But Wohlwill’s (1971) use of
atonal (and to the adult ear rather strange sounding) melodies probably led to
his result that first graders could distinguish targets from different-contour lures
at a level barely better than chance. At any rate ,Wohlwill’s conclusion that ‘‘the
establishment of pitch as a directional dimension is a relatively late phenome-
non’’couldnotbetrueinthelightofThorpe’sresultwithinfants(1986,citedin
Trehub ,1987). What is true is that first graders have trouble using words to
describe pitch direction (Hair ,1977; Zimmerman & Sechrest ,1970).
During later childhood ,the child continues to develop sophistication in the
use of the tonal scale framework determined by the culture. This progress is
illustrated by Zenatti (1969) ,who studied memory for sequences of three ,four ,
and six notes with subjects from age 5 years up. On each trial ,a standard mel-
ody was followed by a comparison melody in which one note of the standard
had been changed by 1 or 2 semitones. The subject had to say which of the
notes had been changed—a very difficult task. Zenatti found that for the three-
note sequences ,5-year-olds performed at about chance with both tonal and
atonal stimuli. From ages 6 through 10 ,the results for tonal and atonal sequences
diverged ,with better performance on tonal sequences. Then ,at around age 12 ,
processing of the atonal sequences caught up. For four- and six-note sequences,
the same pattern appeared ,but the tonal-atonal difference remained until adult-
hood. Experience with the tonal scale system leads people to improve on rec-
ognition of tonal melodies but not atonal melodies. With simple stimuli such
as the three-note melodies ,atonal performance catches up relatively soon ,but
longer sequences continue to benefit from the tonal framework throughout
childhood. (This result converges with that of Morrongiello & Roes ,1990.)
Superiority of recognition with tonal materials has been often observed with
adults (Dowling ,1978; France`s ,1958/1988); Zenatti’s study shows that the ef-
fect can be used as an index of the child’s acquisition of the scale structures of
the culture.
Trainor and Trehub (1994) took the development of the role of tonality in the
ability to detect melodic pitch changes one step further. In addition to alter-
ations that either remained within key or departed from the key ,Trainor and
Trehub introduced changes that remained in the key but departed from the
particular harmony implied by the melody. For example ,the first four notes of
‘‘Twinkle ,Twinkle’’ (Figure la: C-C-G-G) imply harmonization with the tonic
triad (C-E-G). A change of the third note from G to E would remain within both
the key and the implied harmony. A change to F would remain within the key,
but violate the harmony. Trainor and Trehub found that 7-year-olds ,like
adults ,could detect the out-of-key and out-of-harmony changes much more
easily than the within-harmony changes ,whereas 5-year-olds reliably detected
only the out-of-key changes. As Trainor and Trehub (1994 ,p. 131) conclude ,
‘‘5-year-olds have implicit knowledge of key membership but not of implied
harmony ,whereas 7-year-olds ,like adults ,have implicit knowledge of both
aspects of musical structure.’’ In a result that converges with these studies,
Imberty (1969 ,chapter 4) found that 7-year-olds could tell when a melody had
492 W. Jay Dowling