result,infants can detect contour changes even when comparison pat-
terns are presented at different pitch levels than the original (Trehub,
Thorpe,and Morrongiello 1987),as shown in figure 23.2.In other words,
the pitch contour of a melody seems to be central to its identity.Rhythm
also makes important contributions to the identity of a pattern.For
example,infants consider faster or slower versions of a tone sequence as
functionally equivalent, provided the rhythm or temporal pattern
remains unchanged (Trehub and Thorpe 1989).Comparable perceptual
compensations for differences in pitch level (Kuhl 1979) and speaking
rate (Eimas and Miller 1980;Miller and Eimas 1983) are evident in
infants’ and adults’ perception of speech.
It is also clear that infants group or chunk components of tone
sequences on the basis of similar pitch,timbre,or loudness (Demany
1982;Thorpe et al.1988;Thorpe and Trehub 1989;Trehub,Endman,
and Thorpe 1990) in much the same way as adults (see Bregman 1990).
For example,they detect a pause inserted withina group of similar tones
(e.g.,XXXO OO) more readily than a comparable pause inserted
betweengroups of tones (e.g.,XXX OOO;Thorpe et al.1988;Thorpe and
Trehub 1989).Within-group pauses disrupt the perceptual organization
of the pattern,but between-group pauses do not. Similarly,pauses
inserted within musical phrases disrupt infants’ attention whereas com-
parable pauses between phrases do not (Krumhansl and Jusczyk 1990;430 Sandra Trehub
Figure 23.2
Sample stimuli from Trehub,Thorpe,and Morrongiello (1987).Note that successive pat-
terns are presented at different pitch levels.Repetitions in example (a) are exact transpo-
sitions;those in example (b) alter the component intervals but maintain the contour.(From
Trehub et al.1997.)Fig.23.2