fathers,are more expressive with their infant sons than with their infant
daughters (Trehub,Hill,and Kamenetsky 1997b),which is consistent
with claims of greater attachment to same-sex infants (Fagot and Lein-
bach 1987;Morgan,Lyle,and Condran 1988).In short,the infant audi-
ence contributes to the nature of the performance,much like the adult
audience of an oral performance (Rubin 1995).
Mothers’ performances exhibit some fine-tuning to the mood and abil-
ities of their young listeners.For example,the same song (e.g.,“Twinkle,
Twinkle Little Star”) may be sung in a soothing manner on one occasion
and in a playful manner on another (Trainor and Rock 1997;Trehub
et al.1997b).Moreover,mothers subtly alter their performances of the
same song for infant or preschool children (Bergeson and Trehub 1998).
Typically,they sing at a slightly higher pitch level for infants and enun-
ciate the lyrics more clearly for preschoolers.Of interest,their emotive
quality is relatively similar in both contexts,which may reflect compara-
ble affect and nurturant feelings.Vocal differences that are evident may
constitute age-appropriate means of recruiting and maintaining atten-
tion.Despite the subtlety of these performance differences,naive adult
listeners successfully identify infant-directed versions.Infants’ Responsiveness to Infant-Directed Music
Do particular song types and styles of performance make any difference
to the infant audience? On the basis of our pattern perception research
over the past several years,we know that infants can perceive the rele-
vant acoustic distinctions.Thus they are capable,in principle at least,of
exhibiting attentional and affective preferences for different musical
materials,as they do for contrastive speech registers (Werker and
McLeod 1989;Cooper and Aslin 1990;Papousˇek et al.1990;Fernald
1991,1993;Papousˇek,Papousˇek,and Symmes 1991;Pegg,Werker,and
McLeod 1992;Werker,Pegg,and McLeod 1994).As noted, infants
“prefer”consonant harmonizations of melodies to dissonant harmo-
nizations (Zentner and Kagan 1996;Trainor and Heinmiller 1998).More-
over,they show enhanced attention to recordings of a woman singing to
her infant relative to a comparable performance (same song and singer)
with no infant audience (Trainor 1996).Men’s infant-present and infant-
absent songs do not generate different responsiveness except when the
pitch of both versions is artificially raised to the characteristic pitch range
of women (O’Neill 1997).
Different responsiveness was assessed by videotaping infants as they
listened to two contrastive recordings.Adults judged for each infant
which of two soundless video segments suggested greater infant439 Human Processing Predispositions