A series of four experiments examined infants' capacities to detect re
peated words in fluent speech. In Experiment 1, 7 1/2-month old Americ
an infants were familiarized with two different monosyllabic words and
subsequently were presented with passages which either included or di
d not include the familiar target words embedded in sentences. The inf
ants listened significantly longer to the passages containing the fami
liar target;words than to passages containing unfamiliar words. A comp
arable experiment with 6-month-olds provided no indication that infant
s at this age detected the target words in the passages. In Experiment
3, a group of 7 1/2-month-olds was familiarized with two different no
n-word targets which differed in their initial phonetic segment by onl
y one or two phonetic features from words presented in two of the pass
ages. These infants showed no tendency to listen significantly longer
to the passages with the similar sounding words, suggesting that the i
nfants may be matching rather detailed information about the items in
the familiarization period to words in the test passages. Finally, Exp
eriment 4 demonstrated that even when the 7 1/2-month-olds were initia
lly familiarized with target words in sentential contexts rather than
in isolation, they still showed reliable evidence of recognizing these
words during the test phase. Taken together, the results of these stu
dies suggest that some ability to detect words in fluent speech contex
ts is present by 7 1/2 months of age. (C) 1995 Academic Press, Inc.