TY - JOUR
T1 - From pauses to clauses
T2 - Prosody facilitates learning of syntactic constituency
AU - Hawthorne, Kara
AU - Gerken, Lou Ann
N1 - Funding Information:
This research was supported by National Science Foundation Grant 0950601 and NICHD #R01 HD042170 to LouAnn Gerken and a National Science Foundation Graduate Research Fellowship to Kara Hawthorne. Thank you to the anonymous reviewers, as well as to Carolyn Quam and Rebecca Gomez for comments on this work. I am also grateful to Sara Knight and other members of the Tweety Lab, as well as the families whose participation made this work possible.
PY - 2014/11
Y1 - 2014/11
N2 - Learning to parse the speech stream into syntactic constituents is a crucial prerequisite to adult-like sentence comprehension, and prosody is one source of information that could be used for this task. To test the role of prosody in facilitating constituent learning, 19-month-olds were familiarized with non-word sentences with 1-clause (ABCDEF) or 2-clause (ABC, DEF) prosody and were then tested on sentences that represent a grammatical (DEF, ABC) or ungrammatical (EFA, BCD) 'movement' of the clauses from the 2-clause familiarization sentences. If infants in the 2-clause group are able to use prosody to group words into cohesive chunks, they should discriminate between grammatical and ungrammatical movements in the test items, even though the test sentences have a new prosodic contour. The 1-clause, control, group should not discriminate. Results support these predictions and suggest that infants treat prosodically-grouped words as more cohesive and constituent-like than words that straddle a prosodic boundary. A follow-up experiment suggests that these results do not merely reflect recognition of words in boundary positions or acoustic similarity of words across the familiarization and test phases.
AB - Learning to parse the speech stream into syntactic constituents is a crucial prerequisite to adult-like sentence comprehension, and prosody is one source of information that could be used for this task. To test the role of prosody in facilitating constituent learning, 19-month-olds were familiarized with non-word sentences with 1-clause (ABCDEF) or 2-clause (ABC, DEF) prosody and were then tested on sentences that represent a grammatical (DEF, ABC) or ungrammatical (EFA, BCD) 'movement' of the clauses from the 2-clause familiarization sentences. If infants in the 2-clause group are able to use prosody to group words into cohesive chunks, they should discriminate between grammatical and ungrammatical movements in the test items, even though the test sentences have a new prosodic contour. The 1-clause, control, group should not discriminate. Results support these predictions and suggest that infants treat prosodically-grouped words as more cohesive and constituent-like than words that straddle a prosodic boundary. A follow-up experiment suggests that these results do not merely reflect recognition of words in boundary positions or acoustic similarity of words across the familiarization and test phases.
KW - Constituency
KW - Language acquisition
KW - Prosodic bootstrapping
KW - Prosody
KW - Syntax acquisition
UR - http://www.scopus.com/inward/record.url?scp=84906335021&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84906335021&partnerID=8YFLogxK
U2 - 10.1016/j.cognition.2014.07.013
DO - 10.1016/j.cognition.2014.07.013
M3 - Article
C2 - 25151251
AN - SCOPUS:84906335021
SN - 0010-0277
VL - 133
SP - 420
EP - 428
JO - Cognition
JF - Cognition
IS - 2
ER -