|
Abstract:
The purpose of this experiment is to examine the interaction
of prosodic cues and pronoun referent assignment during early
stages of processing. In particular, we investigate whether
contrastive stress on a pronoun effects the real-time activation
of the pronoun's referent. Judgments from 128 native English
speakers were gathered on sentences such as the following:
"The cowboy pushed the robber into the chairs by the bar and
the waiter pushed him/HIM into the poker table by the
staircase."
Offline results show that most listeners assign 'robber' to
the pronoun in sentences without contrastive stress, but prefer
'cowboy' when emphasis is placed on the pronoun.
A cross-modal naming task was used to investigate the time
course of referent activation during on-going comprehension of
the same sentences. A subject listened to uninterrupted sentences
over headphones while watching a computer screen. At some point
during each sentence, a string of letters appeared on the screen
and the subject read the word aloud into a microphone. Probe
words appeared either at a baseline position 800 ms before the
pronoun or at the offset of the pronoun. Naming reaction times
for probes semantically related to the second NP were compared to
unrelated control probes matched for a prior reaction time and
frequency. We observed reactivation of the correct pronoun
referent immediately at the pronoun in sentences without
contrastive stress, and also in sentences with contrastive stress
on the pronoun, (the later contrary to the offline preferences).
Our results indicate that a pronoun's referent is accessed
immediately upon hearing the pronoun and that contrastive stress
does not interfere with this early syntactic process.
Further research is being undertaken to determine whether all
structurally legal referents are accessed in these sentence types
and whether prosodic cues affect activation patterns
downstream.
|
|
|
|
Probe Type |
| Sentence Type |
|
Related |
Control |
Difference |
Sentences with
Stressed Pronoun |
|
|
Before Pronoun |
543 |
544 |
d = 1 |
|
|
At Pronoun |
537 |
546 |
d = 9* |
Sentences with
Unstressed Pronoun |
|
|
Before Pronoun |
543 |
544 |
d = 1 |
|
|
At Pronoun |
541 |
549 |
d = 8* |
* p < .05 (paired comparisons)
N = 80
|