| |
Abstract:
In sentences like (1), containing a relative clause (RC)
attachment ambiguity, English speakers prefer to attach the RC to
the later-mentioned site
(actress;
Cuetos & Mitchell, 1988).
(1) Someone shot the servant of the actress who was on the
balcony.
However, these preferences are often weak, especially in on-line
studies (e.g., Carreiras & Clifton, 1993), and can be
influenced by properties of the ambiguity such as the type of
preposition connecting the two potential attachment sites (e.g.,
of
in (1); Gilboy et al., 1995). Mitchell (1994) suggests that RC
attachment preferences can be shaped by experience, and individual
differences in working memory capacity have been considered as
differences in experience (Pearlmutter & MacDonald, 1995).
Individual differences in working memory capacity might therefore
provide insight into this ambiguity.
The role of individual differences in working memory as well as
the role of the connecting prepositions were examined in two
experiments. Experiment 1 contrasted two types of
of
(Gilboy et al., 1995): functional or occupational (e.g.,
the translator
of
the ambassador)
and representational (e.g.,
the sketch
of
the house).
Twenty sentences like (1) were created, ten with each type of
of.
Target nouns were either both animate or both inanimate. Each
experimental item was followed by a question about the attachment
such as (2). A questionnaire was composed of these sentences and 60
filler sentences (each followed by a question). Subjects were also
given the Daneman and Carpenter (1980) reading span task.
(2) Who was on the balcony? _________
Low span subjects (span < 4, N = 57) preferred early
attachment (64%), but the high span subjects (span ( 4, N = 12) did
not show a preference (51%). The of types did not differ.
Experiment 2 investigated whether this pattern held across other
prepositions, locatives and
with,
which attach as adjuncts and not as arguments. Ten sentences with
locatives and ten with
with
were created. The
of
sentences from Experiment 1 were also used.
Overall low span subjects preferred early attachment. The
of
and the locative sentences showed similar patterns of preference
to Experiment 1. The
with
sentences differed in that all subjects were much less likely to
attach early (see Table 1).
The difference between the of's and the with's replicates
previous findings, but it cannot be explained by a simple adjunct
vs. argument distinction because the locatives did not display the
same pattern as the with's. A possible explanation for the span
differences is that low span subjects focus on the matrix verb and
its arguments and pay less attention to deeper structure. The data
are discussed with respect to this and other theories of
attachment.
References
Carreiras, M., & Clifton, C. (1993). Relative clause
interpretation preferences in Spanish and English.
Language & Speech,
36, 353-372.
Cuetos, F., & Mitchell, D. C. (1988). Cross-linguistic
differences in parsing: Restrictions on the use of the Late Closure
strategy in Spanish.
Cognition,
30, 73-105.
Daneman, M., & Carpenter, P. A. (1980). Individual differences
in working memory and reading.
Journal of Verbal Learning and Verbal Behavior,
19, 450-466.
Gilboy, E., Sopena, J.-M., Clifton, C., & Frazier, L. (1995).
Argument structure and association preferences in Spanish and
English complex NPs.
Cognition,
54, 131-167.
Mitchell, D. C. (1994). Sentence parsing. In M. A. Gernsbacher
(Ed.),
Handbook of Psycholinguistics,
pp. 375-409. San Diego, CA: Academic Press.
Pearlmutter, N. J., & MacDonald, M. C. (1995). Individual
differences and probabilistic constraints in syntactic ambiguity
resolution. Journal of Memory and Language, 34, 521-542.
|
