| |
Abstract:
Kelly Nafie, University of California, San Diego
Recent research investigating the time course of processing
capacities and their neuralcorrelates have revealed striking
dissociations between overt behavioral deficits and unconscious
processing capacities in brain damaged populations (Swinney et
al., 1996; Zurif et al., 1994, among others). The current work
examines both the real-time processing capacities and final
conscious understanding of complex sentence constructions
containing explicit pro-forms (pronouns or reflexives) by Broca's
and Wernicke's aphasics, Right hemisphere damaged (RHD) subjects,
and unimpaired control subjects.
In the on-line portion of this study, subjects heard (in a
within-Ss, counterbalanced design) modified sentences taken from
Nicol (1988) such as:
The doctor i said that the skier j in the hospital had blamed
him i / himself j *1 for the recent injury.
These sentences contain an unambiguous link between the
pronoun (him) and the antecedent (doctor) and the reflexive
(himself) and its antecedent (skier). In many prior on-line
experiments employing the cross modal priming paradigm, this
design has been used successfully in demonstrating re-activation
of antecedents when the pro-form is encountered (see, e.g.,
Nicol, 1988; Love and Swinney, 1996; Swinney et. al., 1996). For
simplicity, we probed for re-activation at the offset of the
pro-form (*1) and examined for the critical 2nd NP pattern of
activation.
Our findings reveal that like unimpaired individuals,
Wernicke's and RHD patients demonstrate re-activation of the
structurally correct antecedent (skier) to the reflexive, and,
importantly, no reactivation of this same (but here structurally
incorrect) antecedent when the pro-form was a pronoun. This was
not the case for Broca's aphasics who showed an aberrant priming
pattern: facilitation for skier after processing the pronoun.
Here the normal structurally guided co-referential reflex is
disrupted, suggesting that the neural region typically defined as
Broca's area plays a critical role in the automatic syntactic
reflexes underlying auditory sentence comprehension.
The off-line portion of this study tested the subjects'
conscious understanding of these same materials. After hearing
these sentences, subjects answered basic comprehension questions
which targeted the understanding of the ronouns and reflexives in
the sentence. Results demonstrated that both Broca's and
Wernicke's aphasics have difficulty with the overt comprehension
of both types of explicit pro-forms whereas the RHD population
demonstrated near perfect comprehension. Together, the findings
of both studies suggest that the comprehension difficulties
encountered in Broca's aphasia result from damage to an area
involved in the automatic grammatical processes critical in
comprehending grammatically complex sentences while comprehension
problems in Wernicke's aphasia results from a disruption in a
later stage involved in the final interpretation of linked
relationships.
Love, T., and Swinney, D. (1996). Coreference Processing and
Levels of Analysis in Object-Relative Constructions;
Demonstration of Antecedent Reactivation with the Cross-Modal
Priming Paradigm.
Journal of Psycholinguistic Research,
25:1, 5-24.
Nicol,J. (1988). Coreference processing during sentence
comprehension. Doctoral Dissertation, MIT.
Swinney, D., Zurif, E., Prather, P. and Love, T. (1996).
Neurological distribution of processing operations underlying
language comprehension.
Journal of Cognitive Neuroscience,
8:2, 174-184.
Zurif, E., Swinney, D., Prather, P. and Love, T. (1994).
Functional localization in the brain with respect to syntactic
processing.
Journal of Psycholinguistic Research,
23, 487-497.
|
