| |
Abstract:
Previous reports in the literature indicate that the left
hemisphere is involved in category representation, specifically in
hierarchical typicality structures. Roschs prototype theory (1975)
proposed that categories are structured according to a "best"
member, or prototype. Additional members of a category are then
ranked for typicality based on the number of shared features with
the prototypical member.
A large body of work exists investigating how basic categories
(e.g., bird, fruit, etc.) are structured by neurologically-intact
subjects as well as by brain-damaged subjects. However, the
majority of these reports have been based on data collected via
temporally unconstrained (off-line) measures, such as
prototypicality ratings and category naming tasks (Grossman, 1978,
Uyeda & Mandler, 1980).
The present study examines the influence of context on the
nature of category representations during on-line sentence
comprehension in anterior-lesioned and posterior-lesioned aphasics
as well as unimpaired control subjects. An on-line method (Cross
Modal Lexical Priming) was used to measure how the process unfolds
over time. Sentences semantically biased towards the atypical
exemplars were presented aurally to 150 unimpaired control
subjects, 7 anterior-lesioned aphasics, and 3 posterior-lesioned
aphasics (in a within-subjects, counterbalanced design). At one of
three probe positions (the offset of the category name, 450
milliseconds, or 1000 milliseconds downstream in the sentence from
category name offset) a visual probe word related or unrelated to
the basic category term appeared on the computer screen.
The following is an example of the sentential materials and
probes.
The sleepy farmer grabbed a bucket and tossed a handful of grain
to the clucking bird*1 as he hur*2ried out to th*3e barn because he
still needed to feed the livestock.
Typical exemplar/related: ROBIN
Typical exemplar/control: COMET
Atypical exemplar/related: CHICKEN
Atypical exemplar/control: WORSHIP
It was found that the pattern of results for the unimpaired
control subjects resembled the categorical feature pattern found by
Whitney, McKay, Kellas, & Emerson, Jr. (1985) in a modified
Stroop task, with exhaustive semantic priming occurring immediately
after the category name and the continued activation of less
typical (contextually relevant) meanings being maintained via the
biasing context. Results for the anterior-lesioned subjects support
the protracted rise-time of lexical activation hypothesis (Prather,
Shapiro, Zurif, & Swinney, 1991; Prather, Zurif, Love, &
Brownell, 1997), with faster lexical access for only the typical
exemplar immediately upon hearing the category, but activation for
both exemplars later on. Data for the posterior-lesioned subjects
indicated an underlying impairment in the ability to recognize
atypical exemplars, with faster lexical access only for the most
typical exemplars at the point immediately following the offset of
the category name. These findings provide empirical on-line
evidence in support of off-line literature implicating
categorization disruptions for these subjects (Grober, Perecman,
Kellar, & Brown, 1980).
We conclude that the comprehension difficulties demonstrated by
anterior-lesioned and posterior-lesioned subjects implicate
different processing and neural mechanisms. Deficits found in the
anterior aphasics result from damage to an area involved in the
immediate activation of multiple meanings while deficits observed
in the posterior aphasics result from damage to an area involved in
the structuring of semantic field boundaries.
Grober, E., Perecman, E., Kellar, L., & Brown, J. (1980).
Lexical knowledge in anterior and posterior aphasics. Brain and
Language, 10, 318- 330.
Grossman, M. (1978). The game of the name: An examination of
linguistic reference after braindamage. Brain and Language, 6,
112-119.
|
