| |
Abstract:
Abstract: The current best method for localizing critical
language areas prior to the surgical excision of brain tumors is
intraoperative electrical stimulation mapping (ESM), and the most
widely-used task is picture naming. Pictures are typically
presented without preference for semantic category. However,
category-specific effects of brain lesions, and patterns of brain
activation have been reported. Prior to the operation, all items
are tested to insure that they can be named readily without
electrical stimulation. 96 items were selected from the Snodgrass
and Vanderwart (1980) corpus of line drawings, arranged within 8
categories (wild animals, domestic animals, plants, machines,
static objects, manipulable objects, tools, and musical
intruments). 64 of these items were selected for intraoperativ e
presentation, divided into these 8 categories, and presented
randomly for as many repetitions as needed using Psyscope software.
In a consecutive series of 18 patients, within the cortical fields
mapped, six sites that consistently disrupted naming were
identified in 5 patients. Although disruption at four of these
sites appeared to be non-specific by semantic category, two sites
in the left temporal lobe appeared to be selectively affected by
semantic category: one by inanimate objects, and one by animals.
Methods for obtaining a larger sample of items subdived by category
at essential language sites, and for analyzing less-consistently
disrupted secondary language sites are needed to substantiate and
extend these results.
|
