MEMORY FOR SENTENCES IN BRAIN-DAMAGED ADULTS.
Item
-
Title
-
MEMORY FOR SENTENCES IN BRAIN-DAMAGED ADULTS.
-
Identifier
-
AAI8119660
-
identifier
-
8119660
-
Creator
-
IPPOLITO, PAULA MARIA.
-
Contributor
-
LOUIS GERSTMAN
-
Date
-
1981
-
Language
-
English
-
Publisher
-
City University of New York.
-
Subject
-
Psychology, Experimental
-
Abstract
-
In order to distinguish general memory functioning from linguistic memory functions in aphasics, the Peterson and Peterson (1959) technique for preventing rehearsal was superimposed on the Shewan and Canter (1971) design.;Forty seven adults within 5 groups participated in the study: 10 normals hospitalized for non neurological conditions, 10 right brian-damaged (RBD), 5 left brain-damaged without aphasia (LBD-NA), 14 fluent aphasics and 8 non-fluent aphasics.;The stimuli consisted of 96 sentences each indexed to a four-fold picture selection task. A quarter of the sentences were responded to immediately after being read (0 delay), the remainder after one of three delay intervals (6, 12, 18 seconds) during which the subject continuously responded to the Peabody Picture Vocabulary Test. Half of the sentences contained 3 critical items and 7 syllables (L(,1)), the other half having 5 critical items and 11 syllables (L(,2)). Half the critical items were obtained from the thousand most frequent Thorndike-Lorge (1944) words (V(,1)), the other half from the 25-40 words per million range (V(,2)). Both of these linguistic factors were completely crossed with six exemplars in each cell.;Following administration it was found that the six items in any cell varied enormously in difficulty as a consequence of variations in picture salience, i.e., the prominence of the discriminandum required to make a correct choice. Reliability analyses identified the four most consistant exemplars in each cell, which thereupon became the final data base.;Each subject's scores were next examined at each delay interval to determine where performance dropped to chance, defined as the point where two of the four cells (L(,1)V(,1), L(,1)V(,2), L(,2)V(,1), L(,2)V(,2)) had less than two correct choices. By this criterion 4 RBD, 2 LBD-NA, 6 non-fluents and all 14 fluents failed to complete the experiment. A total score was posted for each S up to the point of departure into chance.;Group contrasts established that the performances of both non-aphasic groups were indistinguishable, as was also the case for both aphasic groups. In final contrasts it was found that the aphasic performances were significantly inferior to the non-aphasic performances which were in turn significantly inferior to the normals, who made essentially no errors in the whole experiment.;Analyses of variance of performances at zero delay established that the fluent aphasics were influenced by the vocabulary variable but not the length variable, and only then at the shorter sentence length. Conversely, the non-fluent aphasics were influenced by the length variable but not the vocabulary variable. These disparities confirm the prior findings of Caramazza and Berndt (1978).;A score was computed for each fluent aphasic expressing his personal responsiveness to the vocabulary variable, which was indeed found to correlate highly negatively (-.741) with total performance, indicating that the worse one's non-linguistic memory the worse one's performance with low frequency words. When an influence of length effect was computed for the non-fluent aphasics it was found not to correlate with total performance. In both groups, not surprisingly, total score correlated positively with the Peabody task on which the experiment was modeled, but paradoxically, for the non-fluents total score was related to aphasia test performance while for the fluents it was not.;It was concluded that all the aphasics had a general memory deficit more severe than the deficits of the non-aphasic brain-damaged. In the fluents this deficit was correlated with a linguistic factor in the task but not with those in an aphasia test. In the non-fluents the correlation was with the aphasia test but not with the task.
-
Type
-
dissertation
-
Source
-
PQT Legacy CUNY.xlsx
-
degree
-
Ph.D.
-
Program
-
Psychology
