Automatic and volitional processes in the verbal responses of brain-damaged and normal subjects
Appelbaum, Stephen Arthur
MetadataShow full item record
This research was stimulated by the clinical observation that a brain-damaged person, with impairment of recent memory, did well on an apparent recent memory task, reproduction of his 50 responses to a word association test. A model of how such responses come about was postulated, the main elements of which are grouped into two intervening variables: (1) The Automatic Process refers to the immediate arousal and ordering, according to probability of occurence, of previously associated possible responses. (2) The Selective Process refers to the choosing of the behavioral response according to current situational requirements. It was hypothesized that brain-damage alters the normal process in specified ways. The predicted consequences of this were tested through repetition of a word list under changed cognitive requirements. Brain-damaged people were said to differ from normals in that their threshold for less dominant responses is raised, with consequent shortening of their hierarchy and tendency for most dominant responses to recur; and in their impaired ability to modify established response patterns according to current stimulus conditions. The following experimental hypotheses were derived:. I. Upon repeated presentation of identical stimuli to normal and brain-damaged people, if capability for recent recall is controlled, the brain-damaged will be better able to reproduce previous responses than will the normal. II. Upon repeated presentation of identical stimuli to normal and brain-damaged people, with the requirement to give responses different from those already given, brain-damaged people will tend more than normals to repeat their previous responses. The measure of recent recall was a paired associate test in which the pairing of numbers and words reduces conformity to previously established patterns to a minimum. This measure was also used to discriminate psychologically impaired brain-damaged from those not so impaired. Two groups of 20 subjects each, one group normal and the other selected on the basis of psychological impairment and medical diagnosis of brain-damage, gave one free association to each of 50 words taken from the Kent-Rosanoff List. Immediately following, the list was repeated with the instruction to give the same word they had just given. This was the recall test of Hypothesis I. Following this the words were again repeated with the instruction to give a response different from the previous ones. This was the new-response test of Hypothesis II. Analysis of covariance indicated that recent memory played a very small role in the recall task and that with or without correction for recent memory the brain-damaged did as well, but not better, than the normals. With correction the mean of the brain-damaged group was higher than that of the normals, but the difference was not significant statistically. Thus Hypothesis I was not confirmed. Results of the test of Hypothesis I revealed that a recall task apparently dependent upon memory may in fact be performed through operation of the Automatic Process, and that on such a task brain-damaged people do as well as normals. This provides a plausible explanation for the clinical observation which stimulated the research and supports the postulation of the Automatic Process. The hypothesized difference between the groups in Selective Process functioning tested by Hypothesis II was supported, a Student t test yielding a probability less than .01. When, through analysis of covariance, the effect of recent memory, as measured by the paired associate test, was removed, there was no difference between the groups. It is inferred from this that Selective Process factors operating in this task are highly correlated with recent memory. The demonstrated behaviors, often described as perseveration, rigidity, and concreteness were shown to differ from other behaviors similarly labelled.
Thesis (Ph.D.)--Boston University