THE MODALITY SHIFT EFFECT IN SCHIZOPHRENIA -- FACT OR ARTIFACT?
Item
-
Title
-
THE MODALITY SHIFT EFFECT IN SCHIZOPHRENIA -- FACT OR ARTIFACT?
-
Identifier
-
AAI8302531
-
identifier
-
8302531
-
Creator
-
MANNUZZA, SALVATORE.
-
Contributor
-
Mitchell L. Kietzman
-
Date
-
1982
-
Language
-
English
-
Publisher
-
City University of New York.
-
Subject
-
Psychology, Experimental
-
Abstract
-
The general finding of crossmodal reaction time research has been that schizophrenics are disproportionately retarded when the stimulus of the previous trial is in a different modality: i.e., they are significantly more slowed than normals on "crossmodal," relative to "ipsimodal," trials. The interpretation of this finding has been that schizophrenics are impaired in their ability to shift attention across stimulus modalities.;An alternative interpretation is that the "modality shift effect" is simply reflecting the tendency of schizophrenics to respond more slowly than normals, coupled with the greater ability of the cross-modal condition to discriminate slow from fast responders. Chapman and Chapman contend that, whenever schizophrenics and normals are compared, experimental and control variables must be matched on "discriminating power" (i.e., the ability to discriminate "less competent" subjects from "more competent" subjects) to control for the possibility that psychometric artifact is accounting for the findings.;The purpose of the current research was to assess whether psychometric artifact could account for the modality shift effect. The sample consisted of 15 hospitalized, male patients (mean age = 29.6 yrs., SD = 8.2 yrs.) who fulfilled diagnostic criteria for schizophrenia, and 50 male controls (mean age = 29.3 yrs., SD = 7.8 yrs.) who were screened for a past or present episode of mental disorder. Ten intensities of auditory stimuli (2000 Hz tones) and 10 intensities of visual stimuli were presented in quasi-random order. Stimuli within each modality ranged from 1.8 to 2.7 log units above the threshold determined for one normal. All stimuli were 10 msec. Subjects were instructed to lift their finger from a key, as quickly as possible, when any stimulus was presented.;The major finding was that psychometric artifact could not account for the modality shift effect. Highly significant (p < .001) modality shift differences between patients and controls were found for both modalities. The reliabilities and variances (the critical determinants of discriminating power) of the crossmodal and ipsimodal conditions of normals (N = 50) and of a subgroup of slow-responding normals (N = 15), were assessed. Reliability coefficients were high (.98-.99), and variances did not differ significantly. Therefore, psychometric artifact could not account for the modality shift effect since crossmodal and ipsimodal conditions were comparable on estimates of discriminating power.
-
Type
-
dissertation
-
Source
-
PQT Legacy CUNY.xlsx
-
degree
-
Ph.D.
-
Program
-
Psychology
