حواس پرتی شنوایی و اختلال در تفکر در بیماران بستری مزمن اسکیزوفرنی: شواهد برای کمک جداگانه با ناتوانی و تخصیص ضعیف و سندرم فرعی مربوط به نقصان تخصیص

Abstract This study investigated auditory attentional processes associated with schizophrenic thought disorder. Thirty-five chronically schizophrenic, state hospital inpatients were assessed for thought disorder using the Thought, Language, and Communication Disorders Scale (TLC) and tested in an attentional task. Two measures of attention were derived from the Digit Span Distraction Test (DSDT) (Oltmannns, T.F., Neale, J.M., 1975. Schizophrenic performance when distractors are present: attentional deficit or differential task difficulty. J. Abnorm. Psychol. 84, 205–209), a digit recall task in which distractor digits were interspersed with target digits. The two measures were Distractibility — the overall inaccuracy of recall — which measured attentional incapacity, and Distractible Intrusion — the number of irrelevant digits recalled — which was developed in this study to separately measure the inability to allocate attentional resources. These two measures predicted thought disorder strongly and independently. Distractibility did not significantly correlate with any TLC subscale. In contrast, Distractible Intrusions correlated with the TLC subscales Distractible Speech, Incoherence, Loss of Goal, and Word Approximations. The present findings suggest that these subtypes may comprise a distinct subsyndrome of thought disorder, characterized by a dysfunctionally low threshold for selecting appropriate speech information, and that an attentional allocation deficit is related to this language dysfunction.

1. Introduction There is a long tradition of literature linking attention to schizophrenic thought disorder. Kraepelin (1919) suggested that the impairment of thought and speech he observed in dementia praecox patients might be associated with inattention. Since then, many investigators have considered that impaired attention is fundamental to schizophrenia. Chapman and McGhie (1962) suggested that a failure of perceptual gating produced an overload in the information processing system of schizophrenics. Oltmanns and Neale (1975) suggested that active, controlled, limited capacity operations are more vulnerable to distraction in schizophrenic than normal subjects, leading to the observed attentional deficits. Using the Digit Span Distraction Test (DSDT), Oltmanns and Neale (1975) showed that schizophrenics are more vulnerable to auditory distraction than normals. Harvey and colleagues (Harvey et al., 1986, Harvey and Serper, 1990 and Hotchkiss and Harvey, 1990) have shown that auditory distractibility is associated with positive thought disorder. It has been hypothesized that failure to exclude distracting stimuli is related to thought disorder (failure to exclude irrelevant associations). Maher (1972) wrote: The attentional disturbances believed to affect the processing of sensory input … also underlie the failure to inhibit associations from intruding into language utterance. Intrusions of associations into language may be regarded as similar in character to the ‘intrusions’ of background auditory and visual stimuli into the perceptual processes of the schizophrenic patient. (p. 12) This study tests the hypothesis that intrusions of associations into language are related to intrusions of background auditory stimuli and attempts to elucidate the nature of this attentional disturbance. Recent studies of attentional functioning in schizophrenics have used dual task performance to examine allocation (Asarnow et al., 1991 and Granholm et al., 1996). These investigators found that the poor performance of schizophrenic patients in a dual task (simultaneously searching for letters and shadowing text) is not due to impaired allocation of resources; in other words, that the limited capacity of patients with schizophrenia is not the result of their inability to direct attentional resources as instructed by the experimenters. We hypothesized that poor allocation would be related to the severity of thought disorder in schizophrenic patients. To measure the allocation deficit, an alteration in the DSDT constructed by Oltmanns and Neale (1975) was devised. These investigators measured the recall of target digits when distractor digits were interposed. The same task was used, only with sets of target and distractor digits in each trial that were non-overlapping. This permitted the separate measurement of capacity and allocation. The inaccuracy of recall under conditions of distraction was used as a measure of attentional incapacity, and the number of distractors incorrectly recalled as a measure of impairment of allocation. This study seeks to advance our understanding of the relation between auditory distraction and thought disorder in schizophrenia in two ways. First, we looked at whether deficits in capacity and allocation of resources are specifically and separately related to the severity of thought disorder. It is well established that poor recall of digits under conditions of distraction is related to the severity of thought disorder (Harvey et al., 1986, Harvey and Serper, 1990 and Hotchkiss and Harvey, 1990). We asked if an inability to allocate resources, such that distractors are included, is specifically and independently related to thought disorder. The second issue examined by this study is whether distraction is specifically related to any particular subtype or dimension of schizophrenic thought disorder. Despite widespread agreement that thought disorder is a multidimensional construct (Andreasen, 1982, Barch and Barenbaum, 1996 and Cutting and Murphy, 1986), it is often treated as a unitary process, explainable by a single cognitive deficit or language dysfunction. We hypothesized that certain subtypes of thought disorder, namely those characterized by disorganized thinking, would be more strongly correlated with distraction than others. Recently McGrath (1996) proposed a model of the pathogenesis of thought disorder based on the notion of cognitive set, suggesting that poor error monitoring and disruptions in establishing, maintaining, and changing set are processes that underlie thought disorder. According to McGrath, poverty of speech is related to inability to initiate a set; inability to maintain a set underlies tangentiality, derailment, loss of goal and distractible speech; while inability to change sets is related to perseveration. Inability to allocate resources so as to exclude distractor digits is an example of inability to maintain set, and we hypothesize that poor allocation will be related to tangentiality, derailment, loss of goal, and distractible speech as well as to incoherence. Even though McGrath does not include incoherence in this group, it clearly should be included since it is a severe and highly characteristic symptom of disorganized and disconnected thinking. Illogical thinking might be included as a type of disorganization, but it is not disorganized because of disconnectedness as the others are, so its status is ambiguous. In this study the Scale for Thought, Language, and Communication Disorders (TLC) (Andreasen, 1986) was used to measure thought disorder. This instrument was selected because it is widely used, comprehensive, and clinically relevant. Moreover, the TLC is relatively unconstrained by theory, consisting simply of 18 scales to rate commonly observed abnormalities of speech, and it can be scored on any clinical interview of sufficient length.

3. Results The mean Global TLC score for males (2.67, S.D.=1.12) was significantly greater than that for females (1.57, S.D.=1.02) (t=2.88, P<0.007). There were no significant differences on TLC global scores among racial groups. The mean Global TLC score was 2.23 (S.D.=1.23, max.=4, min.=0) indicating moderate–severe thought disorder. Derailments, incoherence, illogicality, and perseveration had the highest rate of occurrence in this sample, consistent with the chronicity and severity of illness in this sample of patients. We evaluated the possibility that patients gave distractor digits as responses due to random guessing by measuring the percentage of non-included digits relative to the total number of non-target responses given by patients. Ten per cent of the non-target digits were non-included, and 90% were distractors. If patients gave distractors because of random guessing, the percentage of non-included digits would be 25%. The observed and expected frequencies were compared by Chi square analysis. The rate at which non-included digit was given was significantly less than chance (χ2=5.95, P=0.015), suggesting that individuals attempted to include only target digits and were prevented from doing so primarily by the distractors. Furthermore, when random errors occurred, they did not significantly correlate with TLC measures. Table 2 presents the results of a two-part correlational analysis. In part 1, Global Ratings were correlated with Distractibility and Distractible Intrusion, and the correlations were significant. Because the correlation between Distractibility and Distractible Intrusion was nearly significant (r=0.32, P=0.06), an analysis using partial correlations was conducted to determine whether the two variables predicted thought disorder independently, i.e. to determine if the correlation between Distractiblilty and thought disorder was spurious: due merely to the association between Distractibility and Distractible Intrusion. The partial correlation between Distractibility and Global TLC, controlling for Distractible Intrusion, was significant (r=0.37, P=0.03). The partial correlation between Distractible Intrusion and Global TLC, with Distractibility held constant, was also significant (r=0.40, P=0.02). This shows that Distractibility and Distractible Intrusion are independently correlated with Global TLC scores. Together, these variables account for 30% of the variance in Global Rating. Table 2. Correlations of attentional variables with TLC subscales Distractible Intrusiona Distractibility Global TLC Rating 0.48 0.46 Subscales: Poverty of Speech 0.08 0.17 Poverty of Content of Speech −0.10 −0.06 Pressure of Speech 0.17 0.32 Distractible Speech 0.62* 0.11 Tangentiality 0.16 0.11 Derailment 0.42 0.29 Incoherence 0.52* 0.18 Illogicality 0.41 0.34 Clanging 0.38 0.19 Neologisms 0.17 −0.01 Word Approximations 0.52* 0.25 Circumstantiality 0.04 0.28 Loss of Goal 0.49* 0.23 Perseveration 0.13 0.14 Echolalia 0.05 0.18 Blocking −0.01 0.07 Stilted Speech 0.17 0.06 a *P<0.0029 (significance level with Bonferroni correction). Table options In part 2 of the correlational analysis, Distractibility and Distractible Intrusion were correlated with the 17 TLC subscales to determine if any particular aspects of thought disorder are specifically related to distraction. This analysis showed that, after the application of a Bonferroni correction for multiple tests, there were no significant associations between Distractibility and particular subtypes of thought disorder. On the other hand, Distractible Intrusion was strongly and specifically related to certain subtypes of thought disorder, namely, Distractible Speech, Incoherence, Loss of Goal, and Word Approximations. There was a trend of association which did not reach significance for Derailment and Illogicality. An analysis was conducted to determine whether the process measured by Distractible Intrusion is unidimensional. A stepwise multiple regression analysis was performed, using Distractible Intrusion as the dependent variable and the scores on the four TLC subscales as predictors. The results are presented in Table 3. Distractible Speech entered the equation in the first step, and no variables entered the equation after Distractible Speech, demonstrating that none of the other three variables correlated with Distractible Intrusion independently of Distractible Speech. This suggests that all four TLC subscales are associated with the same segment of variance of Distractible Intrusion, i.e. that Distractible Intrusion is unidimensional with respect to these four measures. Table 3. Results of stepwise regression analysis relating TLC subscales to Distractible Intrusions Variable Entered R R2 Adj. R2 Step 1 Distractible speech 0.622 0.387 0.368 F(1,33)=20.8; P=0.000 Table options In order to show that the relations between distraction and thought disorder were not due to an association between global psychopathology and weak performance on cognitive tests, we computed the correlations with possibly confounding variables. Correlations of Distractibility and Distractible Intrusion with Positive, Negative, and Total symptoms on the PANSS were not significant, which shows that distraction was not associated with thought disorder because more globally ill patients tended to be more easily distracted. The correlation between Global Rating and IQ was not significant, which shows that the significant correlation between thought disorder and distraction did not arise simply because the patients who were more thought disordered did more poorly on cognitive tests.