ASSOCIATIVE INTRUSIONS IN VOCABULARY PERFORMANCE OF SCHIZOPHRENIC, NONSCHIZOPHRENIC AND NONPSYCHIATRIC PATIENTS by Jan Klinka Charles University, Prague, 1964-1969 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in the Department of Psychology We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA May, 1975 In presenting th i s thes is in par t i a l fu l f i lment of the requirements for an advanced degree at the Univers i ty of B r i t i s h Columbia, I agree that the L ibrary sha l l make it f ree ly ava i lab le for reference and study. I further agree that permission for extensive copying of th i s thes is for scho lar ly purposes may be granted by the Head of my Department or by his representat ives. It is understood that copying or pub l i ca t ion of th is thes is fo r f inanc ia l gain shal l not be allowed without my writ ten permission. Department of PSYCHOLOGY The Univers i ty of B r i t i s h Columbia Vancouver 8, Canada Date APRIL 1, 1975 ABSTRACT Recently (1973),Rattan and Chapman reported the use of the f i r s t testing instrument capable of separating s p e c i f i c from generalized human performance d e f i c i t s and were able to show the presence of s p e c i f i c asso-c i a t i v e i n t rusion d e f i c i t s i n the vocabulary performance of chronic schizophrenics. The Rattan and Chapman Multiple-Choice Vocabulary Test which con-s i s t s of subtests with and without associative d i s t r a c t o r s that are matched on discriminating power was administered i n the present study to 84 hospitalized patients who were equally d i s t r i b u t e d into s c h i -zophrenic, nonschizophrenic, and nonpsychiatric patient groups. The patients i n these groups were also divided on the basis of length of h o s p i t a l i z a t i o n into long- and short term samples. The subjects' vocabulary accuracy was analyzed i n order to ascertain whether a heightened s u s c e p t i b i l i t y toward associative intrusions can be found i n patients other than chronic schizophrenics and whether length of ho s p i t a l i z a t i o n has any effect on the patients' vocabulary performance. The r e s u l t s showed that a l l long-term patients exhibited s i g n i -f i c a n t performance d e f i c i t s on the subtest with associative d i s t r a c t o r s compared to th e i r performance on the matched subtest without associative d i s t r a c t o r s . Of the short-term patients, only schizophrenics showed a tendency toward a s i m i l a r s u s c e p t i b i l i t y to associative d i s t r a c t o r s . These r e s u l t s , though supportive of the associative interference theory of schizophrenia, also indicate that t h i s d e f i c i t i s not peculiar to schizophrenia; rather, i t may be related to prolonged i n s t i t u t i o n -a l i z a t i o n regardless of psychiatric or nonpsychiatric diagnosis. i T A B L E O F C O N T E N T S P a g e A b s t r a c t i L i s t o f T a b l e s iv L i s t o f F i g u r e s v A c k n o w l e d g e m e n t s v i C h a p t e r 1 . I n t r o d u c t i o n a n d F o r m u l a t i o n o f t h e P r o b l e m . 1 C h a p t e r I I . R e v i e w o f R e c e n t L i t e r a t u r e o n S c h i z o p h r e n i c A s s o c i a t i v e I n t e r f e r e n c e a n d S t a t e m e n t o f t h e H y p o t h e s e s 1 0 C h a p t e r I I I . M e t h o d 2 4 S u b j e c t s 2 ^ M a t e r i a l s 3 1 P r o c e d u r e 3 4 C h a p t e r I V . R e s u l t s a n d D i s c u s s i o n 3 6 P e r f o r m a n c e o n t h e R C M u l t i p l e - C h o i c e V o c a b u l a r y T e s t 3 6 V o c a b u l a r y P e r f o r m a n c e 4 6 C o v a r i a n c e A n a l y s i s 5 1 C h a p t e r V . G e n e r a l D i s c u s s i o n 5 6 M a i n R e s u l t s 5 6 P o s s i b l e L i m i t a t i o n s o f t h e P r e s e n t S t u d y I n f l u e n c e o f P s y c h o a c t i v e M e d i c a t i o n . . . 5 9 T h e P r o b l e m o f D i a g n o s t i c R e l i a b i l i t y . . 6 1 C o m m e n t o n t h e N a t u r e o f t h e C o n t r o l S u b j e c t s 6 2 i i Page Age Differences 64 S u i t a b i l i t y of the Vocabulary Measure 64 Nonschizophrenic Subjects 66 H o s p i t a l i z a t i o n Effects 67 Chapter VI. Summary and Conclusions 71 Directions for Further Research 73 Bibliography 76 Appendix 1 . . . . 83 Appendix 2 87 Appendix 3 97 Appendix 4 . . . . 98 Appendix 5 99 Appendix 6 100 i i i LIST OF TABLES Table Page 1 Subject Characteristics 29 2 Means and Standard Deviations of the RC Multiple-Choice Vocabulary Subtest Raw Scores, F i l l e r Items, and Distractor Alternatives for a l l Three Groups at the Two Levels of Hospitalization 37 3 _t - Values and Probabilities for the Difference between ND/D Means for a l l Three Groups at the Two Levels of Hospitalization 38 4 Mean Percentages of Associative Distractor Errors in Relation to Total Errors Made by the Patient Subgroups on the D Subtest 39 5 Summary of the Analysis of Variance of the RC Multiple-Choice Vocabulary Subtest Scores 42 6 Summary of the Analysis of Variance of the RC Multiple-Choice Vocabulary Subtests and Simple Main Effects Analyses for the G x L and T x L Interactions. 45 7 Means and Standard Deviations of the WAIS Vo-cabulary Raw Scores for a l l Three Groups at the Two Levels of Hospitalization 46 8 Summary of the Analysis of Variance for the WAIS Vocabulary Subtest and Simple Main Effects Analyses for the H x G Interaction 48 9 Summary of the Analysis of Covariance for the RC Multiple-Choice Vocabulary Subtest Scores . . . . 52 10 Covariance-Adjusted Means of the RC Subtests for a l l Three Patient Groups at the Two Levels of Hospitalization 52 iv LIST OF FIGURES Figure Page 1 D and ND RC Vocabulary Accuracy as a Function of Patient Group and Length of Ho s p i t a l i z a t i o n 4 3 2 D and ND RC Vocabulary Accuracy as a Function of Patient Group and Length of Hosp i t a l i z a t i o n 4 3 3 Combimed D and ND RC Vocabulary Accuracy as a Function of Patient Group and Length of Hospital i z a t i o n 4 4 4 Combined D and ND RC Vocabulary Accuracy as a Function of Patient Group and Length of Hosp i t a l i z a t i o n 4 4 5 RC Vocabulary Accuracy as a Function of Subtest and Length of Hosp i t a l i z a t i o n 4 7 6 RC Vocabulary Accuracy as a Function of Subtest and Length of Ho s p i t a l i z a t i o n 4 7 7 WAIS Vocabulary as a Function of Length of Hospi t a l i z a t i o n and Patient Group 5 0 8 WAIS Vocabulary as a Function of Length of Hospi t a l i z a t i o n and Patient Group 5 0 9 D and ND RC Covariance-Adjusted Vocabulary Accuracy as a Function of Patient Group and Length of Hospi t a l i z a t i o n 5 4 1 0 D and ND RC Covariance-Adjusted Vocabulary Accuracy as a Function of Patient Group and Length of Hosp i t a l i z a t i o n . . . . . 5 4 1 1 . Combined D and ND RC Adjusted Vocabulary Accuracy as a Function of Group and Length of Hosp i t a l i z a t i o n 5 5 1 2 Combined D and ND RC Adjusted Vocabulary Accuracy as a Function of Group and Length of Hospi t a l i z a t i o n 5 5 v ACKNOWLEDGEMENTS My special thanks go to Dr. Demetrios Papageorgis of the University of B r i t i s h Columbia and to Dr. A l l a n Clark of Riverview Hospital, Essondale, B.C., for t h e i r encouragement, counsel, and patience throughout a l l phases of t h i s study. Further thanks go to Professor Park Davidson of the University of B r i t i s h Columbia who read and c r i t i c a l l y evaluated the present study as a member of the thesis committee. My thanks are also extended to Drs. W.J.G. McFarlane, W.T. Brown and R. Arrowsmith of Riverview Hospital for approving the study and making t h e i r f a c i l i t i e s a v a i l a b l e , and to residents and st a f f of Riverview Hospital for assistance i n obtaining information and experimental subjects. I am indebted to Dr. L. Chapman of the University of Wisconsin for his test material, and to Drs.J. Johnson and V. Green of the University of B r i t i s h Columbia for help with analyzing the data. Dr. G.E. Wakefield, Mrs. D. Kaiou and Mr. S. Murkin of Pearson Hospital, Mr. D.J. Weeks, Miss T. Sabatino and Mr. T. Teranishi of Shaughnessy Hospital deserve a l l my warm thanks for their valuable assistance i n data c o l l e c t i o n . v i CHAPTER I INTRODUCTION AND FORMULATION OF THE PROBLEM A person with disordered thought processes, which are manifested for the most part i n his verbal behavior, can usually be recognized as deviant, even by an untrained observer, without much reference to the context i n which he i s seen. Even though unclear utterances, incorrect assertions, i n v a l i d reasoning, and the l i k e are common enough i n the general population, the i r frequency and pervasiveness are not such that they cause serious interference with l i f e adjustment and interpersonal r e l a t i o n s . On the other hand, pronounced elements of thought disorder result i n a host of problems i n r e l a t i o n to a person's l i f e adjustment and are as a rule considered c h a r a c t e r i s t i c of psychosis and especially schizophrenia. In general, disturbances i n thought and language are considered by most investigators to be central and essential features of the schizophrenic psychoses. The pioneering and c l i n i c a l l y astute contributions of Emil Kraepelin notwithstanding, the modern concept of schizophrenia can undoubtedly be credited to Eugen Bleuler. In his c l a s s i c monograph, Dementia Praecox or the Group of Schizophrenias, f i r s t published i n 1911, Bleuler (1950) not only suggested the name "schizophrenia" as a replacement for the less accurate l a b e l of "dementia praecox," but also highlighted the notion of pervasive associative intrusions into the schizophrenic t r a i n of thought. Unlike his equally famous contemporary Kraepelin (1919), who more or less confined himself to the l i s t i n g of the m u l t i p l i c i t y of c l i n i c a l symptoms that he had observed i n schizophrenic patients without much e f f o r t toward the i s o l a t i o n of a few unifying underlying p r i n c i p l e s , Bleuler concluded 2 that disturbance in association, disturbance in affect, ambivalence, and autism were the fundamental symptoms of the schizophrenic psychoses: they were to be found in a l l schizophrenics and they were absent in other types of psychosis. Other symptoms, however striking (e.g., hallucinations, catatonia), were viewed by Bleuler as accessory (i.e., neither necessary nor unique to the schizophrenias). Furthermore, Bleuler assigned to the associative disturbance the primary position among the four fundamental symptoms of the schizophrenic psychoses. For Bleuler, association was a basic mechanism of a l l psychical activity; in the schizophrenias, the thought disorder resulted from a complete or partial break-up (splitting) of the associative threads that characterize and direct normal thought processes. Thus, the progression of schizo-phrenic thought seemed only pa r t i a l l y determined by a specific guiding central idea. Since words of the same, similar, or even opposite meaning as well as irrelevant or nonsensical associates find their way into the broken associative pathways of the patients, much of the schizophrenics' ideation and verbalization gets beyond the normal listener's expectations and comprehension and i s then judged to be fragmentary, i l l o g i c a l , tan-gential, or simply bizarre. Since Bleuler, and perhaps largely as a result of his influence, every major conceptualization of schizophrenia has included thought disorder as either at the core or as a very prominent feature of the psychosis (e.g., A r i e t i , 1955; Cameron, 1938; Chapman & Chapman, 1973a; Golstein, 1944; Maher, 1972; Meehl, 1962; Sullivan, 1925; Von Domarus, 1944). The most recent revision of the Diagnostic and S t a t i s t i c a l Manual of Mental Disorders (American Psychiatric Association, 1968) 3 continues to emphasize thought disorder as the primary feature of schizophrenia: This large category (i.e., schizophrenia) includes a group of disorders manifested by characteristic disturbances of thinking, mood and behavior. Disturb-ances in thinking are marked by alterations of con-cept formation which may lead to misinterpretation of reality and sometimes to delusions and hallucina-tions, which frequently appear psychologically s e l f -protective. Corollary mood changes include ambivalent, constricted and inappropriate emotional responsiveness and loss of empathy with others. Behavior may be with-drawn, regressive and bizarre. The schizophrenias, i n which the mental status i s attributable primarily to a thought disorder, are to be distinguished from the Major affective illnesses (q.v.) which are dominated by a mood disorder, (p. 33) Moreover, data from the World Health Organization's International Pilot Study of Schizophrenia (Carpenter et a l . , 1973) have shown twelve signs and symptoms to be especially discriminating between schizophrenia and other psychiatric diagnoses in patients from nine different countries (Colombia, Czechoslovakia, Denmark, India, Nigeria, China, USSR, UK, and USA): restricted affect, poor insight, thoughts believed to be spoken aloud or broadcast, poor rapport, widespread delusions, incoherent speech, unreliable or incredible information, bizarre delusions, n i h i l i s t i c delusions, absence of early morning awakening, absence of depressed facies, and absence of elation. As can be readily seen, several of these signs and symptoms are closely tied to disordered thought processes. In line with the assumed and widely accepted centrality of thought disorder in schizophrenia, there have been numerous suggested explana-tions of i t s nature and/or origins (e.g., A r i e t i , 1955; Bleuler, 1950; Broen, 1968; Cameron, 1938; Chapman & Chapman, 1973a, Cohen & Camhi, 1967; 4 Cromwell & Dokecki, 1968; Goldman, 1962; Golstein, 1944; Haley, 1959; Jung, 1936; McGaughran & Moran, 1956; McGhie & Chapman, 1961; McReynolds, 1960; Mednick, 1958; Payne et a l . , 1959; Payne, 1973; Salzinger, 1971; Shakow, 1962; Su l l i v a n , 1944; Tutko & Spence, 1962; Venables, 1964; Van Domarus, 1944). These attempts at explanation, together with the empirical support obtained to date, have been summarized and evaluated by Chapman & Chapman (1973a). Numerous methodological inadequacies i n the relevant research (Chapman a Chapman, 1973a) preclude any unequivocal conclusions. Several of the above viewpoints about the nature and origins of thought disorder i n schizophrenia are actually variants of a hypothesis that attributes the thought disorder (as wel l as other forms of performance d e f i c i t i n schizophrenia) to int e r f e r e from competing responses, often of an associative nature (Land & Buss, 1965). Especially noteworthy i n terms of i t s empirical support i s the Chapman version of the associative interference hypothesis which postulates that the schizophrenic thought disorder consists of accentuations of normal associative response biases (response bias being defined as a predisposition toward making a p a r t i c u -l a r one of the various possible kinds of responses that one might make to a given stimulus). A recent summary of the Chapman hypothesis and of i t s supportive evidence i s available i n Chapman & Chapman (1973a); more recently, further support has been provided by M i l l e r (1974), Rattan & Chapman (1973), and Roberts & Schuman (1974). Having recognized the tendency toward certain kinds of verbal errors i n normal'people under somewhat exceptional conditions such as fatigue, lack of sleep, sensory deprivation, hallucinogenic drugs, and the l i k e , 5 Chapman et a l . (1964) contended that these manifestations of thought disorder closely resemble those found i n schizophrenics. To turn the point around, the features of the thinking disturbance that are viewed as peculiar to the schizophrenic psychoses are also found in normal subjects, though, to be sure, to aiusually much lesser extent (Chapman, 1958; Chapman & Chapman, 1965; Chapman et a l . , 1964; Miller & Chapman, 1968; Rattan & Chapman, 1973). On experimental tasks that c a l l for con-ceptual sorting and word definitions — the kinds of tasks that had been the primary focus of the earlier research by Chapman and his associates — schizophrenics make the same kinds of errors as normal subjects, but to a greater degree. For example, the schizophrenic, when confronted with a word of multiple meaning on a vocabulary test, follows a response bias that tends to favor the commonly preferred meaning of that word despite the fact that, i n the particular context, the correct response clearly requires the use of a less preferred meaning of the word (e.g., Chapman & Chapman, 1965). Other investigators who have inquired into schizophrenic cognitive d e f i c i t , have demonstrated that associative disturbance i n schizophrenic samples i s more frequent than i n some non-schizophrenic psychiatric groups or normal subjects (e.g., Broen & Storms, 1967; Payne et a l . , 1959; Storms & Broen, 1972; Weckowicz & Blewett, 1959). Thus, the notion of the schizophrenics' heightened susceptibility to associative distractors seems to have been substantiated by compelling evidence from a large number of empirical studies carried out to date. Unfortunately, with the exception of recent investigations by Chapman 6 and his associates within the associative interference frame of reference (Chapman et a l . , 1974; Rattan & Chapman, 1973), the above studies as well . as other studies concerned with the demonstration of a specific schizo-phrenic d e f i c i t may be lacking in parsimony. A specific d e f i c i t , in terms of excessive associative intrusions into schizophrenic verbal behavior, may simply reflect a generalized intellectual d e f i c i t i n schi-zophrenic performance on tests, regardless of the investigators' special focus on a hypothesized specific d e f i c i t (Chapman & Chapman, 1973b). In general, previous investigations have used designs that have not allowed the unequivocal demonstration of a specific d e f i c i t (see Chapman & Chapman, 1973a, especially Chapters 3 and 4, for a detailed discussion of the methodological problems). It has been only very recently that a testing instrument capable of separating out the effect of generalized performance d e f i c i t has been developed and used to demonstrate the specific schizophrenic susceptibility to associative distraction. Rattan & Chap-man (1973) constructed two multiple-choice vocabulary subtests, one sub-test containing items with an associative error built in as an alterna-tive to the correct answer, while items in the other subtest did not offer associates among the incorrect alternatives. These subtests, i.e., one with associates as distractors and the other without associates as dis-tractors — when matched on discriminating power (power to distinguish the more able from the less able subjects) permit the differentiation of subjects who are deficient in the specific a b i l i t y measured by the test. Two tests adequately matched on discriminating power have to be, according to Chapman & Chapman (1973b), matched for a normal population on r e l i a b i l i t y 7 ( c o e f f i c i e n t alpha),mean item d i f f i c u l t y , variance and shape of the d i s -t r i b u t i o n of item d i f f i c u l t y , and shape of the d i s t r i b u t i o n of test scores. Then, any cognitively disturbed group of individuals that obtains d i f f e r e n t scores on the two subtests and shows a greater loss i n one a b i l i t y than i n one or more other a b i l i t i e s i s displaying a s p e c i f i c and not a generalized d e f i c i t . Using t h e i r newly developed te s t , Rattan & Chapman (1973) compared the vocabulary performance of chronic schizophrenic patients to the voca-bulary performance of normal controls (firemen and prison inmates) and found that chronic schizophrenics not only made more errors than normals on both subtests, but that they also erred more on the with-associates subtest. Since the investigators controlled for general d e f i c i t by mat-ching the subtests on discriminating power, the schizophrenics' less ac-curate performance on the with-associates subtest could be attributed to a s p e c i f i c d e f i c i t * i n that area of cognitive function. Tasks which are mat-ched on discriminating power not only permit the demonstration of s p e c i f i c d e f i c i t s : i n other instances, a presumed d e f i c i t can be shown to have been spurious i f schizophrenic performance shows no d e f i c i t on any of the matched subtests. Such, for example, was the case i n the very, recent Chapman et a l . (1974) investigation which showed that schizophrenics are not prone to voca-bulary performance d e f i c i t s with affect-laden as opposed to a f f e c t i v e l y neutral words. The question that remained unanswered i n the Rattan & Chapman (1973) study i s whether the s p e c i f i c overresponsiveness to associa-tes i s a phenomenon that i s unique to chronic schizophrenia (and to schizophrenia i n general) or i f i t i s common to ' I t i s recognized that the term ' d e f i c i t ' may contain unwarranted evaluative connotations i n addition to i t s descriptive status. The author wishes to point out that ' d e f i c i t ' , as used throughout the present study, i n no way implies judgements of i n f e r i o r i t y , d i s a b i l i t y , etc. I t r e f e r s , very simply to d i f f e r e n t i a l performance, and i t s use was dictated by a long established t r a d i t i o n i n investigation of schizophrenia. 8 psyche-pathology or to psychosis i n general. There are some suggestions from previous work on the subject (Blumberg & G i l l e r , 1965;Feinberg & Mercer, 1960; Moran, 1953) that this kind of cognitive d e f i c i t i s not peculiar to schizophrenia, but can be found i i other psychotically dis-turbed or normal individuals even more often than the accentuation of normal response biases notion would seem to indicate. If Rattan & Chapman's (1973) findings hold for hospitalized mental patients diagnosed as chronic schizophrenics, i t i s reasoned that the same effect can be found in hospitalized acute schizophrenics and not found i n nonschizophrenic psychiatric patients or i n patients who are hospitalized exclusively because of physical i l l n e s s . Such results would confirm the Chapman data obtained on chronic schizophrenics and would reinforce the claim that thought disorder resulting from associative interference i s unique to schizophrenia. If, on the other hand, i t is shown that hospitalized psychiatric patients, other than those suffering from schizophrenia, perform in a similar fashion as schizophrenics do on the experimental task, the as-sociative interference notion should be modified to include these groups as well; in addition, claims for i t s specific relevance to schizophrenia may have to be abandoned. The present study undertakes to test the aforementioned hypotheses, and, additionally, to check on effects of institutionalization — a variable that has been too often l e f t uncontrolled, particularly in studies comparing dif f e r e n t i a l performance in schizophrenic and normal control samples (Mednick & McNeil, 1968; Strauss, 1973; Wynne, 1963). The inclusion 9 of additional samples of experimental subjects ( i . e . , non chronic schizophrenics and nonschizophrenic psychiatric patients), and control subjects ( i . e . , nonpsychiatric patients) as we l l as the arrangement of a l l subjects into groups with two l e v e l s of h o s p i t a l i z a t i o n (short and long-term duration) extends the design used by Rattan & Chapman (1973) to these other groups of patients and thus allows the assessment of as-sociative interference i n disorders other than chronic schizophrenia. 10 CHAPTER I I REVIEW OF RECENT LITERATURE ON SCHIZOPHRENIC ASSOCIATIVE INTERFERENCE AND STATEMENT OF THE HYPOTHESES In the f i r s t of three experiments dealing with t h e i r theory of associative interference, Chapman et a l . (1964) compared normal and schizophrenic samples with regard to a b i l i t y to i n h i b i t or s i f t out i r -relevant or misleading words on a number of tasks. To begin with, Chap-man et a l . constructed l i s t s of words with more than one meaning and calculated the r e l a t i v e strength of these meanings. The l a t t e r was done by means of ratings of associative strength of words by normal i n d i v i -duals^ usually college students. On the task i t s e l f , subjects were pre-sented with items i n which the context determined the correct meaning response. In some of the items the correct response involved the stronger (or preferred) meaning of the word while i n other items the correct re-sponse was based on the word's weaker (or non-preferred) meaning. As pre-dicted by the theory, schizophrenics approximated the performance of normal subjects when the proper response depended on the stronger meaning, but made more errors when the proper response was based on the weaker meaning. An example of an item where schizophrenics made more errors by responding to the stronger but incorrect meaning i s as follows: When the farmer bought a heard of c a t t l e he needed a new pen. This means a) He needed a new wri t i n g implement b) He needed a new fenced enclosure c) He needed a new pick-up truck Here, the correct answer i s alternative (b) which involves the weaker 11 denotative meaning of "pen". Alternative (a), which involves the stronger meaning, is incorrect in this context and can function as an associative distractor. Finally, alternative (c) i s an incorrect and irrelevant error, included for control purposes. On the other hand, schizophrenics made fewer errors where the cor-rect response was also the one with the stronger meaning: The professor loaned his pen to Barbara. This means a) He loaned her a pick-up truck b) He loaned her a writing implement c) He loaned her a fenced enclosure. In this context, the correct response (b) also involves the stronger denotative meaning of "pen". ; The errors are presumably mediated by a response bias toward the stronger meaning. Maher (1966) describes the effect as follows: ..."normal communication requires that the stronger meaning response not be emitted when the context calls for the use of a weaker one. A failure to inhibit the strong meaning response might occur either because the subject is particularly insen-s i t i v e to the cues that c a l l for the weaker meaning, or because some hypothetical inhibitory process is not functioning correctly. Whichever of these two processes i s inferred i s somewhat irrelevant to the empirical prediction to be made. This is that the schizophrenic w i l l produce strong meaning responses whether or not the context c a l l s for them. Where the context does c a l l for them,the schizophrenic communi-cation w i l l be correct and meaningful to the listener; where the context c a l l s for the weaker meaning response, then the schizophrenic communication w i l l be incorrect and confusing." (Maher, 1966, p. 418) Another study by Chapman and his associates (Chapman et a l . , 1964) tested for d e f i c i t in assigning objects to common conceptual classes, an error tendency which for decades has been viewed as one of the most prominent features of schizophrenic thought disorder. According to the hypotheses, schizophrenics, by relying on the strongest normal meaning responses to a class name (and at the same time ignoring weaker responses), would be expected to make more errors than normals when re-quired to sort out cards into conceptual categories having more than one meaning. In the experimental task, subjects were presented with cards containing names of animate, inanimate, and irrelevant objects and were asked to sort the cards into the four categories: 12 a) things that have a head b) things that have legs c) things that have teeth d) things that have skin The animate category included words l i k e r a t , dog, cow, horse, and man while the inanimate category contained words such as pin, n a i l , match; chair, bed, table; saw, rake, comb; and prune, potato, banana. Because a group of student judges interpreted the four conceptual classes primarily i n terms of animate examples, i t was predicted that schizophrenics, more than normals, would choose cards with animate class names at the expense of inanimate ones. The findings supported the hypo-theses: there were more exclusions of inanimate examples i n the schizo-phrenic sample while exclusions of animate objects were about equal for both normal and schizophrenic subjects. The t h i r d study i n the Chapman et a l . (1964) series showed that strong contextual cues can help schizophrenics make responses that are mediated by the weaker meanings of words, and consequently, to reduce their error rate to the l e v e l of normal subjects. Experimental subjects who took part i n t h i s study were matched on vocabulary, and then adminis-tered a multiple-choice vocabulary test for words of double meaning; they were asked to choose the correct meaning (strong or weak) under conditions where the other meaning did not appear i n the same context. Relative strength of the meanings was obtained from a normal group of students. An example of these vocabulary items follows: The word BEAR may mean The word BEAR may mean a) to carry a) a sharp end b) to command b) an animal c) neither of the above c) neither of the above d) I don't know d) I don't know 13 There were no differences between error rates of schizophrenics and normals (both groups appeared less accuaate on weaker meaning re-sponses) , the implication being that schizophrenic patients can respond to weaker meanings in situations where the stronger meaning i s absent, and, hence, cannot intrude. The above finding bears some similarity to the results reported by Hamlin et a l . (1965) who attempted to alter the alleged concreteness of schizophrenic patients in explaining proverbs. Instead of the usual single-proverb format used i n test situations of this kind, the subjects were provided with groups of three proverbs. Under such conditions of enriched stimulus input, where the patients were given more instances from which to generalize (and presumably also had the opportunity to u t i l i z e stronger contextual cues), i t was found that schizophrenics with a mild degree of concreteness were able to improve their explanations, but that severely disturbed schizophrenics who showed i n i t i a l l y a high degree of concreteness, were not. Thus, the results of both studies suggest that some schizophrenics may benefit from strong context when the basic stimulus i s weak or ambiguous. Chapman & Chapman (1965) presented further supportive evidence for the associative interference theory when they studied the schizo-phrenic's excessive reliance on the strongest (dominant) meaning of words. F i r s t , the investigators obtained a measure of the degree of similarity between pairs of words from college students; these word pairs, varying from high to low similarity, were then administered to schizophrenic, brain-damaged, and normal individuals. Schizophrenic 14 patients judged similar-meaninged word pairs to have the same meaning and did so even more when the strongest meaning response for both words was identical. The foregoing exposition dealt with the most recent theoretical formulation of the associative interference hypothesis (Chapman et a l . 1964; Chapman & Chapman, 1965). The position, however, stems from a number of earlier empirical investigations by Chapman and his co-workers and by other scientists. Some of these previous studies w i l l be b r i e f l y touched upon i n what follows. It had already been demonstrated by Chapman (1958) that schizo-phrenic subjects exceed normal subjects i n the number of associates which occur as inappropriate intrusions on a conceptual card sorting task. In this study, each subject was required to sort stimulus cards bearing the name of a common object with one of three other words on the basis of membership i n the same conceptual class. One of the three words shared a concept with the word on the stimulus card, another had an as-sociative (competing) connection to i t , and the third was irrelevant. On one of the card combinations, for example, the word RAINCOAT was to be sorted conceptually with the easy conceptual word UMBRELLA, or with the high associative distractor WATER, or with the irrelevant word PENCIL. As anticipated,the response word WATER, evoking an incorrect as-sociation with the stimulus word RAINCOAT, was on the average chosen more often by schizophrenics than by normals. Downing et al.(1963, 1966) elaborated on Chapman's (1958) three types of verbal distractor test. Using associatively linked words, 15 contiguity words, or rhyme-clang words as d i s t r a c t o r s , these i n v e s t i -gators compared the number of errors produced by acute schizophrenics. The results confirmed those obtained by Chapman (1958): the errors made by schizophrenics on meaningful thought irr e l e v a n t associates exceeded their errors roduced by the other two types of d i s t r a c t o r s . S i m i l a r l y , Faibish's (1961) hypothesis that more associative interference intrusions would be noted i f words presented to schizophrenic patients re of multiple meaning than i f they have only one meaning, was confirmed and explained i n l i n e with the associative interference frame of reference. Having exposed normal and schizophrenic groups to words of a single or of several meanings with instructions to define them and to free-as-sociate to them, Faibish found that both normals and schizophrenics did even poorer than normals. Faibish's findings were further supported by Willner (1965) whose chronic schizophrenic patients were able to give fewer unfamiliar abstract meanings of words than brain-damaged patients on the unusual meaning vocabulary t e s t . In one of the e a r l i e r studies, Moran (1953) reported that schizo-phrenics were less able to deal with abstract analogies and less l i k e l y to give abstract d e f i n i t i o n s of words than normals. Moran also found that schizophrenic subjects would accept a larger number of synonyms for for words than ouuld normal subjects, suggesting that schizophrenics have a tendency to choose* words of exceptionally wide range of meanings when presented with the task of selecting synonyms from a group of words with d i f f e r e n t proximity i n meaning. Other studies, by means of which L. Chapman and his co-workers 16 documented the evidence of accentuation of "normal" error propensities to misinterpret and misuse words i n schizophrenia, are also relevant i n this connection. Following the lead of frequent c l i n i c a l observa-tions on the confusion of the l i t e r a l and f i g u r a t i v e usages of some words (or phrases) among brain-damaged and schizophrenic patients i n p a r t i c u l a r , Chapman (1960) conducted a study i n which he required s c h i -zophrenic and normal subjects to select meanings for words that could be interpreted either l i t e r a l l y or f i g u r a t i v e l y . He found that schizo-phrenics made more errors of l i t e r a l misinterpretation (while brain-damaged patients made more f i g u r a t i v e misinterpretations) of words than did normal subjects. On the other hand, a subsequent attempt at r e p l i -cation of th i s study by Eliseo (1963), who controlled for length of ho s p i t a l i z a t i o n and vocabulary l e v e l , f a i l e d to find differences between schizophrenics and a normal control group. On the basis of these equi-vocal data, Maher (1972) has explained instances of the l i t e r a l - f i g u -r a t i v e confusion as " r e f l e c t i o n s of the inappropriate i n t r u s i o n of dominant meaning" (Maher, 1972, p. 11). Investigating another area of disturbed thinking, Gottesman & Chapman (1960) compared 30 schizophrenic and 30 normal control subjects on a l o g i c a l reasoning task which involved the selection of the correct conclusion to be derived from a set of f i v e types of premises. Only the t o t a l number of errors made distinguished the two groups; no s i g n i f i c a n t differences were found between the groups on experimental items designed to e l i c i t error i n l o g i c . A more detailed study along the same Unas, involving checks on verbal i n t e l l i g e n c e (which had not been controlled 17 by Gottesman & Chapman) was carried out by Williams (1964). The results showed that schizophrenics were not inferior to the normals in their total error score; both groups appeared least inclined to draw an affirmative inference on the basis of similar predicates. In a further study, Chapman (1961) asked his subjects to sort out l i s t e d objects on the basis of their belonging or not belonging to a specific category. The objects that did not belong together were of two types, either sharing or not sharing similarities with the cate-gory. Only objects that were similar in some way to the category could presumably evoke a competing response (association) with the category. Schizophrenics, in contrast to normals, responded with greater numbers of associated though incorrect objects while the incidence of unassociated choices did not separate the performance of the two groups. Similarly, Burstein (1961) found that schizophrenics tend to treat homonyms and antonyms (both regarded as associates) as i f they were synonyms to a graater degree than do normals. In a related investiga-tion, Blumberg & G i l l e r (1965) had nonschizophrenic and chronic and acute schizophrenic patients choose a synonym from a group of words including a homonym, an antonym and an irrelevant word. Although the sample of chronic schizophrenics made the most errors, the other groups also showed evidence of distrac t i b i l i t y by the antonyms and homonyms, and the authors concluded that this kind of cognitive dysfunction i s not peculiar to schizophrenia. A recent report by Boland & Chapman (1971) showed that schizo-phrenics, compared to normals, displayed a heightened intrusion of as-sociates on ^ multiple-choice vocabulary test in which the available 18 incorrect alternative included an associate to the stimulus word. More recently, in a study designed to compare the performance of schizophrenics and alcoholics, Roberts & Schuham (1974) modified Chapman's (1958) card sorting test by adding a medium-associative distractor to the high and low associative distraction conditions, and found that schizophrenic subjects made more errors than alcoholic subjects on a l l levels of distraction. In addition, the schizophrenics' associat-ive error scores closely approximated a straight-line function which was interpreted as supporting the notion of hierarchical responding (at least i n three levels of responsiveness) that i s central to Chapman's 'response bias' theory. Miller (1974) also provided further experimental support for the notion of primacy response bias (i.e., the inclination to select the primary, strongest or preferred meanings of multiple-meaning words which i s not attributable to the lack of knowledge of secondary, weaker or less preferred meanings); however, he failed to find a greater primacy error preference specifically for schizophrenics. Chronic and acute schizophrenic subjects combined only made more abstract than concrete errors on a homograph test, compared to hospital employees. Finally, Rattan & Chapman (1973) demonstrated once more the schizo-phrenics' overresponsiveness to associates by using two multiple-choice vocabulary subtests matched on discriminating power (see Chapter I ) . To summarize this brief review, certain research problems that appeared in the studies should be pointed out. The majority of the studies that have been carried out to date on schizophrenic cognitive d e f i c i t , including associative interference, have compared schizophrenic 19 to normal, usually non-hospitalized, samples. The studies c i t e d e a r l i e r may be open to c r i t i c i s m on many grounds: (a) not c o n t r o l l i n g for s a l i e n t variables such as educational l e v e l , age, i n t e l l i g e n c e (Chapman & Chapman, 1973a); (b) not c o n t r o l l i n g for length of h o s p i t a l i z a t i o n (Mednick & McNeil, 1968; Wing, 1962; Wynne, 1961); (c) not c o n t r o l l i n g for l e v e l of cooperativeness (Moran, 1953; Shakow, 1963); (d) not c o n t r o l -l i n g for drug status (Chapman & Chapman, 1973a; Salzinger, 1973; Spohn, 1973); and (e) f a i l i n g to specify the meaning of chronicity (Chapman & Chapman, 1973a; Strauss, 1973). Salzinger (1973), apart from the problem of diagnostic r e l i a b i l i t y , discusses two other methodological p i t f a l l s p a r t i c u l a r l y relevant to research with schizophrenic subjects that are related to 'committal' ( i . e . , what are the selective features bringing about subjects' h o s p i t a l i z a t i o n ) and to ' t e s t a b i l i t y ' which depends on patients' c a p a b i l i t y and willingness to l e t themselves : be tested. Since some of the patients available for the formation of the samples are too confused and others do not agree to p a r t i c i p a t e i n experiments, any sample of schizophrenics i s probably a biased sample of the population of hospitalized patients. Chapman & Chapman (1973a) have broadened the t e s t a b i l i t y variable by adding an aspect to i t which they c a l l e d 'demandingness'. Further methodological flaws, besides the problem of precise matching on the variables most inc l i n e d to affect the dependent variable scores, include absence of control groups, i n s u f f i c i e n t sample s i z e , and invariably the use of inadequate and/or imprecise tools for measuring disordered thought. As mentioned In the introductory chapter and 2 0 elsewhere i n th i s presentation, previous investigators of associative interference i n schizophrenia, with the exception of Rattan & Chapman (1973) and Chapman et a l . (1974) had not equated thei r experimental tasks on discriminating power and, therefore, have f a i l e d to provide an unequivocal demonstration of the schizophrenics' s p e c i f i c s u s c e p t i b i l i t y to d i s t r a c t i o n by associates. The findings of increased associative i n t r u s i o n i n schizophrenia have remained open to either a s p e c i f i c or a generalized d e f i c i t i nterpretation. However, following Chapman & Chapman's (1973b) analysis of the methodological problems i n measuring differences i n s p e c i f i c a b i l i t i e s and their suggestions for procedural improvements, Rattan & Chapman (1973) constructed a suitable testing instrument (a multiple-choice vocabulary test) which allows the separa-t i o n of a d e f i c i t s p e c i f i c to associative interference from a generalized d e f i c i t . With t h i s instrument, i t i s now possible to determine the oc-currence of s p e c i f i c a l l y associative disturbance i n subjects and to compare the extent of such disturbance i n a variety of psychi a t r i c and nonpsychiatric subject populations. I t i s obvious that only l i m i t e d conclusions can be drawn from the previous studies that dealt with schizophrenic cognitive d e f i c i t (and from any future studies on the subject as well) i n which the proper care has not been taken to control for the extraneous sources of variance l i s t e d above. The present study w i l l attempt throughout to either i n -clude the relevant variables that might affect the main dependent measure into the design or to control them s t a t i s t i c a l l y ; where neither of these conditions can be met, the possible sources of error or of results that 21 are discrepant to those of other investigators w i l l be made e x p l i c i t . The Rattan & Chapman (1973) investigation found greater proneness toward associative d i s t r a c t i o n i n chronic, presumably long-term schizo-phrenics when these patients were compared to normal firemen and prison inmates. The present research i s an attempt to r e p l i c a t e the associative d i s t r a c t i o n effects for long-term schizophrenics and, i n addition, to test i t s generality on a sample of short-term schizophrenic patients. I t appears that, i f associative interference i s a c h a r a c t e r i s t i c feature of schizophrenic disturbance, i t s presence should be also demonstrable i n a schizophrenic sample that i s less chronic and less subject to effects of long-term i n s t i t u t i o n a l i z a t i o n . Moreover, as Rattan & Chapman pointed out, t h e i r investigation did not address i t s e l f to the question of whether or not d e f i c i t r e s u l t i n g from associative interference i s a condition unique to schizophrenia. Accordingly, the present study incorporates samples of nonschizophrenic hospitalized psychiatric patients i n an attempt to deter-mine the extent, i f any, of s p e c i f i c associative interference d e f i c i t i n patients who are not diagnosed as schizophrenics. In th i s case such def-i c i t should be absent or at least minimal. F i n a l l y , as a control, the present investigation includes groups of individuals who are suffering from long- and short-term physical i l l n e s s and who are free of psychiatric disorder. Associative interference i n these groups should also be absent or minimal and their performance should be comparable to that of the non-schizophrenics. Since long-term and short-term patients were included i n a l l three patient groups (schizophrenic, nonschizophrenic, and nonpsychiatric), 22 the present study also permitted the assessment of i n s t i t u t i o n a l i z a t i o n effects (or chronicity) on associative interference. In summary, the main hypotheses of the present investigation may be stated as follows: la) long-term schizophrenics w i l l make s i g n i f i c a n t l y more errors on the with-associates (distractor) subtest, than on the without-associates subtest; e s s e n t i a l l y , t h i s r e s u l t would re p l i c a t e the findings of the Rattan & Chapman (1973) study, lb) the same s i g n i f i c a n t tendency for errors on the with-as-sociates subtest to exceed errors on the without-associates subtest w i l l be found i n the case of short-term schizophre-ni c s ; t h i s r e s u l t would extent the Rattan & Chapman (1973) findings to a schizophrenic sample that i s characterized by a r e l a t i v e l y lower degree of chronicity. 2) other patient groups ( i . e . , long-and short-term nonschizo-phrenic psychiatric patients and long- and short-term nonpsychiatric patients) w i l l make the same number of er-rors on the with-associates and the without-associates subtests. Confirmation of th i s hypothesis would lend credence to the view that associative interference i s unique to the schizophrenic psychoses. I t should be noted that these hypotheses have been stated i n terms of findings that would be consistent with the associative interference theory and i t s s p e c i f i c application to schizophrenic disorders regardless of length of h o s p i t a l i z a t i o n . Outcomes that would appear inconsistent with 23 the hypotheses as stated would carry a number of implications and sug-gestions for rev i s i o n of the associative interference theory of s c h i -zophrenic d e f i c i t . 24 CHAPTER I I I METHOD To test the hypotheses that were presented at the end of the previous chapter, the experimental layout called for the formation of three basic groups of subjects: schizophrenic,nonschizophrenic, and nonpsychiatric, each group being divided with respect to length of h o s p i t a l i z a t i o n into a short-term and long-term sample of patients. A l l subjects selected were presented i n d i v i d u a l l y with the Wechsler Adult Intelligence Scale (WAIS) vocabulary subtest (Wechsler, 1955) and with the Rattan S Chapman (RC) multiple-choice vocabulary test (Rattan & Chapman, 1973), i n th i s order. This order was chosen i n order to avoid contamination of the WAIS performance by any associative intrusion sets that might be generated by the RC items that contained associative d i s t r a c t o r s . The score on the WAIS vocabulary subtest served as an estimate of the subjects' current i n t e l l e c t u a l functioning, while the number of correct answers (accuracy) earned by subjects on the two subtests of the RC multiple-choice vocabulary instrument constituted the main dependent variable measure. Subjects A l l three basic groups of patients — schizophrenic (S), non-schizophrenic (NS), and nonpsychiatric (NP) — were selected according to the following h o s p i t a l i z a t i o n c r i t e r i a : for short-term (ST) subjects, continuous stay i n hospital had not exceeded s i x months, whereas long-term (LT) subjects had been continuously i n s t i t u t i o n a l i z e d for more than 30 months. Since each of the s i x subgroups had an equal number of sub-jects (n=14), the t o t a l sample was composed of 84 hospitalized patients. S and NS subjects were hospitalized psychiatric patients at River-view Hospital i n Essondale, B.C.; NP subjects were physically disabled 25 individuals hospitalized at Shaughnessy, Pearson, and Vancouver General (Banfield Pavilion) Hospitals i n Vancouver, B.C. Most of the NP subjects were suffering from tuberculosis, advanced rheumatoid a r t h r i t i s , i r r e -v e r s i b l e damage of the spinal cord, and from multiple-sclerosis without cerebral disorder. As stated e a r l i e r , the psychiatric patients were either schizophrenics or nonschizophrenics, but no attempt was made to control for diagnostic subtypes within the broad categories. Among ST schizophrenics, three had been diagnosed as suffering from acute schizo-phrenic reaction, four as simple type, three as paranoid, two as chronic undifferentiated, and one each as chronic, and chronic paranoid type; of the LT schizophrenic subjects, s i x had been diagnosed as chronic undiffe-rentiated, three as chronic, three as chronic paranoid, and two as paranoid. The diagnosis of ST nonschizophrenic patients were: four depressive reac-tions, four personality disorder, three manic-depressive reactions, and one each as depressive reaction combined with personality disorder, severe depressive neurosis, and acute anxiety reaction combined with obsessive-compulsive neurosis; 12 LT nonschizophrenics appeared d i a g n o s t i c a l l y as manic-depressive, and one each as depressive and in v o l u t i o n a l psychotics. Both male and female subjects were included regardless of th e i r marital status. I t was attempted to set up ST samples to be as homo-geneous as possible — only those subjects who were undergoing t h e i r f i r s t h o s p i t a l i z a t i o n and could be roughly equated on i t s duration, were to be *Diagnoses assigned mainly according to the second editi o n of Diagnostic and S t a t i s t i c a l Manual of Mental Disorders, American Psyc h i a t r i c Association, 1968. 26 selected to form sample groups. Unfortunately, however, the f i r s t admission prerequisite was not met because the f a c i l i t y that supplied psychiatric patients for the project usually accommodates psychiatric cases requiring longer treatment and/or custodial care, and the majority of patients contacted by the author had been transferred from psychiatric wards of general hospitals, boarding homes, penitentiaries and j u d i c i a l or other agencies. Despite these un-favorable circumstances, of the. 14 subjects in each ST group, there were five f i r s t admissions among schizophrenic patients; NS and NP samples both had six f i r s t admissions subjects each. Subjects with two or more previous admissions were included i f the f i r s t admission had been under six months duration; the sane applied i f the patient concerned was in and out of institutions many times, that i s , the total number of months had not ex-ceeded six months i n every single hospitalization period. The subject population was limited to Caucasians over 18 years of age who had received their primary education in the English language and who had completed at least the sixth grade of school. A further r e s t r i c -tion on the psychiatric subjects was that their diagnosis upon the f i r s t admission had not changed during the course of institutionalization (ex-cept for alterations between subtypes), and that none of the subjects had received electroconvulsive therapy three months prior to the study. Additionally, psychiatric residents who were in direct contact with the psychiatric subjects chosen for the study were asked to exclude those individuals with evidence of additional physical disorder, central nervous system pathology, drug addiction, or mental retardation and also to 27 exclude those subjects whose diagnosis with regard to the schizophrenia/ nonschizophrenia dichotomy appeared doubtful. For t h i s purpose, the residents were given a checklist (New Haven Schizophrenia Index — see Appendix 1) designed to ensure uniform c r i t e r i a and allow for future rep l i c a t i o n s of the schizophrenic diagnoses (Astrachan et a l . , 1972), and only those subjects who met the c r i t e r i a of the ch e c k l i s t were included i n the project. The NP subjects' stay i n the h o s p i t a l was determined s o l e l y by the physical i l l n e s s as diagnosed by the attending medical personnel; a control participant i n the study had no history of psychiatric d i s -turbance, and generally was showing adequate s o c i a l functioning. I t was not possible to withdraw subjects who were on psychoactive drugs from medication i n preparation for the study. Because a large portion of schizophrenic patients, when taken off drugs worsen i n their psychotic symptomatology (Chapman, 1963), i t would be unethical to remove these patients from drug therapy s t r i c t l y for l i m i t e d research aims. Further, schizophrenics who can be removed from t r a n q u i l i z i n g drugs are probably r e l a t i v e l y undisturbed and therefore unrepresentative of the schizophrenic population usually seen i n i n s t i t u t i o n s . In a s i t u a -t i o n where the investigator has no way of assigning subjects to a medica-tion/no-medication condition including dosage levels on a random basis, i t i s necessary, i n order to f a c i l i t a t e comparison of his study to r e -lated investigations, to record i n d e t a i l the patients' intake of phar-macological agents — the presence or absence of antipsychotic (and a n t i -Parkinson) medication, type of antipsychotic medication, d a i l y dosage 28 level, and ideally, duration o f drug treatment (Spohn, 1973). A detailed drug status of both experimental and control subjects i s presented in Appendix 2- At this point, i t suffices to note that the present study dealt predominantly with medicated subjects. In total, 59 subjects were taking either one, two or various combinations of psychoactive and/or other drugs, while 14 ST nonpsychiatric, 6 LT non-psychiatric, 3 LT nonschizophrenic, and 2 LT schizophrenic patients were drug-free (or at least free o f neuroleptic drugs). Major demographic characteristics, such as age, education, length of hospitalization, sex, and marital status of the various patient sub-groups are presented in Table 1. A number of one-way analyses of variance and subsequent multiple comparisons across these characteristics were performed to see whether or not the patient samples were properly matched in terms of the varia-bles noted previously. There were altogether six (out of a possible fifteen) significant mean differences found on the age variable (see Appendix 3). Basically, with the exception of the nonpsychiatric patients, the short-term subjects were found to be younger than the long-term subjects: the ST schizo-phrenics were younger than the LT schizophrenics (p <.05), the LT non-schizophrenics (p < .01), and the LT nonpsychiatrie subjects (p < .01); further, the ST nonschizophrenics were younger than the LT nonschizophre-nics (p < .01), and the LT nonpsychiatric patients (p < .05), but not s i g n i f i c a n t l y younger than the schizophrenic patients. While the groups of long-term hospitalization appeared compareble on the age variable, one TABLE 1 SUBJECT CHARACTERISTICS Years Months Diagnosis Hospital Years of age Completed of Sex Marital Status Status i n School* Hospitalization NS NP SD M SD M SD Male Female Si M Se D W ST 28.42 12.61 10.35 2.02 0.85 0.63 12 2 14 - - - -LT 46.07 17.02 8.71 1.72 139.92 171.11 8 6 10 - - 1 3 ST 34.78 12.66 9.64 1.94 1.60 1 .16 5 9 7 3 .1 3 _ LT 60.64 11.23 9.78 1.63 96.00 56.43 3 i i 2 5 4 - 3 ST 46.71 18.50 10.42 1.91 3.57 2.36 12 2 3 7 1 1 2 LT 52.64 13.08 9.21 1.76 69.21 69.45 8 6 5 6 - 2 1 *In averaging the level of education, the number of years beyond 12th grade was not taken into account. In fact, however, there was 1 BA holder among the LT schizophrenics and one subject with an MA degree in the ST nonpsychiatric sample; one LT nonpsychiatric subject had completed a year of a college program, and three ST schizophrenics were found to have had incomplete college education: one having completed one year, one with two years, and one with three years of college. 30 significant difference was found among samples of short-term i n s t i t u -tionalization: schizophrenics were younger than nonpsychiatric patients (p < .05). The groups did not d i f f e r significantly in years completed i n school. (see Appendix 4) Because of their quite comparable educa-tional level, the subjects of a l l samples were assumed to be of ap-proximately equal premorbid intelligence. With respect to length of hospitalization, none of the mean d i f -ferences between samples of patients categorized as being of long-term hospitalization were found to be significant. Within short-term i n s t i t u -tionalization patients, the schizophrenics, as well as the nonschizo-phrenics, when compared to nonpsychiatric patients, were found to have spent significantly less time i n hospital (p < .01; see Appendix 5). However, i n view of the nonsignificant difference between ST schizophre-nics and nonschizophrenics, and considering the fact that a l l samples of short-term hospitalization had met the original criterion (i.e., six or fewer months of continuous stay i n hospital), i t i s reasonable to consider that the subgroups within the LT and ST categories were quite comparable i n terms of length of hospitalization. As revealed by the binomial test, the two sexes were divided about equally within the LT schizophrenic, the LT nonpsychiatric, and the ST nonschizophrenic patient samples. By contrast, short-term samples of schizophrenics and nonpsychiatric subjects were found to be predominantly male (12 males and 2 females; p < .012), while the LT nonschizophrenic sample was predominantly female (11 females and 3 males; p < .058). 31 No s t a t i s t i c a l tests were carried out on the marital status variable. Nevertheless, the following points appear worthy of note: a) ST schizophrenics were a l l single (never married) b) the second highest proportion of single subjects was found among LT schizophrenics (of 14, only 4 subjects had been previously married and were now widowed, and one subject was s t i l l married) c) the total number of married subjects across the six samples appeared to be relatively small (21), when compared to the number of single persons (41), or to the number jof individuals subsumed under the combined category of separated, divorced or widowed (22). Materials The major experimental instrument used i n the present study was the Rattan & Chapman 140-item multiple-choice vocabulary test (Rattan & Chapman, 1973) which consists of 60 randomly ordered with-associate items, 60 withoxit-associate items, and 20 f i l l e r items* The following are examples from the with-associates subtest (item #6) and from the without-associates subtest (item #4), respectively: 6. POOL means the same as 4. NURSE means the same as a) PUDDLE (correct) a) CHAIR (irrelevant) b) COLD (irrelevant) b) SUCKLE (correct) c) SWIM (associative distractor) c) BOX (irrelevant) d) NONE OF THE ABOVE d) NONE OF THE ABOVE For the f i l l e r items, the stimulus words were neither associated 32 nor conceptually related to any of the alternative words so that the answer 'none of the above' was correct. (The word RIVER, e.g., i s obviously unrelated to the words CELE3RATI0N, TIE or LIGHT). According to Rattan & Chapman (1973), the associative d i s t r a c t o r s were words of wide range of d i f f i c u l t y that had, as agreed upon by f i v e normal judges, a strong associative connection with the stimulus word but one that could not be considered a correct answer (e.g., POOL-SWIM) . The strength of associative connection between the stimulus words and the alternatives was estimated for a l l items by 75 college students; cf the two incorrect alternatives for each item, these stu-dents had selected the one which had the dominant associative connec-ti o n (e.g., for the word AIR, the associate BREATHE was chosen decided-l y more often than SUN on a with-associate item, contrary to the word POINT where the alternatives BURN and SING were about evenly d i s t r i -buted among the judges' choices i n the case of one of the without-as-sociate items). Next, 55 freshmen students rated the strength of as-sociative connection between the stimulus word and i t s strongest as-sociate on a 6-point scale for each item using the following format: The tendency for WANT to c a l l to mind SLOW i s a) very strong b) strong c) moderate d) s l i g h t e) very s l i g h t f) no tendency at a l l . Each point on the scale was assigned a value of 1; 'no tendency at a l l ' was worth 1 point while 'very strong' rated 6 points. Rattan and Chap-man, reported mean rated strength of association between the stimulus 3 3 words and th e i r most strongly associated incorrect a l t e r n a t i v e to be 1.49 for the without-associates items and 4.48 for the with-associates items. Further, there was no overlap of the two d i s t r i b u t i o n s i n terms of rated association strength. S t i l l according to Rattan & Chapman, another group of 62 students rated the strength of the associative connection of the stimulus words with the correct alternatives for the two subtests. The reported means were 3.28 for the with-associates items, and 3.00 for the without as-sociates items. The test was standardized on 92 normal subjects of whom 65 were firemen and 27 were prison inmates of lower education and of presumably below-average i n t e l l e c t u a l functioning. Although the inmates averaged 10 points lower than the firemen on each subtest, a l l important test parameters, such as d i s t r i b u t i o n of test scores, mean, variance and d i s t r i b u t i o n of item d i f f i c u l t y , and c o e f f i c i e n t alpha, were found to be highly s i m i l a r on the two subtests regardless of the sample. Thus, the with-associates and the without-associates subtests were matched on those c h a r a c t e r i s t i c s required for equivalence of discriminating power. The instrument proved to be capable of separating out the s p e c i f i c cognitive d e f i c i t from a generalized cognitive deterioration of subjects under investigation. A l l subjects i n the present investigation were also asked to define a standard set of words. They were given the 40-item vocabulary scale of the WAIS to assess t h e i r present functional vocabulary l e v e l and to arrive at an estimate of th e i r i n t e l l i g e n c e ( A l l i s o n et a l . , 1968). 34 Procedure A l l subjects were tested i n d i v i d u a l l y at t h e i r "respective i n s t i -t u tions. The experimenter remained with the subjects throughout the te s t i n g period. F i r s t , the subjects were informed b r i e f l y about the nature of the study which was described to them as dealing with the issue of 'how people use or misuse the Engli s h language'. E f f o r t s were made to e s t a b l i s h proper rapport, to ensure cooperation, and to secure the informed consent of each subject who took part i n the pr o j e c t . (Only two of the p o t e n t i a l subjects who were approached refused to p a r t i c i p a t e ) . Basic personal and demographic information was then obtained from each subject and recorded on a s p e c i a l l y prepared form (see Appendix 6 ) . Later, t h i s information was checked and v e r i f i e d with nursing super-v i s o r s and/or the medical record l i b r a r i a n . The form also included the patient's diagnosis and medication status. The next step i n the procedure was the administration of the voca-bulary subtest of the WAIS. This was followed by the administration of the RC multiple-choice vocabulary t e s t , the dependent measure of the study. The items of the l a t t e r were printed i n e a s i l y readable form, i n c a p i t a l l e t t e r s and with only eight items to a page. The subject was asked to c i r c l e 'the word c l o s e s t i n meaning to the f i r s t word given' i n each item. T y p i c a l l y , the experimenter read the test i n s t r u c t i o n s to the subject while the l a t t e r followed these same i n s t r u c t i o n s as they ap-peared on the cover of the test booklet. The subject then pr a c t i c e d on a sample item. Further questions were s o l i c i t e d and assistance was rendered whenever necessary. The subject then proceeded with the RC t e s t . 35 After the f i r s t eight items were oompleted, the experimenter checked the responses i n order to make sure that the subject had comprehended the i n s t r u c t i o n s ; i f necessary, the instructions were repeated and c l a r i f i e d . For a few S and NS subjects, whose v i s i o n was poor, the ex-perimenter read each item aloud and c i r c l e d the alternative indicated. NP subjects who suffered from impaired motor coordination invariably asked the experimenter to c i r c l e t h e i r choices. Three subjects (1 ST schizophrenic, 1 LT schizophrenic, and 1 LT nonschizophrenic) required two testing sessions because of complaints of fatigue. A l l subjects were allowed (though not encouraged) to take b r i e f (on the average up to two minutes) break(s) during the te s t . The average testing time for nonpsychiatric control subjects was 45 minutes. Experimental subjects, both schizophrenic and nonschizo-phrenic, needed about twice an much time on the average, with a range between 30-210 minutes. 36 CHAPTER IV RESULTS AND DISCUSSION Accuracy score, i . e . , the number of correct answers obtained by the subjects on the two Rattan & Chapman (RC) multiple-choice vocabulary subtests, was the main dependent variable. These scores w i l l be pre-sented and analyzed f i r s t , and they w i l l be followed by the Wechsler vocabulary performance data that were used to estimate the subjects' current i n t e l l e c t u a l functioning. Performance on the RC Multiple-Choice Vocabulary Test Of the subject c h a r a c t e r i s t i c s , the i n t e l l e c t u a l functioning variable was regarded as the one most l i k e l y to produce confounding effects on the main dependent variable measure. Since the subjects' education (in terms of the completed years at school) was found to be approximately equal across the sub-samples, the subgroups were assumed to be matched on premorbid i n t e l l i g e n c e . In other words, schizophrenic and nonschizophrenic subjects' premorbid IQ could be looked at as mat-ched with the nonpsychiatric subjects' present functioning IQ. The comparable premorbid IQ levels thus allowed for a d i r e c t i n t e r p r e t a -t i o n of the accuracy scores obtained on the RC multiple-choice subtests. Table 2 shows the obtained means and standard deviations on the two RC subtests for a l l s i x subject subgroups. A d d i t i o n a l l y , the table contains the means and standard deviations of cor r e c t l y answered f i l l e r items (Fi l l . ) , and the means and standard deviations of actual d i s t r a c t o r a l t e r n a t i v e choices (Distr.) In the f i l l e r items, the correct answer was the alternative 'none of the above'; these items provide a check 37 for random responses. Associative distractors that were part of the with-associates items indicate the extent to which subjects actually chose the incorrect associate a l t e r n a t i v e . TABLE 2 Means and Standard Deviations of the RC Multiple-Choice Vocabulary Subtest Raw Scores, F i l l e r Items, and Distractor Alternatives for a l l Three Groups at the Two Levels of Ho s p i t a l i z a t i o n F i l l . f i l l e r items ND subtest without associates D subtest with associates D i s t r . distractor/associates NS NP F i l l . ND D Di s t r . M SD M SD M SD M SD ST 15.14 3.69 35.28 9.02 30.00 11.44 15.85 12.30 LT 11.85 6.45 23.92 7.65 17.78 7.94 27.94 9.86 ST 14.85 6.60 26.78 12.87 24.07 12.30 18.07 12.87 LT 14.07 6.00 33.07 10.95 26.28 11.33 23.71 12.13 ST 18.57 3.97 41.07 13.69 41.07 11.03 4.64 5.70 LT 15.21 5.19 36.00 7.44 29.64 11.88 18.42 14.64 I t can be seen from Table 2 that the subjects i n the various subgroups — with the exception of the short-term nonpsyatric pa-tien t s — performed d i f f e r e n t l y on the two subtests: the without as-sociates (ND) accuracy means are higher than the with-associates (D) means for a l l subgroups other than the ST/NP patients. These ND/D d i f -ferences within groups were evaluated by two-tailed t-tests for correlated samples with the results shown i n Table 3. For 13 degrees of freedom, the values of _t required for s i g n i f i -cance at the .05 and .01 levels (two-tailed) are 2.16 and 3.01 r e s p e c t i -vely. S i g n i f i c a n t differences (p < .05 or beyond) were obtained between 38 the mean number of correct answers (accuracy) on the two subtests for four of the subgroups. The obtained values of _t exceeded the c r i t i c a l values at the 1 percent l e v e l for the long-term schizophrenics and the long-term nonschizophrenics, and the 5 percent l e v e l for the long-term nonpsychiatric and the short-term schizophrenic subjects. Non-s i g n i f i c a n t differences between the means were found for the short-term nonschizophrenic and nonpsychiatric subjects. TABLE 3 jt-Values and P r o b a b i l i t i e s for the Difference between ND/D Means for a l l Three Groups at the Two Levels of Ho s p i t a l i z a t i o n *p < .05 **p < .01 df = 13 Table 2 also shows that the short-term nonpsychiatric patients gave the lowest number of di s t r a c t o r alternatives (X = 4.64) while the highest number of these choices were found i n the long-term schizo-phrenics (X = 27.94), and the long-term nonschizophrenics (X = 23.71). To show how the int r u s i o n of associates i n the D subtest was responsible for the o v e r a l l number of errors on the subtest, proportions between the number of associative distractors given and the number of t o t a l er-rors ( i . e . , incorrect, 'none of the above', and dis t r a c t o r alternatives combined) were calculated for each group. The percentages, shown i n Table 4, represent the contribution of dist r a c t o r s to the t o t a l error score on the D subtest. Short-term Long-term Schizophrenics 2.94* 4.02** Nonschizophrenics 1.47 3.27** Nonpsychiatric subjects 0.00 2.80* 39 TABLE 4 Mean Percentages of Associative Distractor Errors in Relation to Total Errors made by the Patient Subgroups on the D Subtest Short-term Long-term Schizophrenics 47.9 64.3 Nonschizophrenics 46.5 67.4 Nonpsychiatric subjects 26.9 54.3 The correspondence of these percentages with the results of the _t-test analyses just reported i s apparent: the long-term schizophrenic and the long-term nonschizophrenic patients, having chosen the distractor alternatives most often, appeared clearly distinct from the other subgroups. The long-term nonpsychiatric patients and the short-term schizophrenics along with the short-term nonschizophrenics appear to occupy an intermediate position i n terms of distractor errors, while nonpsychiatric patients of short-term hospitalization endorsed much fewer distractors on the average than the patients i n a l l other sub-groups. The significance of the group means differences on the RC subtests and the differences i n mean percentage of associative distractor contribution established by these i n i t i a l data analyses highlighted two important Sidings that c a l l for further investigation. F i r s t , a l -though the short-term nonschizophrenics did not show a significant mean accuracy score difference on the ND and D subtests, they did respond with an almost identical percentage of associative distractors as their schizophrenic counterparts. This observation suggests that the short-term nonschizophrenics performed less differently on the two 40 subtests (matched on discriminative power) than the short-term schizo-phrenics despite the short-term nonschizophrenics' resorting to asso-c i a t i v e d i s t r a c t o r alternatives as frequently as the short-term schizo-phrenics did when they made errors. One possible explanation for t h i s r e s u l t i s that the short-term nonschizophrenics d i f f e r e d from the short-term schizophrenic subjects i n present i n t e l l e c t u a l functioning. To find out the extent of how the scores on the RC subtests were influenced by differences i n the subjects' current i n t e l l i g e n c e i n these subgroups as w e l l as i n the other subgroups, an analysis of variance was performed on the WAIS vocabulary data. The summary of t h i s analysis w i l l be pre-sented i n the subsection immediately following. Some deterioration i n i n t e l l e c t u a l a b i l i t i e s over the course of the disorder and/or i n s t i t u -t i o n a l i z a t i o n was expected, and, as w i l l be seen, i t was indeed found. Second, a l l three groups tended to cluster on the basis of pre-ference for the d i s t r a c t o r alternative as opposed to the correct a l t e r -native or to an incorrect one other than the d i s t r a c t o r . The highly s i g n i f i c a n t subtest differences as w e l l as the predominance of associative d i s t r a c t o r choices on the D subtest found among long-term patients, both psychiatric (regardless of the schizophrenia/nonschizophrenia dichotomy) and nonpsychiatric, seem to e s s e n t i a l l y support the contention that chronicity and the associated lengthy i n s t i t u t i o n a l i z a t i o n may account for increased responsiveness to di s t r a c t o r s and for the r e s u l t i n g d i f -f e r e n t i a l performance on the two matched subtests more meaningfully than does c l i n i c a l diagnosis. In addition, the raw accuracy scores were analysed i n a 3 x 2 x 2 41 Groups (G) x Length of hospitalization (L) x Subtests (T) repeated-measures fa c t o r i a l analysis of variance which permitted the assess-ment of main effects due to diagnostic groups and length of hospi-talization as well as the groups x hospitalization interaction for the between-group comparisons. The within-groups portion of variance allowed for the assessment of the main effect due to the subtest, and for the subtests x groups, the subtests x hospitalization, and the subtests x groups x hospitalization interactions. Data analysis was carried out by means of the computerized program UBC BMDP 2 V - Repeated measures analysis of variance with covariates . The results are shown in Table 5. A l l three between group effects reached s t a t i s t i c a l significance: the groups main effect (F = 8.25, df = 2/78) was significant beyond the .001 level of confidence whereas length of hospitalization (F = 5.14, df = 1/78) and the interaction (F = 4.48, df = 2/78) were significant at the .05 level. Two effects in the within-group variance analysis were found significant: The main effect for the subtests and the subtests x *Adapted by the UBC Computing Centre from the Health Sciences Computing F a c i l i t y , University of California, Los Angeles BMD documentation in March, 1974. This program performs an analysis of variance or an analysis of covariance for a general repeated measures model. The treatment part of the model must be completely balanced and fa c t o r i a l ; the group part can have any hierarchical form of nesting with unequal sizes. In the present study, for each subject, the repeated measures factor had a complete factorial structure with no missing observations, a l l c e l l sizes being equal (n=14). Between subjects responses were assumed to be inde-pendent; whereas within subjects responses were not (according to the program description, they need not be independent), and a l l factors, ex-cept subjects, were assumed fixed. 42 TABLE 5 Summary of the Analysis of Variance of the RC Multiple-Choice Vocabulary Subtest Scores Source SS_ 'df MS F P_ G 3499.34 2 1749.67 8.25 .001 L 1090.33 1 1090.33 5.14 .026 G x L 1898.46 2 949.23 4.48 .014 error 16523.72 78 211.84 T 806.09 1 806.09 35.91 .0001 T x G 31.86 2 15.93 .71 .495 T x L 123.42 1 123.42 5.49 .022 T x G x L 76.10 2 38.05 1.69 .190 error 1750.48 78 22.44 h o s p i t a l i z a t i o n i n t e r a c t i o n . A highly s i g n i f i c a n t subtest main effect (F=35.91, df = 1/78, p < .001) indicated, as expected, that the average number of correct (accurate) test choices produced by the subjects on the ND and D subtests dif f e r e d markedly (the o v e r a l l means were X = 32.68 (ND) and X = 28.30 (D) . Inspection of the means (Table 2) shows a decreased number of correct answers (reduced accuracy) on the D sub-test f o r a l l subgroups except f o r the subgroup composed of short-term nonpsychiatric patients. The subtest x length of h o s p i t a l i z a t i o n i n t e r -action was s i g n i f i c a n t at the .05 l e v e l of confidence (F = 5.49, df = 1/78), suggesting subgroups differences as a function of duration of i n s t i t u t i o n a l i z a t i o n (and, possibly, c h r o n i c i t y ) . In view of the two s i g n i f i c a n t i n t e r a c t i o n s , tests of the s i g n i f i -cance of the component effects were also carried out. The results of these analyses are summarized i n Table 6 and graphically represented for the GxL in t e r a c t i o n i n Figs. 3 & 4. The non-significant but in t e r e s t i n g t r i p l e i n t e r a c t i o n i s i l l u s t r a t e d i n Figs. 1 & 2. I t i s evident (a) that 43 c CJ e cn u CJ CO c CO o o o 40 30 20 -I ;ND D F=1.69 p<-190 NS groups N P Fig. 1. • D and ND RC Vocabulary Accuracy as a Function of Patient Group and Length of Hospita l i z a t i o n a a) e CO CJ CO c a) 4-1 o CJ 1-4 o CJ 40 -J 30 4 20 A ND F-1.69 p<-190 "ST LT 7 length of h o s p i t a l i z a t i o n Fig. 2. D and ND RC Vocabulary Accuracy as a Function of Patient and Length of Hospita l i z a t i o n 44 S NS NP groups Fi g . 3. Combined D and ND RC Vocabulary-Accuracy as a Function of Patient Group and Length of Hos p i t a l i z a t i o n ST LT length of h o s p i t a l i z a t i o n Fig. 4. Combined D and ND RC Vocabulary Accuracy as a Function of Patient Group and Length of Hos p i t a l i z a t i o n 45 TABLE 6 Summary of the Analysis of Variance of the RC Multiple-Choice Vocabulary Subtests and Simple Main Effects Analyses for the G x L and T x L Interactions Source SS df MS F 2. G at 1 3432.66 2 1716.33 8.10 .01 G at 1 2 1965.16 2 .982.58 4.63 .05 L at g 1783.39 1 1783.39 8.41 .01 L at gj 252.87 1 252.87 1.19 L at g 3 952.87 1 952.87 4.49 .05 error term = 211.84 T at l x 149.32 1 149.32 6.65 .05 T at 1 2 780.19 1 780.19 34.76 .001 error term = 22.44 L at ti 240.05 1 240.05 2.04 L at t2 973.76 1 973.76 8.31 .01 error term = 117.14 the highly significant difference (p < .01) in test performance among patients in the short-term hospitalization groups i s largely the result of the performance of the nonpsychiatric patients: these short-term nonpsychiatric patients were more accurate on the test than both the short-term schizophrenic and nonschizophrenic patients; surprisingly, nonschizophrenics were even less accurate than schizophrenics. On the other hand, the long-term schizophrenics appear to account for the significant (p < .05) difference on test accuracy among the long-term subjects. Length of institutionalization appears to be a crucial variable accounting for the decreased test accuracy scores not only for schizophrenic (p < .01) but also for chronic medical patients^ (p < .05), t h o u g h apparently leaving nonschizophrenics relatively un-affected, (b) A highly significant (p < .001) difference in subtest performance was found among the long-term patients while the patients 46 of short-term hospitalization differed to a lesser extent (p < .05) in the number of correctly answered items on the D as opposed to the ND subtest. Markedly lower accuracy scores achieved by the long-term subjects can evidently be attributed to diff e r e n t i a l performance on the D subtest (p < .01) because the contribution of the ND subtest is minor. Graphic representation of the T x L interaction i s pre-sented in Figures 5 and 6> Vocabulary Performance The relative magnitude of variation in the subjects' present intellectual functioning in terms of psychiatric or medical c l a s s i f i -cation and length of hospitalization was assessed by means of two-way analysis of variance. Table 7 presents means and standard deviations of the WAIS vocabulary subtest for a l l three groups of patients at the two levels of hospitalization. It is apparent that the attempt to match the samples on current intellectual functioning was not successful. The raw accuracy scores achieved on the WAIS vocabulary scale by a l l sub-jects were subjected to a 2 x 3 fact o r i a l analysis of variance. As shown i n Table 8, a l l three effects attained s t a t i s t i c a l significance: TABLE 7 Means and Standard Deviations of the WAIS Vocabulary Raw Scores for a l l Three Groups at the Two Levels of Hospitalization Short-term Long-term X SD X SD Schizophrenics 47.78 14.54 Nonschizophrenics 41.21 15.96 Nonpsychiatric subjects 59.57 11.78 34.28 9.81 44.42 12.73 43.85 12.84 47 40 C cd 30 CO u 0) w C cd 4J O o o o o O O O « 1 •> o o o o « o a e o « a t « o o 4. Confusion . , . . 5o Paranoid ideation » 84 Scoring Sg^ft^ To be considered part of the scMrophre>.i:?.c group;, the patieafc ymst scosre on either Item 1 os Iteas 2a«, 2b» 2c end must ettain a total score of at least 4 points,, Ha can achieve a maxlmutn of 4 points on Item Is 2 for the presence of delusions, 2 xor haXlucicacloiis. On Item 2 S ha can score 2 points for any of the symptoms a through c 1 point for either or both sysnpfcosas d through e, and 1 point each for £ and He can thus score a niaxiHHSu of 5 points cm Item 2. Items 3 9 49 5 and 6 each receive 1 point. NOTSs Where 4 t h point necessary for lacluslw* i n ths seisple i s provided by 2d or 2e„ these sysaptcasa ere not scored. 85 Pafinielons Delusions; false beliefs out of keeping with the individual's level of kncrc-r-ledge and cultural background. They may be viewed ca a vaan.ifestat.ion of the individual's misinterpretation or misconception regarding the self or the external world. Hallucinations? a issnif^station ox en individual's aisperceptioa of the self or world, a false sensory perception i n the absence of actual external stimuli. Biasrre or idiosyncratic thinking: Inconsistent, confused or contradictory sta-tement which do not make sense,, statements which dramatically demonstrate an unexplained gap i n the reasoning process 5 logically coherent statement** "which are «arelated to the topic at head, strange or socially inappropriate statements 9 etc. Bisarre thoughts refer to the fona that thoughts take when they are stre s s e d . Avtiss or grossly unrealistic private thoughts; interpretation of outer r e a l i t y a:id inner thoughts and the tendency not to distinguish between the tiro. Objectified facts become obscured, distorted or excluded i n varying degrees and there i s a detachment froa re a l i t y with the relative predominance of inner l i f e . Leoeenesa of association, i l l o g i c a l thinking; refer to the content and form of ik person's expressed thoughts. Thoughts are not continuous, thinking is i l l o g i c a l j, not oriented to any goal and operates with ideas and concepts which have l i t t l e or no connection with the main idea. Overineluslon: d i f f i c u l t y i n maintaining the usual conceptual boundaries arid a tendency to include many elements Irrelevant to the central idea i n thinking. 8 6 Blockings rafero to the • cr the !:rc ' Coneret^aas: the iosa o? •"„'. too l i t e r a l a thoughts or I.'fv;.;, Derealization: a jrentai ;>*• real i t y c o u c : ; that one's ci:-:*<.i . v i r o r a & m f c io ii. v.--. D e p e r s o n a l i z a t i o n , to vhirl. :'. .• unreality or like and the t?c :*!<•.•; Inappropriate affects a Xn~l. < of affect and inrpf' Confusion: a d i s t u r b e d x>r.l'-(•.'.••-Paranoid ideations the preij'.. i especially i n c l u t l . ' . . ..::!::w.oxt i n the flow of thought or speech, n traia of thought due to esatioral factors. •'.!} think i n P.Lstract concepts; reacting i n . considering the relationships of particular characterised b y the loss o f the sense of i n relationship to one's surroundings, have ao'idiot? changed so that the real en-•/ay no longer what i t used to be. • -iraSlon i s cluaely related, i s a sense of „':. frora oneself. Everything seea3 dream-o n e s e l f or others are matched with detachment. • ' and consioteacy of affect, a l a b i l i t y jvir.to affect, '. ;>dth respect to either p e r s o n , place or t i m e , p a r a n o i d thoughts, other than delusions, Mjoieiousnaaa oad oelf-rafarential t h i n k i n g . Subject 1 APPENDIX 2 Patient Drug Status and Raw Scores Obtained on the RC and WAIS Vocabulary Subtests a) Short-term Schizophrenic Subjects (n=14; medicated 14) Diagnosis , Sex RC Scores WAIS Medication(s) Daily 6 7 8 9 10 11 Paranoid Schizophrenia Schiz.reaction, simple Schiz. reaction, simple Schiz. reaction, simple Chronic,undif.schizophrenia Chronic, undif.schizophrenia Chronic schizophrenia Schiz. reaction, simple Paranoid schizophrenia Paranoid schizophrenia ND D Vocabulary Amount(mg.) male 35 19 33 Chlorpromazine 150 Haloperidol 20 Benzohexol 6 ' male 53 50 72 Chi or promazine 500 Trifluoperazine 15 male 31 27 43 Chior promazine 300 Haloperidol 10 Benztropine 2 male 29 30 34 Thioridazine 500 Benztropine 3 male 38 31 64 Chlorpromazine 300 Haloperidol 10 female 20 18 39 Thioridazine 200 male 30 34 40 Thioridazine 600 male 49 39 65 Thioridazine 900 male 39 32 35 Chlorpromazine 400 Trifluoperazine 15 Procyclidine 15 male 29 11 52 Chlorpromazine 400 Procyclidine 15 female 27 21 66 Penfluridol 20 Note (1 x weekly) •a) Short-term Schizophrenic Subjects (n=14; medicated 14) (con't) Subject Diagnosis Sex RC Scores WAIS Medication(s) Daily ND D Vocabulary Amount (mg.) 12 Acute schiz. reaction male 44 48 56 Thioridazine 300 Amitriptyline 150 Procyclidine 10 13 Acute schiz. reaction male 39 39 43 Thioridazine 600 14 Acute schiz. reaction male 31 21 27 Chlorpromazine 300 Haloperidol 20 Benztropine 2 Flurozepan 45 Note (fluctuates between 30-60 m G hs) CO co APPENDIX 2 (Con't) b) Long-term Schizophrenic Subjects (n=14; medicated 12) Subject Diagnosis Sex RC Scores WAIS Medication(s) Daily Note ND D Vocabulary Amount (mg.) 1 Chronic Paranoid Schizophrenia Female 30 18 32 Chlorpromazine Haloperidol Procyclidine 600 15 10 2 Paranoid Schizophrenia Female 40 37 56 Trifluoperazine Diazepam Procyclidine 10 10 10 3 Chronic Paranoid Schizophrenia Female 13 11 29 Trifluoperazine Haloperidol Benztropine 20 10 2 4 Chronic Undif .Schizophrenia Female 21 14 40 - -5 Chronic Undif.Schizophrenia Female 31 28 18 Chlopro thixene Thioridazine Benztropine Fluphenazine 100 100 4 25 6 Chronic Schizophrenia Male 19 12 24 Thioridazine Haloperidol Diazepam Benztropine 300 15 15 6 7 Paranoid Schizophrenia Male 29 23 40 - -8 Chronic, Undif. Schizophrenia Male 20 10 28 Chlordiazepoxide 75 9 Chronic, Undif. Schizophrenia Male 31 26 37 Chlordiazepoxide. 75 VO b) Long-term Schizophrenic Subjects (n=14; medicated 12) (con't) Subject Diagnosis 10 Chronic Schiziphrenia 11 Chronic Schizophrenia 12 13 Chronic, Undif. Schizophrenia Chronic, Undif. Schizophrenia Sex RC Scores WAIS Medication(s) Daily ND D Vocabulary Amount (mg.) Note Male Male Male Male 23 14 17 27 20 14 21 14 Chronic Paranoid Schizophrenia Male 20 20 45 29 39 25 38 Thioridazine 300 Benztropine 6 Chlorpromazine 450 Benztropine 4 Fluphenazine 25 Chlo rpromaz ine 400 Benztropine 4 Chlorpromazine 300 Diazepam 30 Amitriptyline 75 Nicotinic Acid 18 Ascorbic Acid 18 Benztropine Haloperidol 10 Benztropine 4 Pericycline 20 Chlorpromazine (1 x weekly) prn prn VO o APPENDIX 2 (con't) c) Short-term Nonschizophrenic Subjects (n=14; medicated 14) Subject Diagnosis 1 Manic-depres. Psychosis 2 Manic-depres. Psychosis 3 Personality Disorder 4 Personality Disorder 5 Depressive Neurosis 6 Personality Disorder & Depres. Reaction 7 Personality Disorder 8 Manic-depres. Psychosis 9 Psychotic Depression Sex Female Female RC Scores ND D 41 39 26 13 Male Male Female 6 10 10 22 9 17 WAIS Medication(s) Daily Vocabulary Amount (mg.) Note 62 41 23 47 16 Trimipramine 200 Lithium carbonate 1200 Chlorpromazine 300 Haloperidol 6 Metatrimipramine 10 Fluphenazine 25 Benztropine 2 Amitriptyline 50 Haloperidol 6 Benztropine 2 Chlorpromazine 200 Chlorpromazine 400 Every 2nd week Every 2nd week Female 26 32 60 Male 25 15 22 Male 23 19 26 Female 45 35 44 Dilantin 300 Phenobarbital 120 Chlorpromazine 100 Haloperidol 12 Chlorpromazine 500 Lithium Carbonate 900 Chlorpromazine 400 Chlordiazepoxide 30 c) Short-term Nonschizophrenic Subjects (n=14: medicated 14) (con't) .Subject Diagnosis Sex RC Scores WAIS Medication(s) Daily .Note ND D Vocabulary Amount (mg.) 10 Psychotic Depression Female 26 19 38 Thioridazine Benztropine Amitriptyline 400 4 100 11 Personality Disorder Female 18 30 30 Chlordiazepoxide Imipramine 15 150 12 Psychotic Depression Female 18 18 59 Mesoridazine Diazepam 75 75 13 Anxiety Neurosis Male 38 30 47 Imipramine Chlordiazepoxide 150 14 Psychotic Depression Female 51 51 62 Thioridazine Chlorpromazine Benztropine Me tho t r imep r az one 400 400 8 50 VO r-o APPENDIX 2 (con't) d) Long-term Nonschizophrenic Subjects (n=14; on psychoactive drugs 11) • Subject Diagnosis Sex RC Scores WAIS Medication(s) Daily ND D Vocabulary Amount (mg.) Note 1 Manic-depres. Psychosis (c) Female 38 33 64 Lithium Carbonate 900 Haloperidol 20 Benztropine 4 2 Manic-depres. Psychosis (c) Female 17 15 48 Lithium Carbonate 1200 Promethazine 100 Chlorpromazine 75 prn Benztropine 2 prn 3 Manic-depres. Psychosis Female 12 11 28 Potassium Chloride 900 Furosemide 40 4 Manic-depres. Psychosis (c) Female 44 22 48 Lithium Carbonate 900 Haloperidol 4 Benztropine 1 5 Involutional Psychosis Female 39 30 31 Chlorpromazine 50 Furosemide 40 6 Manic-depres. Psychosis 7 Manic-depres. Psychosis Female 38 18 Female 15 11 8 Manic-depres. Psychosis (c) Female 43 44 45 Amitriptyline 200 Thioridazine 200 Benztropine 2 Chlorprothixene 25 18 Lithium Carbonate 1200 Chlorprothixene 300 Benztropine 4 Chlorpromazine 100 49 Lithium Carbonate 600 Thioridazine 150 Amitriptyline 100 Benztropine 2 prn prn d) Long-term Nonschizophrenic Subjects (n=14; on psychoactive drugs 11) (Con't) Subjects Diagnosis Sex RC Scores WAIS Medication(s) Daily Note ND D Vocabulary Amount (mg.) 9 Manic-depres. Psychosis Female 38 34 41 Diazepam 10 Amitriptyline 55 10 Manic-depres. Psychosis Male 42 43 59 -11 Manic-depres. Psychosis Male 33 33 50 Fluphenazine 2 Benztropine 2 12 Psychotic Depression Female 33 15 40 Phenytoin 400 Felic Acid 10 Chloral Hydrate 2 hs 13 Manic-depres. Psychosis Female 43 36 60 Lithium Carbonate 900 Chlorpromazine 300 14 Manic-depres. Psychosis Male 28 23 41 Chlorpromazine 400 i vO APPENDIX 2. (con't) e) Short-term Nonpsychiatric Subjects (n=14; none on psychoactive drugs) Subjects Diagnosis Sex RC Scores WAIS Medication(s) Daily ND N Vocabulary Amount (Mg 1 Multiple sclerosis Female 20 31 53 - -2 Glaucoma Male 52 50 72 — -3 Hemiplegia Male 39 36 40 - -4 Multiple Sclerosis Female 47 49 71 - -5 Multiple Sclerosis Male 50 51 69 - -6 Hemiplegia Male 51 40 39 - -7 Tuberculosis Male 58 52 73; 8 Tuberculosis Male 54 48 68; 9 Tuberculosis Male 36 35 45; 10 Tuberculosis Male 33 33 62; Antimicrobial Therapy 11 Tuberculosis Male 9 14 43j 12 Tuberculosis Male 52 39 1 37; 13 Tuberculosis Male 54 54 69! 14 Tuberculosis Male 66 44 46* APPENDIX 2 (con't) f) Long-term Nonpsychiatric Subjects (n=14; on psychoactive drugs 8) Subjects Diagnosis Sex RC Scores WAIS Medication(s) Daily (mg.) Note ND D Vocabulary Amount(mg.) . 1 Multiple Sclerosis Female 33 19 45 Diazepam 20 2 Quadriplegia Male 31 33 57 Diazepam 20 3 Rheumatoid Ar t h r i t i s Female 41 .42 36 - -4 Quadriplegia Female 51 552 62 Dilantin Diazepam Hydrochlorthiazide 400 20 50 5 Rheumatoid A r t h r i t i s Female 39 14 36 Diazepam Tofranil 5 10 6 Rheumatoid Ar t h r i t i s Female 44 ;44 70 - -7 Quadriplegia Male 37 37 30 Diazepam 20 8 Multiple Sclerosis Male 32 11 22 Thioridazine Diazepam 50 10 9 Multiple Sclerosis Male 46 40 53 - -10 Multiple Sclerosis Male 28 27 40 - -11 Multiple Sclerosis Male 25 26 40 - -12 Multiple Sclerosis Male 30 20 44 Diazepam 10 13 Quadriplegia Urosepsis Male 31 26 37 Diazepam 45 14 Hemiplegia Female 36 34 42 - -ON APPENDIX 3 A Posteriori Pairwise Contrasts Among Means on the Age Variable Xi X 2 x 3 x 4 x 5 x 6 s ST X i = 28.42 6.36 17.65* 18.29* 24.22** 32.22** NS ST X 2 = 34.78 - 11.29 11.93 17.86* 25.86** s LT X i = 46.07 - .64 6.57 14.57 • NP ST X 4 = 46.71 - 5.93 13.93 NP LT x 5 = 52.64 - 8.00 NS LT x 6 = 60.64 _ _ V - 0 5 **p<.01 Tukey HSD = q , / MS error \x,k,v = 4.15 /"24~7.65 14 =17.45; (20.79) Level of Significance a = .05; (.01) Number of Means k =6 Degrees of freedom for MSe v = 78 The value of q was obtained from the distribution of the studentized range s t a t i s t i c (Table D7 in Kirk, R.D., Experimental Design: Procedures for the Behavioral Sciences, Belmont, Cal i f . : Brooks/Cole, 1968. 98 APPENDIX 4 X l x 2 x 3 Xit x 5 x 5 s LT X i = 8.71 - .50 .93 1.07 1.64 1.71 NP LT X 2 = 9.21 .43 .57 1.14 1.21 NS ST x 3 = 9.64 - .14 .71 .78 NS LT = 9.78 - .57 .64 S ST x 5 = 10.35 - .07 NP ST x 6 = 10.42 — Tukey HSD = q Ct j tc j V = 4.15 P>.05 J MS error 2.99 14 = 1.91 Level of significance a = .05 Number of means k = 6 Degrees of freedom for MSe v = 78 c f . footnote in Appendix 3 99 APPENDIX 5 A Posteriori Comparisons among Means on Length of Hospitalization Variable a) Short-term Hospitalization x l x 2 x 3 S X 1 = .85 - -75 2.72** NS X2= 1.60 - 1.97** NP X3= 3.57 **p<.01 Tukey HSD - q , /MSe - 3.44 14 = 1.46; (1.83) b) Long-term Hospitalization Xi X 2 X 3 NP X1 = 69.21 - 26.79 70.71 NS X 2 = 96.00 - 43.92 S X 3 = 139.92 p>.05 Tukey HSD = q / ^ e a>K>v _n - 3.44 / 1 2 4 2 9 - 2 1 14 = 102.49 Level of significance = .05; (.01) Number of means k = 6 Degrees of freedom for MSe =78 c.f. footnote in Appendix 3 100 APPENDIX 6 Author's Form for Recording Basic Personal, Demographic, Diagnostic, and Medication Data Name Date of birth Sex day month male female Last grade completed 6 7 8 college 1 2 Marital status single married (duration; frequency i f applicable Diagnosis Date year 10 4 11 12 degree separated divorced widowed Date of admission Number of admissions Therapy JLength of hospitalization months Overall length of hospitalization _years D r u g Amount per day Started date Discontinued Comment