• 沒有找到結果。

Psychopathological Dimensions in Schizophrenia: A Correlational Approach to Items of the Sans and Saps

N/A
N/A
Protected

Academic year: 2021

Share "Psychopathological Dimensions in Schizophrenia: A Correlational Approach to Items of the Sans and Saps"

Copied!
10
0
0

加載中.... (立即查看全文)

全文

(1)

Psychopathological dimensions in schizophrenia: a

correlational approach to items of the SANS and SAPS

Albert Shi-Kwang Lin

a

, Chun-Houh Chen

b

, Hai-Gwo Hwu

c,U

, Hsin-Nan Lin

c

,

Jih-An Chen

d

a

National Taiwan Uni®ersity, College of Medicine, Department of Psychiatry, 7 Chung-Shan South Road, Taipei, Taiwan 100, People’s Republic of China

b

Institute of Statistical Science, Academia Sinica, Taipei Taiwan, People’s Republic of China

c

Department of Psychiatry, College of Medicine, National Taiwan Uni®ersity, Taipei Taiwan, People’s Republic of China

d

Institute of Mathematical Statistics, National Chung-Cheng Uni®ersity, Taipei Taiwan, People’s Republic of China

Received 22 April 1997; revised 1 May 1997; accepted 9 October 1997

Abstract

Seventy DSM-III schizophrenic patients were assessed for positive and negative symptoms using Andreasen’s

Ž .

scales for the assessment of positive and negative symptoms SANS and SAPS on admission. The correlation structure of the items in the SANS and SAPS was explored in dimension and item levels by use of correlation plots through a distinct analytical method displaying the proximity matrix. The results revealed at least three major dimensions of symptoms delineated as Negative Symptoms, Disorganized Thoughts and Delusions and Hallucina-tions. The latter two dimensions were derived from the SAPS, while Negative Symptoms comprised most of the items in the SANS. Items in Disorganized Thoughts were more correlated to Negative Symptoms than to the other items in the SAPS. ‘Loss of ego boundary’ delusions and experience of auditory hallucinations appeared as two sub-clusters in the group of Delusions and Hallucinations. The relative independence of persecutory, grandiose, religious, somatic and reference delusions gives support to the concept that paranoid schizophrenia stands as a distinct clinical subtype of schizophrenia. The graphical method introduced here well expresses the information of correlation matrix and is useful for exploring inter-item or inter-cluster associations. Q 1998 Elsevier Science Ireland Ltd.

Keywords: Positive symptoms; Negative symptoms; Disorganized thoughts; Correlation

UCorresponding author. Tel.: q886 2 3970800, ext. 6785; fax: q886 2 3753663; e-mail: haigohwu@ha.mc.ntu.edu.tw

0165-1781r98r$19.00 Q 1998 Elsevier Science Ireland Ltd. All rights reserved. Ž .

(2)

1. Introduction

Various rating scales for measuring positive andror negative symptoms have been proposed

Ž

since the work of several groups Strauss et al., 1974; Crow, 1980, 1985; Andreasen and Olsen,

.

1982; Andreasen et al., 1982 led to extensive interest in this dichotomy of schizophrenic

symp-Ž

tomatology. A series of reports Arndt et al., 1991; Minas et al., 1992, 1994; Malla et al., 1993; Klimidis et al., 1993; Andreasen et al., 1995;

Stu-. w

art et al., 1995 using the SANS and SAPS Scale for the Assessment of Negative Symptoms ŽAndreasen, 1983 ; Scale for the Assessment of.

Ž .x

Positive Symptoms Andreasen, 1984 and some Ž

using other psychometric scales Liddle, 1987; Kay and Sevy, 1990; Sarai and Matsunaga, 1993;

.

Lindenmayer et al., 1995b indicated that, in con-trast to the conventional positive and negative syndromes, an independent third dimension of thought disorder or disorganization syndrome may be an underlying form of symptomatology in schizophrenia.

The major analytic strategies in the literature to explore underlying dimensions of measured psychopathology usually involve multidimensional

Ž . Ž .

scaling MDS or factor analysis FA techniques. Both approaches address issues of the empirical cohesiveness of sets of symptoms and the inde-pendence of dimensions. However, the approach of analysis of ratings is crucial. There is a sub-stantial risk of obliterating the relationships between individual symptoms, or even subgroups

Ž

of symptoms, and external validators e.g. illness course, treatment outcomes and biological

corre-.

lates if the composite of scale items is segregated and grouped inappropriately. Some researchers employed a global or subscale rating approach ŽArndt et al., 1991; Sarai and Matsunaga, 1993;

.

Klimidis et al., 1993 and others applied an ap-Ž

proach to individual items Liddle, 1987; Minas et al., 1992, 1994; Malla et al., 1993; Stuart et al.,

.

1995 . Several investigators indicated that differ-Ž ent approaches had distinct implications Minas

.

et al., 1992, 1994; Stuart et al., 1995 . Whether the strategy of analysis would have influence on the measured profile of psychopathological di-mensions needs further attention.

The SANS and SAPS are widely known for

their high reliability and used to demarcate the dimensions of schizophrenic symptoms. However, questions have been raised as to the validity of their internal construct and subgroup

composi-Ž

tion Minas et al., 1992, 1994; De Leon et al., 1993; Klimidis et al., 1993; Malla et al., 1993;

.

Sarai and Matsunaga, 1993 . Given this scenario, the strategy of an item-level approach for depict-ing the more homogenous clusters of psy-chopathology in schizophrenia appears important. The aim of this study was to explore the major and sub-cluster structures that delineate the psy-chopathological dimensions of the SANS and SAPS items in a sample of Chinese schizophrenic patients. We used correlation plots with an item-level approach to reveal clustering of the SANS and SAPS items through a new graphical method. 2. Methods

2.1. Subjects and clinical rating

Ž .

Seventy patients 41 male; 29 female with

Ž .

DSM-III American Psychiatric Association, 1980 diagnoses of schizophrenia were recruited after giving written consent. Patients ranged in age

Ž

between 16 and 54 years mean age s 28.2; .

S.D.s 7.0 ; mean duration of illness was 42.8

Ž .

months S.D.s 50.7 . The subjects were all expe-riencing a significant array of psychotic symptoms when admitted to the acute inpatient wards of two university-affiliated hospitals.

Clinical diagnoses were assessed with a semi-structured psychiatric diagnostic assessment

in-Ž .

terview schedule as described by Hwu et al. 1986 and supplemented with clinical data from the medical charts. Patients with the DSM-III diag-noses of organic mental disorders and mental retardation were excluded from the study. The SANS and SAPS were completed by the end of the first week after admission by the attending psychiatrist. All of the subjects were receiving antipsychotic treatment when assessed. The raters were familiar with these measuring scales. The inter-rater reliabilities were between 0.82 and 0.92 for the SAPS and between 0.73 and 0.88 for the SANS as determined by the Spearman rank corre-lation. Items of low base rate, i.e. frequency of

(3)

Ž

occurrence ratings scored G 2 in the SANS or .

SAPS less than 10%, were excluded, yielding 43

Ž .

items for analysis Table 1 .

2.2. Statistical and graphical method

Ž .

We used generalized association plots GAP wChen 1996 : summarized PC program availableŽ .

x

on request , a graphical environment for general purpose multivariate analyses, as the major ana-lytical tool. This approach takes advantage of the computer’s superior computing power and graph-ing ability to retrieve important pieces of infor-mation embedded in all different kinds of raw data matrices and proximity matrices.

In our study, a Pearson’s product-moment cor-relation matrix of the 43 symptoms was used as the input proximity matrix to the GAP program. Given a proximity matrix, GAP first represents

Ž

the matrix graphically as a color map correlation .

plot with variables in the matrix sorted by a

Ž .

random order Fig. 1a or by a user specified

Ž .

order Fig. 1b . Each of the proximity measures Žcorrelation coefficients in the matrix is repre-. sented by a colored dot. For our correlation ma-trix, the blue]red color spectrum with 100 levels was employed to illustrate the bi-directional na-ture of the correlation coefficients.

In Fig. 1a, the randomly-sorted correlation plot does not give us much more information than the usual correlation matrix of numbers, while the overall pattern in Fig. 1b presents a somewhat clearer clustering effect. Delineating an overall profile of the symptom structure, Fig. 1b can be further sorted to study the within- and between-group symptom relations in a more revealing manner.

3. Results

3.1. Tree diagram

Among many sorting algorithms from the GAP

Table 1

a Forty-three SAPS and SANS items included in analysis

SAPS SANS

AH1 Auditory hallucinations NA1 Unchanging facial expression

AH2 Voices commenting NA2 Decreased spontaneous movements

AH3 Voices conversing NA3 Paucity of expressive gestures

VH Visual hallucinations NA4 Poor eye contact

DL1 Persecutory delusions NA5 Affective non-responsivity

DL4 Grandiose delusions NA6 Inappropriate affect

DL5 Religious delusions NA7 Lack of vocal inflections

DL6 Somatic delusions NB1 Poverty of speech

DL7 Ideas and delusions of reference NB2 Poverty of content of speech

DL8 Delusions of being controlled NB3 Blocking

DL9 Delusions of mind reading NB4 Increased latency of response

DL10 Thought broadcasting NC1 Grooming and hygiene

DL11 Thought insertion NC2 Impersistence at work or school

DL12 Thought withdrawal NC3 Physical anergia

BEH1 Clothing and appearance ND1 Recreational interest and activities BEH2 Social and sexual behavior ND2 Sexual interest and activity

BEH3 Aggressive and agitated behavior ND3 Ability to feel intimacy and closeness

TH1 Derailment ND4 Relation with friends and peers

TH2 Tangentiality NE1 Social inattentiveness

TH3 Incoherence NE2 Inattentiveness during MSE

TH4 Illogicality TH5 Circumstantiality TH7 Distractible speech aSeven items excluded due to low base rates.

(4)

Ž .

Fig. 1. Correlation plots. The numeric correlation coefficients y1 to 1 are illustrated in a blue]red color spectrum of 100 block

Ž . Ž .

levels. a Symptoms are sorted by a random order. b Symptoms are sorted by the order as the Andreasen’s SANS and SAPS table. Ž .

Fig. 2. Divisive clustering tree with sorted correlation plot. a Number in the horizontal scale represents the average correlation Ž .

coefficient for the submatrix under each node. b The sequence of symptoms in the correlation plot is sorted according to the order of the items determined by the terminal nodes in the clustering tree.

(5)

program, the tree diagram can provide us, in addition to the sorted correlation plot, further information for clustering the symptoms. The cor-relation matrix of 43 symptoms was decomposed into two disjoint sub-matrices of 26 and 17 items according to the converging property of the itera-tively formed sequence of correlation matrices wsee also McQuitty 1968 and Breiger et al. 1975Ž . Ž . for earlier applications of the dichotomous con-verging pattern of the sequence of correlation

x

matrices . This process of decomposition contin-ues until each terminal submatrix contains only

Ž .

one symptom item. The tree diagram Fig. 2a illustrates this recursive partitioning of the sub-matrix sequence.

The branches of the tree diagram are then permuted according to the relative mean levels of

Ž .

between-group submatrices correlation so that symptom groups with higher means of between-group correlation stay close to each other in the clustering tree diagram. The horizontal scale in the tree diagram indicates the average correlation coefficient for the submatrix under each corre-sponding node.

3.2. Correlation plots

The correlation map sorted by the order of the terminal nodes in the tree diagram is then plotted as Fig. 2b. Fig. 3 further reveals the relations among purported items or clusters by excluding the items within indicated ranges of correlations. It is apparent that the SANS and SAPS items do not merge into two discrete, coherent groups that can adequately delineate the structure of the correlation matrix. Instead at least three major symptom clusters can be delineated here as Nega-tive Symptoms, Disorganized Thoughts and Delu-sions and Hallucinations. The Negative Symptoms comprise most of the items in the SANS. The items of the SANS clearly form a more coherent

Ž .

group than do those of the SAPS. Blocking NB3

Ž .

and increased latency of response NB4 , though closely correlated to each other, show different patterns of relation to other items in the SANS. NB4, unlike NB3, shows fair correlation to other items of Negative Symptoms. However, NB3, but not NB4, has moderate correlation with several items of Disorganized Thoughts and ‘loss of ego

Ž

boundary’ delusions i.e. thought withdrawal, thought broadcasting, delusions of being con-trolled, delusions of mind reading and thought

. insertion .

A second cluster, Disorganized Thoughts, in-cludes derailment, tangentiality, incoherence, il-logicality and less coherent, distractible speech. Lying in between are several items

simultane-Ž .

ously correlated most of them loosely with Neg-ative Symptoms and Disorganized Thoughts, such

Ž .

as clothing and appearance BEH1 , social and

Ž . Ž .

sexual behavior BEH2 , circumstantiality TH5 ,

Ž .

inappropriate affect NA6 and items of

attentio-Ž .

nal problems NE1, NE2 . With little correlation to most of the other items in the plot, grandiose

Ž . Ž .

delusions DL4 and religious delusions DL5 lie in between the Disorganized Thoughts and an-other major group.

The third major dimension, Delusions and Hal-lucinations, contains two moderately correlated subgroups, namely, delusions of ‘loss of ego boundary’ and items of auditory hallucinatory

ex-Ž

periences i.e. voices conversing, voices comment-.

ing and auditory hallucinations . Aggressive and

Ž .

agitated behavior BEH3 and visual

hallucina-Ž .

tions VH had mildly negative correlations to Negative Symptoms. Like items of grandiose delusions and religious delusions, persecutory

Ž . Ž .

delusions DL1 , somatic delusions DL6 and

Ž .

reference delusions DL7 manifest relative inde-pendence of other items of the SANS and SAPS. 4. Discussion

We present a graphical tool, the correlation plot, to illustrate the major dimensions and sub-dimensions of the SANS and SAPS items. The analytical strategy we used has distinct implica-tions in exploring the dimensions of schizophrenic psychopathology as measured by the SANS and SAPS.

Factor analysis and MDS are the most common techniques used in the literature to study the structure of correlation matrices on schizophrenic

Ž

psychopathology Minas et al., 1992; Klimidis et al., 1993; Andreasen et al., 1995; Stuart et al.,

.

1995 . These statistical strategies are dimension reduction approaches which approximate the structure of correlation matrices by low

(6)
(7)

dimensio-nal spatial coordinates. Because of approxima-tion, both methods suffer from loss of

informa-Ž

tion contained in the original proximity correla-.

tion matrices. FA uses one minus the total vari-ance explained by the leading factors with large

Ž .

eigenvalues usually greater than one , while MDS employs the stress score to represent this loss of

Ž

information by approximation dimension

reduc-. Ž .

tion . To achieve a low usually two dimensional

Ž 2 .

result with low stress score or high R score , MDS usually needs to exclude some objects Žsymptoms of low commonalities with other ob-.

Ž .

jects from the analysis Minas et al., 1992 . For FA, relevant information may be disregarded in the factors with small eigenvalues, although FA does not need to exclude the low commonality objects in advance.

The sorted correlation plot we used has some advantages over FA and MDS. First, the correla-tion plot requires only a 2-dimensional color map with blocks of dark red points on the main-diago-nal to reveal all the possible equivalent factor loadings in FA and groups of symptoms in MDS. Second, the blue points in a correlation plot indi-cate symptoms with negative correlations, while two far-separated symptoms in an MDS plot do not necessarily have negative correlation. Third, the correlation plot retains each piece of correla-tion in the matrix as a colored point. The user has easy access to all the information of correlation in the plot. Fourth, the sequence of clustering of the symptoms was revealed by the tree diagram. Vari-ous seriation algorithms may find slightly differ-ent permutations of the correlation matrix. As long as the permuted seriation is not much dif-ferent from the presumably best one in all 43 possible seriations, one should be able to recog-nize most of the correlation information con-tained in the original matrix by examining the sorted correlation matrix map. One of the major advantages of the GAP environment is that every single correlation coefficient from the input prox-imity matrix is maintained in the output sorted proximity matrix map. We think that the

correla-tion plot can be used to complement, not to replace, the existing multivariate tools in per-forming a more comprehensive data analysis.

Our data revealed that at least three major dimensions of symptoms were measured by the SANS and SAPS, i.e. Negative Symptoms, Disor-ganized Thoughts and Delusions and Hallucina-tions. Our results support a valid distinction between Delusions and Hallucinations and Nega-tive Symptoms, both representing discrete, but not co-exclusive or inversely related, dimensions of schizophrenic symptoms. Meanwhile, in

paral-Ž

lel with several previous reports Minas et al., . 1992, 1994; Malla et al., 1993; Stuart et al., 1995 , our findings showed that the simple dichotomy of positive and negative syndromes does not suffice to demarcate adequate profiles of psycho-pathology in schizophrenia.

Another main finding is that Disorganized Thoughts may well be independent from Negative Symptoms and Delusions and Hallucinations. This has been shown through various lines of research:

Ž

those employing the SANS and SAPS Minas et al., 1992, 1994; Malla et al., 1993; Klimidis et al., . 1993; Andreasen et al., 1995; Stuart et al., 1995 ,

Ž

those with other scales Liddle, 1987; Sarai and .

Matsunaga, 1993 and one with a different ethnic

Ž .

sample group Kulhara et al., 1986 . As reflected in our results and these reports, it seems that differences in ethnicity and cultural group of patients do not bring about a significant discrep-ancy in the dimensionality. Furthermore, given the notion that our patients were suffering from schizophrenia in an acute or exacerbated phase, the segregation of the disorganization syndrome appeared not to differ from that in the samples of

Ž

chronic cases Liddle, 1987; Sarai and Matsunaga, .

1993 .

In the present study, Disorganized Thoughts are composed of items from the SAPS. Interest-ingly, the items associated with Disorganized Thoughts are more positively correlated with Negative Symptoms than with Delusions and

Hal-Ž .

lucinations. Fenton and McGlashan 1994

indi-Fig. 3. The sectional correlation plots in series. The blank bars indicate the ranges of correlations within which corresponding items were excluded.

(8)

cated that, over the course of many years,

symp-Ž .

toms of thought disorder and bizarre behavior might increase among patients who also have prominent persistent negative symptoms. It is necessary to accommodate the syndromal inde-pendence of disorganized thoughts and observe its long term relation with Negative Symptoms in future studies.

The items of the SAPS appeared not a very coherent group as we examined the structure of the correlation plot. Clearly, Disorganized Thoughts can be demarcated from the other SAPS items. The SAPS includes several Schneiderian first-rank symptoms. Our results revealed that those first-rank symptoms adopted in the SAPS emerged into two moderately correlated

sub-clus-Ž

ters as ‘loss of ego boundary’ delusions i.e. delu-sions of being controlled, deludelu-sions of mind read-ing, thought broadcastread-ing, thought insertion and

.

thought withdrawal and the group of auditory Ž

hallucinatory phenomena i.e. auditory hallucina-. tions, voices commenting and voices conversing , which is consistent with a recent report of Stuart

Ž .

et al. 1995 . Moreover, the evidence that ‘loss of ego boundary’ delusions were correlated only loosely with the other delusions supports the view that they should be treated as distinct from hallu-cinations and other types of delusions. Unlike the other items in ‘loss of ego boundary’ delusions, thought withdrawal shows moderate or fair corre-lation with items of Disorganized Thoughts, which

Ž .

is consistent with the finding of Mellor 1970 . These findings are compatible with the classical

Ž .

description by Carl Schneider Schneider, 1942

Ž .

of a thought withdrawal ‘Gedankenentzug’ group of acute schizophrenics with symptoms including thought withdrawal, verbal derailment, blocking and breaking off of thoughts. The psychopatho-logical implication of the relationship between ‘loss of ego boundary’ delusions and formal thought disturbances is an issue that needs fur-ther investigation.

Furthermore, persecutory, religious, grandiose, reference and somatic delusions showed indepen-dence from other delusions, hallucinations and thought disturbances. Given the empirical im-pression that paranoid schizophrenia is distinct from its non-paranoid counterpart in several

as-Ž .

pects e.g. treatment outcome and prognosis , it is reasonable to consider these paranoid delusions

Ž

as an even further symptom dimension Minas et .

al., 1992; Stuart et al., 1995 . Taken together, our results of relation profile among the items in the SAPS concur with the notions that the SAPS does not assess a unitary construct of ‘positive’ psy-chotic symptoms and that the strategy of taking the SAPS total score as a direct measure of

Ž

‘positive’ symptoms is not appropriate Minas et .

al., 1992; Stuart et al., 1995 . The interrelations among the items of Delusions and Hallucinations in the SAPS need further exploration.

Both the SANS and SAPS have several items in the same subscale and show disparity in the corre-lations with respective symptom subgroups. The bizarre behaviors subscale items of social and

Ž .

sexual behavior BEH2 and clothing and

appear-Ž .

ance BEH1 of the SAPS are more closely re-lated to Negative Symptoms and Disorganized Thoughts than to Delusions and Hallucinations. Meanwhile, aggressive and agitated behavior ŽBEH3 has little relation to the Negative Symp-.

Ž .

toms. Thought blocking NB3 and increased

la-Ž .

tency of response NB4 , both being items of the alogia subscale, differed in their correlations with most other Negative Symptoms, Disorganized Thoughts and several ‘loss of ego boundary’

delu-Ž .

sions Miller et al., 1993 . This may imply that thought blocking should be considered a symptom closer to formal thought disorders than to Nega-tive Symptoms. Furthermore, unlike most other Negative Symptoms to which they are closely

cor-Ž .

related, the items in avolition-apathy NC1]3 , a subscale intended to examine dysfunction in self-care and persistence at work, are also moderately

Ž associated with Disorganized Thoughts Liddle,

.

1987 . Another parallel finding is that, though closely associated with each other, the items of attentional impairment were respectively corre-lated with different subgroups of symptoms. That

Ž .

is, inattentiveness during MSE NE2 was more closely associated with Disorganized Thoughts;

Ž .

social inattentiveness NE1 , with Negative Symp-toms. While usually recognized as significant dys-functions, attentional problems have been allo-cated to various dimensions of schizophrenic symptomatology in different reports. They were

(9)

taken as symptoms of the disorganization

syn-Ž .

drome Liddle, 1987; Sarai and Matsunaga, 1993 , Ž

items of the negative component Andreasen, . 1983; Kay and Sevy, 1990; Malla et al., 1993 , or manifestations secondary to positive features ŽCarpenter et al., 1985 , even as equally presented.

Ž

with negative and positive syndromes Bilder et .

al., 1985; Kay et al., 1986 . This implies that items allocated together in the same subscale by seman-tic presumption may not share common psycho-pathology and may be apparent in different con-texts. Whether they represent separate processes or similar psychopathology mediated by different

Ž

coexisting disturbances e.g. disorganization or .

negative symptoms per se warrants further clari-fication.

This study has inherent limitations. First, the sample size may be relatively small. We suspect that different sample sizes may have some minor changes on the finer structure, but not the overall pattern. Second, our sample consisted of patients in acute phases of illness. However, it would be more informative to further validate these symp-tom dimension entities in specific subgroups of patients for a longer course through different

Ž

phases of illness e.g. acute vs. chronic; exacer-.

bated vs. stable . Third, this study explored the psychopathological dimensions derived from items in the SANS and SAPS. However, it has been pointed out that the SANS and SAPS do not measure mood disturbances which are not un-commonly present in patients with schizophrenia ŽMinas et al., 1994 . Different lines of research. employing scales with items of mood profile, such as the Positive and Negative Syndrome Scale ŽPANSS. ŽKay and Sevy, 1990; Lindenmayer et

.

al., 1995a,b , reported additional dimensions of depression and excitement in schizophrenia.

The symptom clusters discussed in this study were statistically segregated and may well be treated as dimensions rather than discrete cate-gories of schizophrenia. Carefully interpreted, the identification of clinically correlated symptom clusters may help define more homogenous di-mensions of psychopathology in schizophrenia. Taking the above notions together, we can more critically examine the actual relationships between

symptom assessment, psychopathological con-struct and external validation measures.

Acknowledgements

This work was supported by grants from the National Health Research Institute, Taiwan,

Ž .

R.O.C. DOH-83-HR-306 and DOH-84-HR-306 . References

American Psychiatric Association, 1980. DSM-III: Diagnostic and Statistical Manual of Mental Disorders, 3rd ed. Ameri-can Psychiatric Press, Washington, D.C.

Andreasen, N.C., 1983. The Scale for the Assessment of

Ž .

Negative Symptoms SANS . University of Iowa, Iowa City, IA.

Andreasen, N.C., 1984. The Scale for the Assessment of

Ž .

Positive Symptoms SAPS . University of Iowa, Iowa City, IA.

Andreasen, N.C., Arndt, S., Alliger, R., Miller, D., Flaum, M., 1995. Symptoms of schizophrenia: methods, meanings and mechanisms. Archives of General Psychiatry 52, 341]351. Andreasen, N.C., Olsen, S., 1982. Negative v. positive

schizophrenia: definition and validation. Archives of Gen-eral Psychiatry 39, 789]794.

Andreasen, N.C., Olsen, S.A., Dennert, J.W., Smith, M.R., 1982. Ventricular enlargement in schizophrenia: relation-ship to positive and negative symptoms. American Journal of Psychiatry 139, 297]302.

Arndt, S., Alliger, R.J., Andreasen, N.C., 1991. The distinction of positive and negative symptoms: the failure of a two-di-mensional model. British Journal of Psychiatry 158, 317]322.

Bilder, R.M., Mukherjee, S., Rieder, R.O., Pandurangi, A.K., 1985. Symptomatic and neuropsychological components of defect states. Schizophrenia Bulletin 11, 409]419. Breiger, R.L., Boorman, S.A., Arabie, P., 1975. An algorithm

for clustering relational data with applications to social network analysis and comparison with multidimensional scaling. Journal of Mathematical Psychology 12, 328]383. Carpenter, W.T., Jr., Heinrichs, D.W., Wagman, A.M.I., 1985.

Treatment of negative symptoms. Schizophrenia Bulletin 11, 440]452.

Chen, C.H., 1996. The properties and applications of the convergence of correlation matrices. In: ASA Proceedings of the Section on Statistical Computing, American Statisti-cal Association.

Crow, T.J., 1980. Molecular pathology of schizophrenia: more than one disease process? British Medical Journal 280, 66]68.

Crow, T.J., 1985. The two-syndrome concept: origins and current status. Schizophrenia Bulletin 11, 471]485.

(10)

De Leon, J., Cuesta, M.J., Peralta, V., 1993. Delusions and hallucinations in schizophrenic patients. Psychopathology 26, 286]291.

Fenton, W.S., McGlashan, T.H., 1994. Antecedents, symptom progression and long-term outcome of the deficit syndrome in schizophrenia. American Journal of Psychiatry 151, 351]356.

Hwu, H.G., Yeh, E.K., Chang, L.Y., 1986. Chinese diagnostic interview schedule: I. Agreement with psychiatrist’s diagno-sis. Acta Psychiatrica Scandinavica 73, 225]233.

Kay, S.R., Fiszbein, A., Lindenmayer, J.-P., Opler, L.A., 1986. Positive and negative syndromes in schizophrenia as a function of chronicity. Acta Psychiatrica Scandinavica 75, 507]518.

Kay, S.R., Sevy, S., 1990. Pyramidical model of schizophrenia. Schizophrenia Bulletin 16, 537]545.

Klimidis, S., Stuart, G.W., Minas, I.H., Copolov, D.L., Singh, B.S., 1993. Positive and negative symptoms in the psy-choses: re-analysis of published SAPS and SANS global ratings. Schizophrenia Research 9, 11]18.

Kulhara, P., Kota, S.K., Joseph, S., 1986. Positive and negative subtypes of schizophrenia: a study from India. Acta Psychi-atrica Scandinavica 74, 353]379.

Liddle, P.F., 1987. The symptoms of chronic schizophrenia: a re-examination of the positive]negative dichotomy. British Journal of Psychiatry 151, 145]151.

Lindenmayer, J.-P., Bernstein-Hyman, R., Grochowski, S., Bark, N., 1995a. Psychopathology of schizophrenia: initial validation of a 5-factor model. Psychopathology 28, 22]31. Lindenmayer, J.-P., Grochowski, S., Bernstein-Hyman, R., 1995b. Five factor model of schizophrenia: replication across samples. Schizophrenia Research 14, 229]234. Malla, A.K., Norman, R.M.G., Williamson, P., Cortese, L.,

Diaz, F., 1993. Three syndrome concept of schizophrenia: a factor analytic study. Schizophrenia Research 10, 143]150.

McQuitty, L.L., 1968. Multiple clusters, types and dimensions from iterative intercolumnar correlational analysis. Multi-variate Behavioral Research 3, 465]477.

Mellor, C.S., 1970. First rank symptoms of schizophrenia: I. The frequency in schizophrenics on admission to hospital, II. Differences between individual first rank symptoms. British Journal of Psychiatry 117, 15]23.

Miller, D.D., Arndt, S., Andreasen, N.C., 1993. Alogia, atten-tional impairment and inappropriate affect: their status in the dimensions of schizophrenia. Comprehensive Psychia-try 34, 221]226.

Minas, I.H., Stuart, G.W., Klimidis, S., Jackson, H.J., Singh, B.S., Copolov, D.L., 1992. Positive and negative symptoms in the psychoses: multidimensional scaling of SAPS and SANS items. Schizophrenia Research 8, 143]156. Minas, I.H., Klimidis, S., Stuart, G.W., Copolov, D.L., Singh,

B.S., 1994. Positive and negative symptoms in the psy-choses: principal components analysis of items from the Scale for the Assessment of Positive Symptoms and the Scale for the Assessment of Negative Symptoms. Compre-hensive Psychiatry 35, 135]144.

Sarai, M., Matsunaga, H., 1993. Symptom segregation in schizophrenia: the significance of thought disorder. Schizophrenia Research 10, 159]163.

Schneider, C., 1942. Die Schizophrenen Symptomverbande. Springer, Berlin.

Strauss, J.S., Carpenter, W.T., Bartko, J.J., 1974. The diagnosis and understanding of schizophrenia: Part III. Speculations on the processes that underlie schizophrenic symptoms and signs. Schizophrenia Bulletin 11, 61]76.

Stuart, G.W., Malone, V., Currie, J., Klimidis, S., Minas, I.H., 1995. Positive and negative symptoms in neuroleptic-free psychotic inpatients. Schizophrenia Research 16, 175]188.

數據

Fig. 1. Correlation plots. The numeric correlation coefficients y1 to 1 are illustrated in a blue ]red color spectrum of 100 block

參考文獻

相關文件

Raising Students’ Awareness of Text Structures, Rhetorical Functions and Language Items in Academic Texts.. Use of

Reading Task 6: Genre Structure and Language Features. • Now let’s look at how language features (e.g. sentence patterns) are connected to the structure

- allow students to demonstrate their learning and understanding of the target language items in mini speaking

< Notes for Schools: schools are advised to fill in the estimated minimum quantity and maximum quantity (i.e. the range of the quantity) of the items under Estimated Quantity

Students are asked to collect information (including materials from books, pamphlet from Environmental Protection Department...etc.) of the possible effects of pollution on our

• Teaching grammar through texts enables students to see how the choice of language items is?. affected by the context and how it shapes the tone, style and register of

By correcting for the speed of individual test takers, it is possible to reveal systematic differences between the items in a test, which were modeled by item discrimination and

volume suppressed mass: (TeV) 2 /M P ∼ 10 −4 eV → mm range can be experimentally tested for any number of extra dimensions - Light U(1) gauge bosons: no derivative couplings. =>