Psychological Medicine, 1993, 23, 745-753. Copyright © 1993 Cambridge University Press Symptom dimensions and cognitive and social functioning in recent-onset schizophrenia A. J. W. VAN DER DOES, 1 P. M. A. J. DINGEMANS, D. H. LINSZEN, M. A. NUGTER AND W. F. SCHOLTE From the Department of Psychiatry, Leiden University; Department of Psychiatry, University of Amsterdam; Institute for Psychotherapy 'De Oosthoek', Limmen, The Netherlands SYNOPSIS The relationships among symptoms, cognitive functioning and social functioning were investigated in 60 patients with recent-onset schizophrenia. Positive symptoms were unrelated to cognitive measures. Disorganization and depressive symptoms were correlated significantly with Card Sorting performance. Furthermore, only negative symptoms were correlated significantly with social functioning. These results replicate earlier studies with chronic and mixed samples, and support the validity of disorganization as a separate symptom dimension. In contrast to most previous studies, no significant correlations were found between negative symptoms and cognitive measures. However, some evidence was found for a non-linear association between negative symptoms and several cognitive measures. The variation explained by a curvilinear model was not high, but for some cognitive measures this model was clearly superior to a linear model. If replicated, this finding supports the position that cross-sectionally measured negative symptoms cannot be viewed as a unitary concept. INTRODUCTION In the past decade supporting evidence has become available for the validity of the so-called type I/type II model of schizophrenia, but the need for refinement of the model has also become apparent. In the type I/type II model, two major syndromes of schizophrenia are postulated, each with a specific pathophysiology, course and response to treatment (Crow, 1980). The syndromes can be recognized by clinical presentation (predominance of positive or nega- tive symptoms). Although biological research has yielded mixed results (e.g. Andreasen & Olsen, 1982; Andreasen et al. 1990), studies have indicated that negative symptoms are a common and valid component of schizophrenia, and that they are related to poor pre-morbid functioning and poor long-term outcome when present beyond the early phase of illness (McGlashan & Fenton, 1992). Research has also shown a relationship between negative symp- • Address for correspondence: Dr A. J. W. Van der Does, Leiden University, Department of Psychiatry, Jelgersma Polikliniek, PO Box 1251, 2340 BG Oegstgeest, The Netherlands. toms and cognitive impairment (e.g. Andreasen et al. 1990; Addington et al. 1991; Schwartz et al. 1991). Recently, however, the type I/type II model has been challenged by Liddle (1987a), who factor-analysed schizophrenic symptoms and concluded that these segregate into three dimensions, with a disorganization factor be- sides a positive and negative symptom factor. This three-dimensional structure has since been replicated with different symptom scales and in different samples (Liddle & Barnes, 1990; Arndt et al. 1991; Van der Does et al. 1993). There are indications that these three symptom dimensions have distinctive patterns of relationships with other variables, for instance neurological signs and cognitive test performance (e.g. Liddle, 19876; Kay, 1990; Liddle & Morris, 1991). The studies by Liddle and co-workers have been with chronic patients only and with rather small sample sizes. The present study was designed to investigate the relationships among symptom dimensions, cognitive and social func- tioning in a homogeneous sample of young, recent-onset schizophrenic patients. In a pre- vious report (Van der Does et al. 1993), a four- dimensional model of symptoms was found with 745