Social dysfunction in chronic schizophrenia : nature, treatment and generalization
The present research explores the impact of neurocognitive status on generalization of social skills training and how these difficulties are further compounded by the difficulty of inadequate matching of problems to treatments. The generalization problem is addressed in two phases. In the first phase, a suggestion that a subgroup of chronic schizophrenia patients may have normal abstract problem solving was investigated by comparing paranoid and non-paranoid chronic schizophrenic in-patients (n= 11 per group) with assessed social dysfunction. Only the paranoid group had profiles comparable to matched normal controls. The non-paranoid group had neuropsychological deficits (executive dysfunction) on the Wisconsin Card Sorting Test (WCST). In the second phase, a longitudinal study compared the differential efficacy of behavioural versus cognitive-behavioural social skills training (SST), and monitored the effects on maintenance and generalization, focusing on social anxiety. Concurrently, the treatment validity of a multidimensional assessment of social problem solving was assessed using an alternating treatments design with a multiple baseline across participants from three groups: skill deficits (n=3), cognitive blocks (n=3) and a control group (n=3) with both problems. All participants received four randomized sessions each of Behavioural SST and Cognitive-Behavioural SST. The group SST used a 'whisper in the ear' game format and included self-instructional training and social problem solving to facilitate generalization. The findings demonstrated the treatment validity of the multidimensional assessment and offered preliminary evidence for the differential efficacy, maintenance and generalization of Behavioural and Cognitive-Behavioural SST which occurred for untrained behaviours and was maintained at the 3- and 9-month follow-ups but only for those not impaired on the WCST. It is concluded that it is more productive for assessmentto focus on social anxiety and the practice of mixing schizophrenia patients in SST, without taking a neurocogitive baseline, may be responsible for confounding generalization efforts. A limitation of the research is that gender differences were not testable as there were insufficient numbers.