Impairments in social functioning are among the most debilitating and treatment refractory aspects of schizophrenia (Bellack et al 2007
). It has become clear that further improvements in social functioning will not occur through gains in psychotic symptom management alone, since psychotic symptoms are typically not closely related to functional adaptation levels in community-dwelling outpatients (Carter 2006
; Heydebrand et al 2004
) and there has been little improvement in community functioning over the past 100 years (Hegarty et al 1994
). Instead, treatments that directly address the key determinants of poor social functioning are required to ameliorate these impairments. There has been good success in identifying basic neurocognitive processes that predict social dysfunction (Green et al 2000
; Green et al 2004
), which has been one rationale for major NIMH initiatives to stimulate the development of new pharmacological treatments for cognitive deficits (Marder and Fenton 2004
). However, it is unlikely that interventions targeting only basic neurocognition will be sufficient to achieve optimal functioning since neurocognitive deficits typically account for only 10% to 40% of the variance in outcome (Green et al 2000
; Green et al 2008
; Penn et al 2006
). Thus, there is a critical need to identify and treat other determinants of poor outcome.
Rapidly growing evidence indicates that impairments in the domain of social cognition are important determinants of functional outcome in schizophrenia. Social cognition is a multifaceted construct that refers to the mental operations underlying social interactions, which include processes involved in perceiving, interpreting, and generating responses to the intentions, dispositions, and behaviors of others (Brothers 1990
; Fiske and Taylor 1991
; Kunda 1999
). Schizophrenia patients show substantial deficits in several aspects of social cognition, including emotional processing, social perception, Theory of Mind, and attributional style (Penn et al 2006
). There is a general consensus that social cognition is distinct from, though related to, basic neurocognition and other clinical features of schizophrenia (Green et al 2005
; Penn et al 1997
; Sergi et al 2007
). Furthermore, social cognition shows unique
relationships to functional outcome, above and beyond basic cognition (Couture et al 2006
). For example, social cognition has been found to mediate the relationship between basic neurocognition and functional outcome (Addington et al 2006
; Brekke et al 2005
; Sergi et al 2006
; Vauth et al 2004
). Hence, social cognition appears to be more proximal to functional outcome than basic cognition and, for that reason, could be an even better target for intervention.
A few research groups have demonstrated that the social cognitive deficits of schizophrenia are modifiable through brief experimental manipulations or more intensive psychosocial interventions (see (Horan et al 2008
). For example, performance on facial affect recognition tests has been enhanced through brief (e.g., an hour or less) intervention probes such as attentional manipulations or monetary reinforcement (Combs et al 2006
; Penn and Combs 2000
; Russell et al 2006
; Silver et al 2004
). In addition, longer-term studies that incorporated social cognitive training exercises into multi-component treatment packages (often including neurocognitive remediation) demonstrate improvements on social cognitive tests (Bell et al 2001
; Hodel et al 2004
; Hogarty et al 2004
; van der Gaag et al 2002
). However, specifically attributing any intervention effects to the social cognitive training in these longer-term treatments is difficult because the procedures were embedded within multi-component rehabilitation programs.
A handful of research groups have begun developing and testing treatment programs that specifically target social cognition. For example, Wolwer and colleagues in Germany developed the Training in Affect Recognition program to remediate facial emotion perception deficits in schizophrenia. This 12-session computer-based training program is administered to pairs of patients at a time. It initially focuses on recognition of specific facial features associated with basic emotions and progresses to more complex facial displays in social contexts. Following an initial uncontrolled feasibility study (Frommann et al. 2003
), a randomized trial demonstrated that inpatients who received this intervention demonstrated significant improvements in facial affect perception (and working memory), whereas patients in a time-matched neurocognitive remediation program or treatment as usual did not (Frommann et al 2003
; Wolwer et al 2005
). Since only facial affect perception was assessed, it is unknown whether this resource-intensive intervention leads to improvements in other social cognitive processes.
Penn and colleagues in North Carolina developed Social Cognitive and Interaction Training (SCIT; (Penn et al 2007b
), an 18-session intervention that addresses three social cognitive processes: emotion perception, attributional bias, and Theory of Mind. The intervention is designed for small groups of six to eight patients and includes a variety of interactive training exercises, such as distinguishing facts from guesses, avoiding jumping to conclusions about suspicious beliefs, and gathering information about others’ emotions and beliefs. An uncontrolled feasibility study of seven inpatients demonstrated improvements in attributional bias and Theory of Mind (but not emotion perception), as well as clinical symptoms (Penn et al 2005
). A subsequent study using a slightly modified treatment manual demonstrated improvements in social attribution and Theory of Mind, as well as facial emotion perception, in 18 forensic inpatients compared to patients receiving treatment as usual (Combs et al 2007
). Recently, a quasi-experimental study by Roberts and Penn (Roberts and Penn in press
) evaluated an outpatient sample that received either SCIT plus treatment as usual or treatment as usual-only (without random assignment to condition). The SCIT group showed significant treatment benefits on a facial affect perception task, but not on measures of the other two targeted social cognitive processes.
These encouraging findings across studies (also see (Roncone et al 2004
)) have several limitations, including: 1) most did not include active control groups matched for time in treatment, 2) it is unclear from these studies whether social cognitive improvements merely reflected changes in basic neurocognitive functioning, and 3) with one exception, all used inpatients who often showed concurrent improvements in clinical state that could influence social cognition results. Because schizophrenia inpatients comprise a small fraction of patients, social cognitive interventions will find greater use in stabilized outpatients. Hence, it is critical to evaluate the efficacy of targeted social cognitive interventions in community-dwelling outpatients.
We report here the initial results of a randomized, controlled clinical trial for a new integrative social cognitive intervention for outpatients with psychotic disorders designed to improve four domains, including facial affect perception, social perception, attributional style, and Theory of Mind. As detailed below, this program combines successful elements from two existing programs (Frommann et al 2003
; Penn et al 2007b
) with a variety of novel training exercises and materials. We evaluated whether this new intervention results in specific improvements on social cognitive tests.