The present work consists of a randomized clinical trial comparing the effectiveness of two interventions based on social cognition training in outpatients with schizophrenia. The investigators sought to compare the effect of a "targeted" (TAR) and a "broad-based" (SCIT) intervention on schizophrenia patients' performance in facial affect recognition, theory of mind and attributional style. Secondarily, the investigators compare the effect on symptomatology, general cognition and functioning. The main hypothesis was that the patient group receiving TAR would exhibit a greater improvement in emotion recognition performance at the post-intervention assessment in comparison to patients receiving the SCIT, and, conversely, patients receiving SCIT would show more effect in ToM and attributional style. To assess the durability of these effects, performance in measures of social cognition, basic cognitive functioning, symptomatology and functional capacity were assessed before (T0), after treatment (T1) and 3 months later (T2).

In recent years, there has been an interest in the development of intervention programs focused on the social cognition for people with schizophrenia (Andres et al., 2001; P Penn et al. 2005). At least, five reviews and one meta-analysis have been conducted to date, which demonstrate promising results of the effectiveness of such interventions on social cognitive deficits and functional outcomes (Tan et al; 2016; Choi et al., 2009; Horan et al., 2008; Kurtz and Richardson, 2012; Statucka and Walder, 2013; Wolwer et al., 2010). Some approaches are focused on one specific domain of social cognition ("targeted" interventions, such as the Training in Affect Recognition (TAR, Wolwer et al., 2005), and others incorporate multiple domains to create more complex, eclectic programs ("broad-based" interventions, such as the Social Cognition and Interaction Training (SCIT; Penn et al. 2007; Kurtz & Richardson, 2012).

TAR (Frommann et al., 2003) is one of the social cognition interventions with greater empirical support (Statucka & Walder, 2016) and has been shown to effectively attenuate facial affect recognition deficits in patients with schizophrenia (Wölwer et al. 2005; Wölwer and Frommann 2011; Sachs et al. 2012; Luckhaus et al. 2013). TAR teaches compensation strategies using errorless learning principles, positive reinforcement, feature abstraction, self-instruction and, most importantly, verbalization of characteristic features of facial affect. In a randomized controlled trial, the TAR group achieved significant improvements in facial affect recognition -in particular in recognizing sad faces- and in the quality of life domain social relationship. Furthermore, the TAR training contributed to enhancing some aspects of cognitive functioning and negative symptoms (Sachs et al, 2012). In the other hand, SCIT is a 24-session manual-based group treatment, including elements of cognitive behavioral therapy and social skills training. It is designed for those with schizophrenia spectrum disorders to improve social functioning by enhancing social cognition. Across different studies and research groups, SCIT has been also shown to improve in emotion perception, theory of mind (ToM), and social functioning (Bartholomeusz et al., 2013; Combs et al., 2007; Hasson-Ohayon, 2014; Parker et al. 2013; Penn et al., 2007; Roberts & Penn, 2009; Roberts et al. 2010; Roberts et al., 2014, 2016; Wang et al., 2013).

The efficacy of both interventions has been demonstrated in randomized controlled trials compared to "treatment as usual", occupational therapy or cognitive remediation (Kurtz et al. 2016) but to date no study has compared the efficacy of two different social cognitive interventions (a direct comparison design). A more precise knowledge about the effect of each intervention on the 4 main domains of social cognition (affect recognition, theory of mind, attributional style and social perception) is needed, and this would enable to identify potential candidates for each program.

The present work consists of a randomized clinical trial comparing the effectiveness of two interventions based on social cognition training in outpatients with schizophrenia. The investigators sought to compare the effect of a "targeted" (TAR) and a "broad-based" (SCIT) intervention on schizophrenia patients' performance in facial affect recognition, theory of mind and attributional style. Secondarily, the investigators compare the effect on symptomatology, general cognition and functioning. The main hypothesis was that the patient group receiving TAR would exhibit a greater improvement in emotion recognition performance at the post-intervention assessment in comparison to patients receiving the SCIT, and, conversely, patients receiving SCIT would show more effect in ToM and attributional style. To assess the durability of these effects, performance in measures of social cognition, basic cognitive functioning, symptomatology and functional capacity were assessed before (T0), after treatment (T1) and 3 months later (T2).

2. Methods

2.1. Participants Outpatients who met DSM-IV criteria for schizophrenia and schizoaffective disorder (SCID-P; First et al. 1994) with stable symptoms in the range from 18 to 65 years were included into the study. Patients were recruited from 4 Mental Healt Centers in Madrid, Barcelona, Zaragoza and Teruel (Spain). All were clinically stable, without any psychiatric hospitalizations in the last 3 months, with the same antipsychotic medication during the previous 6 weeks, and no planned change in the drug regime for the next 3 months. Exclusion criteria were: 1. Disorders other than schizophrenia or schizoaffective disorder, according to DSM-IV diagnosis criteria; 2. Additional axis-I or axis-II diagnosis; 3. Dependence to alcohol or other drugs (except nicotine); 4. Serious somatic disorders or organic brain damage; 5. Mental retardation or difficulty speaking or understanding the Spanish language.

The study was approved by the loval ethics committee and all participants gave their informed consent. Overall 100 participants were randomized either to TAR group (n = 49) or to the SCIT group (n = 51) (Consort diagram, Graphic 1).

2.1.1. Treatment

TAR is a 12-session training on facial affect recognition over a period of 6 weeks. Treatment includes one therapist (psychiatrist or clinical psychologist) and 2 patients. It involves neuropsychological strategies, such as restitution and compensation, as well as principles of errorless learning, direct positive reinforcement, verbalization and self-instruction (Frommann et al., 2003; Wölwer et al., 2005). The program is divided into three blocks, whereas each block consists of 4 sessions: during the first block patients learn to identify and discriminate the prototypical facial signs of the six basic emotions (happiness, sadness, fear, disgust, anger and surprise). The next block aims at a more holistic processing mode with fast decisions, relying on first impression, nonverbal processing and recognition of facial expressions with small intensities. The third block deals with the role of facial emotions in social, behavioral and situational context. Baseline assessments (T0=pre-treatment) were performed after enrolment to the study and post treatment assessments (T1=post-treatment) after the end of the training period (Sachs et al. 2012).

SCIT is a manual-based group intervention that is delivered in 20-24weekly, hour-long sessions. Groups include two clinicians and six to ten patients. SCIT uses a combination of psychoeducation, drill-and-repeat skill practice, strategy games, heuristic rehearsal, and homework assignments to remediate deficits and decrease biases in social cognition. Each SCIT group participant was encouraged to identify a 'practice partner', a family member or acquaintance who was willing to practice SCIT skills with the participant weekly in lieu of, or in addition to, traditional homework. SCIT clinicians attempted to reach practice partners by phone each week to check -in and provide guidance in their efforts to support SCIT participants' learning (Roberts et al. 2014).

Inclusion Criteria:

• Outpatients who met DSM-IV criteria for schizophrenia and schizoaffective disorder (SCID-P; First et al. 1994)
• Clinical stability: without any psychiatric hospitalizations in the last 3 months, with the same antipsychotic medication during the previous 6 weeks, and no planned change in the drug regime for the next 3 months.
• Age in the range from 18 to 65 years

Exclusion Criteria:

1. Disorders other than schizophrenia or schizoaffective disorder, according to DSM-IV diagnosis criteria;
2. Additional axis-I or axis-II diagnosis;
3. Dependence to alcohol or other drugs (except nicotine);
4. Serious somatic disorders or organic brain damage;
5. Mental retardation or difficulty speaking or understanding the Spanish language.

Trained raters who were blind to treatment condition assessed participants on the following variables at baseline, post-treatment and 3-month follow-up