Social cognition constructs

Emotion processing is the ability to perceive and manage emotions [Reference Pinkham, Penn, Green, Buck, Healey and Harvey5]. It includes both low-level and high-level processes. The low-level processes typically include the ability to identify and recognize emotions from facial expressions or other cues, for example, prosody (voice intonations) and posture [Reference Pinkham, Penn, Green, Buck, Healey and Harvey5]. High-level processes, on the other hand, relate to a deeper understanding of the emotions of others, for example, understanding that a combination of simple emotions can create a more complex emotion [Reference Pinkham, Penn, Green, Buck, Healey and Harvey5]. Emotion processing deficits have been reported in different clinical populations, notably schizophrenia spectrum disorders, autism spectrum disorder, and head trauma [Reference Gur and Gur7, Reference Oliver, Moxon-Emre, Lai, Grennan, Voineskos and Ameis8].

Theory of mind (ToM) is a high-level construct referring to the ability to infer the mental states of others (for example, their intentions, beliefs, knowledge, or emotions) [Reference Achim, Guitton, Jackson, Boutin and Monetta9]. To infer a mental state adequately, many sources of information must be considered [Reference Achim, Guitton, Jackson, Boutin and Monetta9], including what is perceived in the current environment (notably the recognition of emotional expressions and contextual elements) and information stored in memory (including general social knowledge). ToM deficits are common in several clinical populations, including people with schizophrenia, bipolar disorder, autism spectrum disorder, Parkinson’s disease, and other disorders [Reference Oliver, Moxon-Emre, Lai, Grennan, Voineskos and Ameis8].

Social perception involves the general processes required for decoding, understanding, and interpreting social cues [Reference Pinkham, Penn, Green, Buck, Healey and Harvey5]. One aspect of social perception that can more readily be evaluated is social knowledge, that is, the understanding of the rules, roles, and social behaviors typically expected in various situations [Reference Pinkham, Penn, Green, Buck, Healey and Harvey5, Reference Achim, Guitton, Jackson, Boutin and Monetta9] (e.g., people would not be expected to interact the same way with their colleagues as they would with their spouse). Social knowledge is affected in people with schizophrenia, but the deficits may depend on the specific task being used or the stage of the illness [Reference Achim, Ouellet, Roy and Jackson10].

Attributional style refers to the way an individual usually interprets the cause of social behavior [Reference Pinkham, Penn, Green, Buck, Healey and Harvey5]. People with schizophrenia who have persecutory delusions tend to blame others for negative situations and negative social outcomes [Reference Pinkham, Penn, Green, Buck, Healey and Harvey5, Reference Lahera, Herrera, Reinares, Benito, Rullas and González-Cases11–Reference Achim, Sutliff, Samson, Montreuil and Lecomte13]. People without psychopathology tend to see themselves as the cause of positive events and tend to attribute negative events to others or the context (self-serving bias) [Reference Achim, Sutliff, Samson, Montreuil and Lecomte13]. Other biases include the negative bias, where people tend to interpret situations more negatively, and the personalizing bias, where people tend to easily blame others for negative situations rather than considering the influence of circumstances. Attribution biases are observed in many different clinical populations, including anxiety disorders, depression, and people with schizophrenia who present with delusions [Reference Achim, Sutliff, Samson, Montreuil and Lecomte13, Reference Plana, Lavoie, Battaglia and Achim14].

Other emerging constructs: empathy and social metacognition

Empathy is an important, complex construct in the field of social cognition. Current models suggest that empathy involves an affective component, a cognitive component, and a regulation component [Reference Decety15]:

1. (1) Affective empathy component: Affective response of shared emotional resonance with what the other person is going through, which has been conceptualized to be fairly automatic and bottom-up (e.g., “I feel what you are feeling”).

2. (2) Cognitive empathy component: The cognitive ability to put oneself in another’s shoes (also called perspective-taking or cognitive empathy), which is similar to ToM skills as it involves an understanding that the other person has a particular affective state that may differ from one’s own (e.g., “I understand what you are feeling”).

3. (3) Emotion regulation: The ability to manage one’s personal response to the emotions of others. This ability is crucial for the distinction of self vs other’s emotions and for the ability to manage one’s personal affective state when facing the other’s affective state. A failure of this process can lead someone to feel overwhelmed by the emotions of other people.

Empathy deficits have been extensively documented in schizophrenia, especially for the cognitive component of empathy [Reference Langdon, Coltheart and Ward16–Reference Achim, Ouellet, Roy and Jackson18]. However, whether individuals with schizophrenia have deficits in the affective component of empathy is still a debate as studies have shown mixed results [Reference Corbera, Ikezawa, Bell and Wexler19–Reference Lehmann, Bahçesular, Brockmann, Biederbick, Dziobek and Gallinat22].

As for metacognition skills, they are the ability to evaluate thoughts, including one’s own and those of others [Reference Pinkham, Penn, Green, Buck, Healey and Harvey5]. While there is yet no consensus on a single model of metacognition that identifies its processes or their definition, such a model will likely group together several of the other aspects of social cognition already mentioned (e.g., ToM, social knowledge, emotion regulation) and include an overarching self-awareness.

Social cognition and social functioning

Extensive evidence has shown that social cognitive deficits are pervasive in individuals with schizophrenia. In their meta-analysis of peer-reviewed studies from 1980 to 2011 targeting social cognition abilities in schizophrenia, Savla and colleagues [Reference Savla, Vella, Armstrong, Penn and Twamley2] revealed that individuals with schizophrenia show large impairments in the domains of social perception (Hedge’s g =1.04), ToM (g = 0.96), emotion perception (g = 0.89), and emotion processing (g = 0.88) [Reference Savla, Vella, Armstrong, Penn and Twamley2], these deficits being present at all stages of the illness although to lesser extent in the early stages of the disease [Reference Savla, Vella, Armstrong, Penn and Twamley2, Reference Green, Horan and Lee4, Reference McCleery, Ventura, Kern, Subotnik, Gretchen-Doorly and Green23–Reference Green, Penn, Bentall, Carpenter, Gaebel and Gur25]. As social cognition involves the use of mental operations and processes, it is intricately related to neurocognition [Reference Savla, Vella, Armstrong, Penn and Twamley2, Reference Ventura, Wood and Hellemann26–Reference Thibaudeau, Achim, Parent, Turcotte and Cellard28]. While neurocognition and social cognition play an important role in predicting everyday functioning [Reference Halverson, Orleans-Pobee, Merritt, Sheeran, Fett and Penn3], the extent to which each predicts functioning can depend on the exact domain of neurocognition or social cognition being measured. Many studies have examined the role of social cognition, neurocognition and social functioning [Reference Fett, Viechtbauer, Dominguez, Penn, van Os and Krabbendam1, Reference Couture, Granholm and Fish27, Reference Ruiz-Toca, Fernández-Aragón, Madrigal, Halverson, Rodriguez-Jimenez and Lahera29–Reference Oliver, Haltigan, Gold, Foussias, DeRosse and Buchanan32], and several meta-analyses have attempted to describe their specific contributions. Importantly, a meta-analysis by Halverson et al. [Reference Halverson, Orleans-Pobee, Merritt, Sheeran, Fett and Penn3] revealed that social cognition explained more unique variance in function than neurocognition and that social cognition mediated the relationship between neurocognition and functional outcome [Reference Halverson, Orleans-Pobee, Merritt, Sheeran, Fett and Penn3]. This mediation has also been supported by many other studies and meta-analyses [Reference Fett, Viechtbauer, Dominguez, Penn, van Os and Krabbendam1, Reference Couture, Granholm and Fish27, Reference Couture, Penn and Roberts30–Reference Oliver, Haltigan, Gold, Foussias, DeRosse and Buchanan32]. A more recent semisystematic analysis by Kharawala et al. [Reference Kharawala, Hastedt, Podhorna, Shukla, Kappelhoff and Harvey33] supported this mediation and went further into dissecting the functional outcomes into functional capacity and real-world functioning. It found that while both neurocognition and social cognition were associated with functional capacity and real-world functioning, social cognition played a larger role in these outcomes [Reference Kharawala, Hastedt, Podhorna, Shukla, Kappelhoff and Harvey33].

ToM has been shown to have a strong relationship with functioning [Reference Fett, Viechtbauer, Dominguez, Penn, van Os and Krabbendam1, Reference Hajdúk, Krajoviová, Zimányiová, Koínková, Heretik and Peeák34, Reference Thibaudeau, Cellard, Turcotte and Achim35], with negative impacts of ToM deficits on several important social functioning impairments, such as poorer personal independence, hygiene, and worse levels of school and work functioning [Reference Thibaudeau, Cellard, Turcotte and Achim35–Reference Ventura, Ered, Gretchen-Doorly, Subotnik, Horan and Hellemann37]. Furthermore, ToM interventions have been shown to improve functioning [Reference Grant, Lawrence, Preti, Wykes and Cella38]. In addition to ToM, emotion perception and recognition also show strong relationships with social functioning in several studies [Reference Hajdúk, Krajoviová, Zimányiová, Koínková, Heretik and Peeák34, Reference Irani, Seligman, Kamath, Kohler and Gur39].

In summary, evidence suggests that social cognition is a stronger predictor of social functioning than neurocognition, explaining a meaningful percent of unique variance in functioning [Reference Fett, Viechtbauer, Dominguez, Penn, van Os and Krabbendam1, Reference Halverson, Orleans-Pobee, Merritt, Sheeran, Fett and Penn3, Reference Couture, Penn and Roberts30]. However, this complex relationship is still a matter of research, as many factors can influence social functioning such as negative symptoms and metacognition [Reference Kharawala, Hastedt, Podhorna, Shukla, Kappelhoff and Harvey33], and the type of instrument utilized for assessment (such as self-report vs observer report) [Reference Kharawala, Hastedt, Podhorna, Shukla, Kappelhoff and Harvey33, Reference Vita, Gaebel, Mucci, Sachs, Barlati and Giordano40]. Despite general scientific agreement on the pervasiveness of the social cognition deficits in schizophrenia and the strong relationship of these impairments to social functioning, there is not yet a commonly accepted approach to testing and treating social cognitive deficits. Although more research is needed, what is already established is sufficient to offer a first attempt at laying out preliminary recommendations and procedures that clinicians can use and integrate into their practice.

Social cognition as a perceived need of patients

People with schizophrenia themselves report links between their social cognitive skills and everyday function. A needs assessment survey on social cognition in people with schizophrenia was conducted in Japan [Reference Uchino, Okubo, Takubo, Aoki, Wada and Hashimoto41]. With a newly developed questionnaire, KEA-SC (Survey Questionnaire on Knowledge, Experience, and Awareness of Social Cognition), data from 232 people with schizophrenia and 494 healthy subjects were analyzed, being weighted according to the demographics of the national population in Japan. The results showed that less than a quarter of both groups were familiar with the term or concept of social cognition and that less than 5% had experienced being assessed or treated for social cognition. In contrast, more than half of both groups were aware of the relationship between social cognition and social functioning when being given additional detailed examples of social situations, with the results for people with schizophrenia being significantly higher than those for the healthy subjects.

To understand the perceived need of patients in this area, questionnaires, such as the Observable Social Cognition Rating Scale (OSCARS), have been developed to assess subjective social cognition difficulties [Reference Graux, Thillay, Morlec, Sarron, Roux and Gaudelus42, Reference Halverson, Hajdúk, Pinkham, Harvey, Jarskog and Nye43]. Studies using this tool have shown that individuals with schizophrenia were conscious of greater difficulties explaining social situations or behaviors that require social cognition, compared with healthy controls. These studies revealed that although people with schizophrenia are usually assumed to have a limited perception of their psychological and cognitive symptoms [Reference Aleman, Agrawal, Morgan and David44, Reference Cella, Swan, Medin, Reeder and Wykes45], their subjective report of social cognition difficulties was associated with their reports of poorer social functioning [Reference Uchino, Okubo, Takubo, Aoki, Wada and Hashimoto41] and mediated the relationship between neurocognition and social functioning [Reference Uchino, Okubo, Takubo, Aoki, Wada and Hashimoto46]. These results suggest that interventions in social cognition might not only address clinical outcomes but also may relate closely to the client’s own perceptions of deficits.

Social cognition as transdiagnostic phenotype

Besides being an important powerful predictor of functioning and an area of subjective functional need for patients with SZ as described above, social cognition deficits are present in other psychiatric disorders highlighting the importance of this as a transdiagnostic domain [Reference Lakhani, Bhola and Mehta47]. A prime example is autism spectrum disorder (ASD), where social communication deficits are required to establish the diagnosis (along with repetitive and restrictive behaviors and interests) [48]. A recent meta-analysis demonstrated qualitatively and quantitatively similar deficits in SZ and ASD across standardized tests assessing different social cognitive constructs, including emotion processing and ToM [Reference Oliver, Moxon-Emre, Lai, Grennan, Voineskos and Ameis8, Reference Morel-Kohlmeyer, Thillay, Roux, Amado, Brenugat and Carteau-Martin49, Reference Martinez, Alexandre, Mam-Lam-Fook, Bendjemaa, Gaillard and Garel50]. Moreover, overlap in psychosis-related negative symptoms and to a lesser degree in positive symptoms [Reference Trevisan, Foss-Feig, Naples, Srihari, Anticevic and McPartland51] as well as co-diagnoses [Reference Lai, Kassee, Besney, Bonato, Hull and Mandy52] have been reported. Social cognitive deficits have also been demonstrated in other psychotic disorders, such as bipolar disorder, mood disorders (e.g., major depression), eating disorders, and personality disorders [Reference Cotter, Granger, Backx, Hobbs, Looi and Barnett53]. While careful clinical assessments of other symptoms for accurate differential diagnosis are warranted, cross-validation of overlap and divergence in underlying mechanisms can enhance individualized treatment optimization, independent of specific clinical categorical diagnosis [Reference Luyten, Campbell, Allison and Fonagy54].

Current cross-cultural landscape of social cognition

As part of regular practice, clinicians and researchers in schizophrenia should conduct a careful social cognitive assessment when examining the overall cognitive profile of an individual. We recommend the use of measures that have been validated and have good psychometric properties, not only in the country of origin where the measure was created, but that also received a thorough cross-national and cross-cultural validation process if available, to ensure construct validity invariance across and within cultures and languages. The NIMH “Social Cognition Psychometric Evaluation” (SCOPE) study was designed to create consensus in the schizophrenia field on the main core domains of social cognition using a panel of experts and to identify measures that would most reliably assess those domains, emphasizing their potential use in intervention studies [Reference Pinkham, Penn, Green, Buck, Healey and Harvey5, Reference Pinkham, Harvey and Penn6, Reference Pinkham, Penn, Green and Harvey55]. It launched in 2013 and involved five phases until 2018. Phase 1 included surveying experts in the field to identify the core domains and assessments of social cognition; in phase 2, the most promising social cognition tasks were selected by the experts using the RAND consensus method; and in phase 3 the candidate measures were administered to 179 individuals with SZ and 104 healthy controls (HCs), and their psychometric properties were examined. In phase 4, the candidate measures with inadequate psychometric qualities were modified and pilot tested for postpsychometric assessment, and in phase 5, a final validation study was conducted with the final measures in 218 individuals with SZ and 154 HCs to assess their psychometric properties and their relationship with functional outcomes. The following measures were recommended for assessing the core constructs of social cognition in schizophrenia, especially in clinical trials [Reference Pinkham, Harvey and Penn6]:

• - Emotion processing: The Bell Lysaker Emotion Recognition Task (BLERT) [Reference Bell, Bryson and Lysaker56] with suitable use in clinical trials and the Penn Emotion Recognition Task (ER-40) [Reference Kohler, Turner, Bilker, Brensinger, Siegel and Kanes57] with adequate psychometric properties but with limitations [Reference Pinkham, Harvey and Penn6].

• - ToM/mental state attribution: The hinting task (Hinting) [Reference Corcoran, Mercer and Frith58] with suitable use in clinical trials and the reading the mind in the eyes test (Eyes) [Reference Baron-Cohen, Wheelwright, Hill, Raste and Plumb59] as suitable but with concerns on its reliance on verbal content and equivalence of alternate forms, with final recommendation to treat it with caution. The awareness of social inferences task (TASIT) [Reference McDonald, Flanagan, Rollins and Kinch60] was considered a possible candidate with similar concerns as the Eyes test [Reference Pinkham, Harvey and Penn6].

• - Social perception: None of the evaluated measures in social perception had strong psychometric properties to be considered advisable for clinical trials. The Mini Profile of Nonverbal Sensitivity (MiniPONS) [Reference Bänziger, Scherer, Hall and Rosenthal61] and the Social Attribution Task-MC (SAT-MC) [Reference Bell, Fiszdon, Greig and Wexler62] were considered to be candidates but they did not show adequate psychometric properties [Reference Pinkham, Harvey and Penn6].

• - Attributional style: Similar to the domain of social perception, none of the measures of attributional style were recommended; however, the intentional bias task (IBT) [Reference Rosset63] was brought up as a possible candidate for further study [Reference Pinkham, Harvey and Penn6].

The expert authors did not recommend the Ambiguous Intentions Hostility Questionnaire (AIHQ) [Reference Combs, Penn, Wicher and Waldheter64], the relationship across domains (RAD) task [Reference Sergi, Fiske, Horan, Kern, Kee and Subotnik65] nor the trustworthiness task (Trust) [Reference Pinkham, Harvey and Penn6, Reference Adolphs, Tranel and Damasio66] because of their inadequate psychometric properties [Reference Pinkham, Harvey and Penn6]. Additionally, a recent large systematic review by Vita et al. [Reference Vita, Gaebel, Mucci, Sachs, Barlati and Giordano40] providing extensive guidance recommendations by the European Psychiatric Association (EPA) on assessment of cognitive and social cognitive impairments agreed with the Pinkham et al. [Reference Pinkham, Harvey and Penn6] study on the use of the BLERT and the Hinting task to assess emotion processing and ToM, respectively [Reference Pinkham, Harvey and Penn6]. However, it also included in their recommendations the use of the TASIT to assess these processes. Based on the inadequate psychometric properties shown in the studies reviewed, they discouraged the use of RAD, MiniPONS, and the SAT-MC for social perception as well as discouraged the use of the AIHQ, and the trustworthiness task to assess attributional style [Reference Vita, Gaebel, Mucci, Sachs, Barlati and Giordano40].

The SCOPE study as well as the EPA recommendations on the assessment of social cognitive impairments also concurred on the use of the Mayer-Salovey-Caruso Emotional Intelligence Test (MSCEIT) [Reference Mayer, Salovey and Caruso67, Reference Mayer, Salovey, Caruso and Sitarenios68]. The MSCEIT was originally developed to measure emotional intelligence (EI) and was divided into four branches or skill groups of EI: (a) perceiving emotion, (b) using emotion to facilitate thought, (c) understanding emotions, and (d) managing emotions; each of them being measured by two tasks. The MSCEIT was originally validated in English in 2,112 speaking adult individuals across seven countries (the United States, South Africa, India, Philippines, the United Kingdom, Scotland, and Canada)[Reference Mayer, Salovey and Caruso67, Reference Mayer, Salovey, Caruso and Sitarenios68], and it has also been validated in Italian [Reference Curci, Lanciano, Soleti, Zammuner and Salovey69], and Chinese [Reference Mao, Chen, Chi, Lin, Kao and Hsu70], Spanish [Reference Extremera, Fernandez-Berrocal and Salovey71], and German [Reference Steinmayr, Schütz, Hertel, Schröder-Abé and Huber72]. Because its good psychometric properties had already been tested before the SCOPE study [Reference Eack, Pogue-Geile, Greeno and Keshavan73, Reference Nuechterlein, Green, Kern, Baade, Barch and Cohen74] and the EPA recommendations, they were not included as part of their studies although it was nominated as a candidate measure for examining social cognition [Reference Pinkham, Penn, Green, Buck, Healey and Harvey5, Reference Vita, Gaebel, Mucci, Sachs, Barlati and Giordano40]. Additionally, the managing emotions (MEs) branch of the MSCEIT was selected to be included in the Measurement and Treatment Research to Improve Cognition in Schizophrenia (MATRICS) initiative [Reference Nuechterlein, Green, Kern, Baade, Barch and Cohen74, Reference Hellemann, Green, Kern, Sitarenios and Nuechterlein75], as the representative measure of social cognition, although it specifically measures emotion management and emotion relations (MSCEIT-ME). Helleman et al. [Reference Hellemann, Green, Kern, Sitarenios and Nuechterlein75] validated the MSCEIT-ME in several cultural contexts and languages such as Hindi, Japanese, Simplified Chinese, Russian, Spanish for Spain, and Spanish for Central and South America, and currently the MATRICS Consensus Cognitive Battery (MCCB), which includes the MSCEIT-ME offers versions in numerous languages and norms in Chinese Simplified, German, Hindi, Italian, Japanese, Russian and Spanish from Spain, and Spanish from Central and South America [Reference Green, Harris and Nuechterlein76] (https://www.matricsinc.org/available-language-versions/).

It is important that clinicians and researchers use the above-mentioned validated measures (i.e., Hinting Task, BLERT, and ER-40) in their own language and cultural framework, if available. However, these may not be readily adapted and validated in many countries with different languages and cultural and societal context. As pointed out by Hajdúk et al. [Reference Hajdúk, Achim, Gouet, Mehta and Pinkham77], even though some measures may have adequate psychometric properties in the country of creation, that cannot be equally assumed for their “translated targeted” counterpart in other countries. A psychological test developed to measure a construct in one language and country cannot be presumed to measure the same construct similarly when translated into another cultural and linguistic environment [Reference Byrne78]. In the realm of social cognition, this is even a greater possibility, as social cognitive stimuli may employ, for example, paragraphs portraying social interactions embedded in a cultural context, facial stimuli with specific ethnicity/race and specific cultural attire, or video clips in which the speaker communicates a message that is appropriate for a specific culture context. For example, the BLERT, as it includes only video clips portraying a Caucasian male, may show lower construct validity in other non-Western-English speaking countries or even within the same country but as a cross-racial effect. Pinkham et al. [Reference Pinkham, Kelsven, Kouros, Harvey and Penn79] examined the race effect on the performance in the BLERT and the Reading the Mind in the Eyes Test in patients with SZ and HCs and found that African American participants in both groups performed worse than their Caucasian counterparts on the Eyes test and BLERT test, both tasks which exclusively use images of Caucasian individuals for visual stimuli. Similarly, Adams et al. [Reference Adams, Rule, Franklin, Wang, Stevenson and Yoshikawa80] compared native Japanese and white American healthy individuals’ performance on variations of the Eyes test, composed of either Caucasian or Asian facial images. Participants showed improved task performance and increased brain activity when tested using images of people with shared cultural background.

Another example of a measure to treat with caution when using it cross-culturally is the Hinting task, as it has a high verbal reliance for measuring ToM and it would need a thorough cultural adaptation besides language. An example of the low cross-cultural construct invariance of this measure is demonstrated by Lim et al. [Reference Lim, Lee, Pinkham, Lam and Lee81]. Similar to the SCOPE study, they conducted a psychometric evaluation of cognitive tasks in Singaporean patients with SZ and HC. Notably, all participants in this study were fluent in English, and tasks were administered verbatim, without any modifications to the original versions. While overall results were relatively consistent with the SCOPE study, the Hinting task, which was one of the most strongly recommended tasks in SCOPE, showed noticeably less favorable psychometric properties when used in an Asian sample. This reinforces the notion that certain tasks may not be suited for use in non-western cultures, irrespective of language [Reference Vita, Gaebel, Mucci, Sachs, Barlati and Giordano40, Reference Lim, Lee, Pinkham, Lam and Lee81], and the need to take into consideration the contrast between Western individualistic cultures and Asian collectivistic cultures [Reference Wu and Keysar82, Reference Markus and Kitayama83].

As research interest grows beyond the anglosphere, such insight has led experts to call for a more “international perspective” that considers the potential impact of cultural differences on various aspects of social cognition. This may involve taking a more contextual approach to adapting tasks (rather than a simple, literal translation), retesting the validity and utility of tasks for individual cultural settings, or finding culture-specific assessment tools [Reference Hajdúk, Achim, Gouet, Mehta and Pinkham77, Reference Mehta, Thirthalli, Gangadhar and Keshavan84].

The need for international measures

As described by Byrne [Reference Byrne78, Reference Byrne85], it is important to note that the need of adaptation of measures can be necessary “cross-nationally” and “cross-culturally,” as there can be national political, language, and cultural differences within one country and cultural differences across nations. Therefore, we aimed for a broader definition of this process, such as “multigroup adaptation”. Despite the methodological difficulties to conduct a thorough “multigroup adaptation” of social cognition measures, it is important to maximize its rigorousness as it is the only manner to propel international scientific research and collaboration. In a letter to the Editor of Schizophrenia Research, Mehta et al. [Reference Mehta, Thirthalli, Gangadhar and Keshavan84] strongly advocated for the need of social cognition measures to be validated cross-culturally, and most current efforts in research agree with that statement [Reference Pinkham, Harvey and Penn6, Reference Vita, Gaebel, Mucci, Sachs, Barlati and Giordano40, Reference Hajdúk, Achim, Gouet, Mehta and Pinkham77], although there is still work to do in this area. Given this imperative need and as a way to move forward this field of research, The Schizophrenia International Research Society (SIRS), in 2022, created and awarded their Research Harmonization Award on the topic of social cognition (SIRS, Research Harmonization Award, 2022).

There has been an attempt to assess the cross-cultural adaptability of neurocognitive measures in schizophrenia [Reference Velligan, Rubin, Fredrick, Mintz, Nuechterlein and Schooler86, Reference Gonzalez, Rubin, Fredrick and Velligan87] which created a precedent for this type of work. Velligan et al. [Reference Velligan, Rubin, Fredrick, Mintz, Nuechterlein and Schooler86] created a survey, called the Cross-Cultural Adaptation Rating Scale (C-CARS) to be completed by researchers and clinicians around the world to assess whether the functional outcome measures they use would be of appropriate use in their culture. In a subsequent study, these authors provided a qualitative assessment of the potential cross-cultural adaptation of functional outcome measures and provided recommendations for international studies [Reference Gonzalez, Rubin, Fredrick and Velligan87]. We propose that future efforts in social cognition research may use, as a starting point, a similar approach and survey researchers around the world on the cultural validity and suitability of the measures in different cultures and languages. With the compilation of their input, researchers can then provide specific cultural adaptation guidelines for each specific social cognitive measure.

Many of the current measures used to assess social cognition have been developed in North America and Western Europe, and substantial current efforts are being conducted in many countries to develop culturally valid social cognitive assessment batteries with adequate psychometric properties to be used locally. Some examples of these current efforts are the Social Cognition Rating Tools in Indian Setting (SOCRATIS) developed by Mehta et al. [Reference Mehta, Thirthalli, Naveen Kumar, Mahadevaiah, Rao and Subbakrishna88] in which the authors validate social cognitive measures for ToM, social perception, and attributional bias. Recently, researchers in Japan, created an expert panel to provide consensus on the use of social cognitive measures for Japanese culture in “The Evaluation Study for Social Cognition Measures in Japan” (ESCoM) [Reference Okano, Kubota, Okubo, Hashimoto, Ikezawa and Toyomaki89, Reference Okano, Kubota, Okubo, Hashimoto, Ikezawa and Toyomaki90]which attempted to systematically investigate the utility of social cognition tasks for Japanese schizophrenia patients and HCs, using both Western-developed tasks translated to Japanese and tasks originally developed in Japan (ESCoM) [Reference Kubota, Okubo, Akiyama, Okano, Ikezawa and Miyazaki91]. In France, the Multicentric Research Group in Psychiatry (GDR3557 – Institut de Psychiatrie) developed a battery for social cognition that would measure the four social cognitive domains identified by the SCOPE study called “ClaCos” [Reference Morel-Kohlmeyer, Thillay, Roux, Amado, Brenugat and Carteau-Martin49, Reference Peyroux, Prost, Danset-Alexandre, Brenugat-Herne, Carteau-Martin and Gaudelus92, Reference Stoppelbein, Kechid, Morel-Kohlmeyer, Thillay, Roux and Amado93].

Adaptation process of social cognitive measures to propel international research

Global cross-cultural international research in psychological assessment has increased exponentially in the last decades, and there has been an increase in psychological assessment scales being translated into multiple languages [Reference Byrne78]. Additionally, societies have become increasingly diverse. For example, the European Consortium on Cross-Cultural Neuropsychology (ECCron) recently published a position statement on the impact of diversity in culture, education, and language in neuropsychological testing and emphasized the pressing need to develop and implement more culturally validated tests on social cognition and language and advocated for improvement in clinical training in culturally sensitive neuropsychological testing [Reference Franzen, Watermeyer, Pomati, Papma, Nielsen and Narme94]. Given the social and cultural nature of social cognitive measures, they may be more vulnerable to cross-nationally and cross-culturally differences, and therefore, different adaptations may be needed within one nation to ensure culture sensitivity and validity and more reason that properly trained clinicians administer them. However, it is important to note that the cross-national and cross-cultural adaptation process of a measure requires extensive work that may need the expertise of psychometricians, linguists, statisticians, and psychologists, and that the mere translation of a psychological test does not guarantee its cross-cultural equivalency [Reference Byrne78]. We recommend using expert cross-cultural guidelines such as the ones provided by the International Test Commission (ITC; https://intestcom.org) for adapting tests cross-nationally and cross-culturally. Now, these guidelines are in their second edition, which was prepared between 2005 and 2015 enhancing the first edition with the newer technology advances and practices https://www.intestcom.org/files/guideline_test_adaptation_2ed.pdf

Additionally, Byrne [Reference Byrne78] expanded on the ITC guidelines, describing that there are three types of options when developing a measure to use in cross-national research: adoption, adaptation, and assembly. The adoption strategy relies on linguistic equivalence between the original and the new translated measure. The adaptation strategy, in addition to translation, includes psychometric adaptation. The assembly option creates a new linguistically and culturally accurate scale. For cross-cultural research in psychology, the most feasible would be adaptation consistent with the ITC Guidelines on Adapting Tests by Byrne [Reference Byrne78]. A starting point that is recommended before beginning the process of adaptation is to ensure that there is cross-cultural construct equivalence across the different national and cultural environments between the original and the target population [Reference Byrne78]. Below are the adaptation stages:

1. 1) Translation stage of adaptation: It involves three substeps, which include a (1) translation from the source to the new language, then (2) the test is translated back to the original language (back translation), and (3) finally using expert translators team approach, the test is examined for any discrepancies in the translations.

2. 2) Pilot and field-testing stage of adaptation: In this stage, poorly functioning and biased items are identified, and it involves testing the instrument with a small (pilot) and/or a large sample (field) representative of the new group. In this stage, sometimes researchers use cognitive interviews in which they survey a representative sample to explain the logic of their responses and their interpretation of the items.

3. 3) Construct validation stage of adaptation: This stage requires extensive psychometric statistical testing of the measure and validation of the underlying factors structure of the new measure, pattern of factor loadings, and relationship among factors with the original measure.

4. 4) Norming stage of adaptation: This stage focuses on creating norms for the new measure in the country where it will be used.

Given the above information, when a clinician and/or researcher decides to conduct social cognitive testing in their practice or research, it is important to use research and clinical judgment to select the appropriate measure to assess a specific social cognitive domain. Even though there is no current existing battery that covers all the social cognitive domains [Reference Vita, Gaebel, Mucci, Sachs, Barlati and Giordano40], we recommend using the following guidelines:

• • Ideally, the four social cognitive domains from the SCOPE trial would be assessed (emotion perception, ToM, social perception, and attributional bias). However, the current measures that assess two of the domains, social perception, and social attribution, have inadequate psychometric properties and we advise their use prudently, taking into consideration their limitations.

• • Emotion perception and ToM are the domains in which the measures retained by SCOPE presented with the largest construct validity and most robust psychometric properties (BLERT, ER-40, and Hinting Task). Therefore, these should be prioritized in the social cognitive evaluation. If the measures do not originated in the same country/culture/language, then the clinician/researcher should seek a thoroughly cross-culturally adapted version of the original measures, desirably beyond mere translation. (See Table 1 adapted from Vita et al. [Reference Vita, Gaebel, Mucci, Sachs, Barlati and Giordano40]).

• • The results from the emotion perception and ToM measures should be utilized first for the customization of a treatment plan, and the outcome of the social perception/attributional style measures (if administered as well) should assist as additional information for confirmation of the initial plan. It is then recommended to follow the steps laid out in Section Assessment and treatment of social cognitive impairments in schizophrenia for the assessment and treatment of social cognitive impairments (See Figure 1).

Table 1. Adaptation and expansion of the table of measures of social cognition in schizophrenia from Vita et al., 2022

^a Original validation study.

^b Translation/adaptation only.

^c Translation/adaptation with psychometric information.

^d Psychometric information without translation.

Figure 1. Flowchart for the assessment and treatment of social cognition in schizophrenia.

Assessment of social cognitive deficits in schizophrenia

Assessment of social cognition performance should always be included in the evaluation of the global cognitive profile of an individual diagnosed with schizophrenia in clinical practice. These assessments should include different domains of social cognition, namely, emotion processing, ToM, and social perception [Reference Vita, Gaebel, Mucci, Sachs, Erfurth and Barlati95]. The assessment should be carried out using validated instruments with good psychometric properties, such as those identified by the SCOPE project [Reference Pinkham, Penn, Green, Buck, Healey and Harvey5, Reference Pinkham, Harvey and Penn6, Reference Pinkham, Penn, Green and Harvey55] or recommended by the EPA, which are described in detail in the previous paragraphs of the present document [Reference Vita, Gaebel, Mucci, Sachs, Erfurth and Barlati95].

This assessment should be aimed at identifying both strengths and weaknesses of the subject in the different social cognitive domains, and, while carried in the context of a broader cognitive evaluation, its results should not be influenced by those obtained in neurocognitive tests (Figure 1; Figure 2). Finally, this assessment should be conducted not only to have a better understanding of the overall cognitive profile of the patient but also and most importantly to implement dedicated and effective treatment, as social cognition deficits represent an important predictor of impaired psychosocial functioning and a clear predictor of worse real-world outcomes [Reference Deste, Vita, Nibbio, Penn, Pinkham and Harvey96–Reference Maj, van Os, De Hert, Gaebel, Galderisi and Green99].

Figure 2. Assessment of social cognitive deficits.

Note.

** Measures recommended by the SCOPE trial and Vita et al., 2022.

* Measures considered by the SCOPE trial, but they did not show adequate psychometric properties.

^† See Table 1 for a list of these clinical tools translated/adapted/validated to other languages.

From the results of this assessment, three different possibilities emerge (see Figure 1 and Figure 2): (I) a social cognitive profile showing no substantial impairment, (II) deficits in specific social cognition domains, and (III) a global social cognition impairment including multiple domains with different degrees of severity; the majority of individuals living with schizophrenia will fall in the latter two categories [Reference Green, Horan and Lee4, Reference Green, Horan and Lee100, Reference Penn, Sanna and Roberts101]. Individuals showing no social cognition impairment represent a minority in the context of schizophrenia and could benefit more from other evidence-based psychosocial interventions such as social skills training [Reference Turner, McGlanaghy, Cuijpers, van der Gaag, Karyotaki and MacBeth102], cognitive-behavioral therapy [Reference Turner, Reijnders, van der Gaag, Karyotaki, Valmaggia and Moritz103, Reference Wood, Williams, Billings and Johnson104] or metacognitive training [Reference Penney, Sauvé, Mendelson, Thibaudeau, Moritz and Lepage105], rather than from interventions specifically targeting social cognitive performance. These patients, having no specific deficits in social cognition abilities, could also represent ideal candidates to be trained to lead peer-led interventions and provide peer support [Reference Ahmed, Doane, Mabe, Buckley, Birgenheir and Goodrum106–Reference Sun, Yin, Li, Liu and Sun108]. For individuals showing social cognition deficits, providing social cognition training is again recommended with the highest grade available by recent EPA guidance [Reference Thibaudeau, Achim, Parent, Turcotte and Cellard28, Reference Vita, Gaebel, Mucci, Sachs, Barlati and Giordano40, Reference Thibaudeau, Cellard, Reeder, Wykes, Ivers and Maziade109]. For individuals showing deficits only in specific social cognition domains, focused interventions targeting only the impairment domains could be considered. However, these patients and patients with global social cognition deficits have been shown to consistently benefit from social cognitive remediation interventions targeting multiple domains [Reference Yeo, Yoon, Lee, Kurtz and Choi110, Reference Nijman, Veling, van der Stouwe and Pijnenborg111].

Several validated social cognition training interventions and programs described later are currently available for implementation into clinical practice. To provide significant improvements in real-world outcomes, however, all these interventions should either include bridging sessions to help transfer social cognition gains into everyday functioning or be carefully integrated into a structured psychiatric rehabilitation program; alternatively, they should be provided alongside other evidence-based rehabilitation interventions that can facilitate the transfer of social cognitive performance improvements into real-world functioning [Reference Yeo, Yoon, Lee, Kurtz and Choi110–Reference Vita, Barlati, Ceraso, Nibbio, Ariu and Deste112].

At the conclusion and during the follow-up period of the social cognition training intervention, social cognition performance should be reassessed to track and measure potential treatment-related improvements. If significant improvements with normalization are observed, the patient should be appropriately followed, provided with other evidence-based interventions, and considered for peer-led interventions and peer-support training. If the response to the social cognition training intervention is partial or absent, secondary causes of social cognition deficits in schizophrenia should be carefully assessed. This includes potential comorbidity with ASD [Reference Dell’Osso, Carpita, Cremone, Gesi, D’Ermo and De Iorio113–Reference Pinkham and Sasson115], as well as primary and secondary negative symptoms [Reference Galderisi, Kaiser, Bitter, Nordentoft, Mucci and Sabé116]. Potential causes of secondary negative symptoms include persistent positive symptoms, antipsychotic medication side effects, such as excessive sedation and extrapyramidal symptoms, illicit substances and alcohol use, depressive symptoms, and social deprivation [Reference Galderisi, Kaiser, Bitter, Nordentoft, Mucci and Sabé116–Reference Kirschner, Aleman and Kaiser118]. Considering these issues, re-evaluation of pharmacological treatment should be considered, and implementing dedicated and targeted interventions might be necessary to tackle these problematic clinical dimensions. In this context, if the underlying cause of secondary negative symptoms is identified, treated, and resolved, persisting social cognition impairments might benefit from another trial with a social cognition training program. Finally, social cognition performance should be periodically reassessed in all patients alongside global cognition and real-world functioning outcomes to track and measure potential treatment-related improvements or other changes, to provide an effective and personalized long-term rehabilitation program. Also see Supplementary Material for clinical approaches for recognizing socio cognitive deficits.

Treatments for social cognitive deficits in schizophrenia

There have been several social cognition interventions developed in the last twenty years, and they vary in terms of modalities and how social cognition is targeted. The interventions listed below provide a representation of the different modalities available, but we note that this is not an exhaustive list of all the available interventions, as it is beyond the scope of the present review. These interventions can be divided into (a) targeted, which are centered on solely improving a specific social cognitive domain, (b) comprehensive, which focus on several social cognitive domains and (c) broad-based, which may target several social cognitive domains, but are embedded within a broad psychosocial treatment that may address other areas of functioning [Reference Fiszdon and Reddy119]. Additionally, newer psychotherapies have also shown to be efficacious in improving social cognition, such as metacognitive therapies and interventions using newer technologies [Reference Horan and Green120–Reference d’Arma, Isernia, Di Tella, Rovaris, Valle and Baglio122]:

1. A) Targeted interventions have mostly been designed to address deficits in emotion processing and facial affect recognition, ToM, and social perception [Reference Fiszdon and Reddy119]. Training of affect recognition (TAR) [Reference Wölwer, Frommann, Halfmann, Piaszek, Streit and Gaebel123] and theory of mind intervention (ToMI) [Reference Bechi, Spangaro, Bosia, Zanoletti, Fresi and Buonocore124] are programs which exclusively target particular social cognitive domains. Imitation training (IT) focuses on observation and imitation of facial emotions and targets at affect recognition and can target at ToM as well [Reference d’Arma, Isernia, Di Tella, Rovaris, Valle and Baglio122, Reference Pino, Pettinelli, Clementi and Mazza125, Reference Mazza, Lucci, Pacitti, Pino, Mariano and Casacchia126]. The instrumental enrichment program (IEP) [Reference Roncone, Mazza, Frangou, De Risio, Ussorio and Tozzini127] targets at ToM by conducting role plays and showing to participants new social situations and assist them in the interpretation [Reference Fiszdon and Reddy119, Reference d’Arma, Isernia, Di Tella, Rovaris, Valle and Baglio122].

2. B) Comprehensive interventions such as the social cognitive and interaction training (SCIT) [Reference Penn, Roberts, Munt, Silverstein, Jones and Sheitman128], the social skills training (SCST) [Reference Horan, Kern, Shokat-Fadai, Sergi, Wynn and Green129] or Remédiation de la Cognition Sociale Dans la schizophrénie (RC2S) [Reference Peyroux and Franck130] target at a wide arrange of social cognitive domains and have been implemented internationally [Reference Horan and Green120]. The SCIT has been translated into multiple non-English languages and employed in several countries such as India [Reference Kumari, Sayeed, Das, Bose, Umesh and Khanande131], Finland [Reference Voutilainen, Kouhia, Roberts and Oksanen132], China [Reference Wang, Roberts, Xu, Cao, Yan and Jiang133], Japan [Reference Kanie, Kikuchi, Haga, Tanaka, Ishida and Yorozuya134], Israel [Reference Hasson-Ohayon, Mashiach-Eizenberg, Avidan, Roberts and Roe135], Turkey [Reference Tas, Danaci, Cubukcuoglu and Brüne136], and Spain [Reference Lahera, Benito, Montes, Fernández-Liria, Olbert and Penn137], and similarly, the SCST in Egypt [Reference Gohar, Hamdi, El Ray, Horan and Green138] and Korea [Reference Lim, Kwon, Jung, Park, Lee and Lee139]. The efficacy of these interventions has been tested over the years. A review and effect-size analysis by Kurtz et al. [Reference Kurtz, Gagen, Rocha, Machado and Penn140] of 16 controlled comprehensive interventions of social cognition using seven different models of treatment revealed large effect sizes in the facial affect recognition domain (ES=0.84), moderate-to-large effect sizes in ToM (ES=0.7), and a medium effect size in attributional style (ES=0.30-0.5). A most recent meta-analysis by Yeo [Reference Yeo, Yoon, Lee, Kurtz and Choi110] examining 42 studies with social cognitive interventions without additional effects of other behavioral interventions, revealed moderate effect sizes in social perception (g=0.46), and emotion recognition (g=0.55) and small to moderate effect in ToM (g=0.36). Overall, evidence indicates that social cognitive interventions may be of value to implement as a means to improve social cognition given the effect sizes observed in the above-described studies [Reference Horan and Green120].

3. C) Broad-based interventions may vary considerably in their methods and the extent of their target on social cognitive modalities [Reference Fiszdon and Reddy119]. Cognitive remediation therapies usually include sessions focusing on social cognition. They primarily address attention, memory, problem-solving skills, and executive functions [Reference Tan, Lee and Lee141]. Despite often not focusing on social cognition specifically, these methods have shown improvements in measures of global social cognition and social adjustment [Reference Thibaudeau, Cellard, Reeder, Wykes, Ivers and Maziade109, Reference Tan, Lee and Lee141]. Additionally, there are social cognitive computerized remediation programs that specifically target social cognition such as SocialVille [Reference Hooker, Carol, Eisenstein, Yin, Lincoln and Tully142]. Cognitive enhancement therapy (CET) also focuses on social cognition providing neurocognitive training as a “building block” [Reference Hogarty and Flesher143], and more recently, a multicomponent intervention, ecological training, has been developed that targets at social cognition by presenting participants with scenarios describing real-world conditions, such as clips of popular movies or real-life clips [Reference Fiszdon and Reddy119, Reference d’Arma, Isernia, Di Tella, Rovaris, Valle and Baglio122, Reference Bechi, Bosia, Spangaro, Buonocore, Cocchi and Pigoni144].

The above-described interventions have developed from multiple early psychosocial interventions which included social skills training, cognitive remediation, and cognitive-behavioral therapy (CBT) [Reference Fiszdon and Reddy119]. These individual or group-based therapies for social cognition include role play, learning through repetition, and computerized exercises with the end goal of enhancing daily functioning [Reference Penn, Roberts, Munt, Silverstein, Jones and Sheitman128, Reference Tan, Lee and Lee141]. Although there are still many questions to answer in this area, such as effect sizes when combining treatment modalities compared to active control conditions, treatment-dose effects and the generalization of social cognitive gains to social functioning, the outcomes observed are promising and support pursuing of more research in this area.

A new social cognition method (SCEILESS-Social Cognition Explicit Learning and Social Skills) is currently developed in France combining explicit learning of emotion and perspective-taking, social scripts, social perception, flexibility, and ToM. This method includes 30 sessions, social cognition and psychosocial skills to increase transfer to daily life. SCEILESS is proposed for persons with schizophrenia and for persons with autism, who have difficulties recognizing emotions, reduction in empathy, and have difficulties mentalizing in ToM. SCEILESS is delivered in a trans-nosographic approach with homogenous profiles of social cognition [Reference Martinez, Alexandre, Mam-Lam-Fook, Bendjemaa, Gaillard and Garel50] after an evaluation of the CLACOS battery [Reference Peyroux, Prost, Danset-Alexandre, Brenugat-Herne, Carteau-Martin and Gaudelus92]. A pilot study is an ongoing process.

Delineating the neural mechanism of social cognition in schizophrenia

Understanding the underlying neural mechanisms of social cognition in schizophrenia has potential implication not only for understanding its origin but also for diagnosis and individualized treatment development. The multidimensional construct of social cognition is also reflected in recent neuroimaging studies demonstrating unique brain networks subserving different social cognitive processes [Reference Schurz, Radua, Tholen, Maliske, Margulies and Mars171–Reference Assaf, Kahn, Pearlson, Johnson, Yeshurun and Calhoun175]. For example, it has been shown that the emotional processing network include the dorsal anterior cingulate cortex, insula and amygdala among other regions [Reference Fusar-Poli, Placentino, Carletti, Landi, Allen and Surguladze173] and the ToM network includes regions in the medial prefrontal cortex, posterior cingulate cortex, precuneus and temporo-parietal regions [Reference Yang, Rosenblau, Keifer and Pelphrey172, Reference Assaf, Kahn, Pearlson, Johnson, Yeshurun and Calhoun175]. In a meta-analysis summarizing neuroimaging studies of brain activity during empathy and ToM processes, Schurz and colleagues [Reference Schurz, Radua, Tholen, Maliske, Margulies and Mars171] proposed innovative models building a hierarchical clustering analysis. The results suggested a multilevel model of social cognitive processes with overarching networks that are flexibly combined and can relate to sensory affective versus more abstract representations across empathy and mental states of others [Reference Lahera, Benito, Montes, Fernández-Liria, Olbert and Penn137]. Several studies to date have revealed altered patterns of brain activation in schizophrenia during the performance of social cognitive tasks. In one of the first studies in the field, Gur and colleagues [Reference Gur, McGrath, Chan, Schroeder, Turner and Turetsky176] demonstrated decreased activation in the left amygdala and bilateral hippocampus in patients compared to controls during facial emotional processing [Reference Gur, McGrath, Chan, Schroeder, Turner and Turetsky176]. Multiple subsequent studies demonstrated similar results, but not all were consistent [Reference Sugranyes, Kyriakopoulos, Corrigall, Taylor and Frangou177, Reference Jáni and Kašpárek178]. Abnormal activations in brain regions of the ToM network were also observed [Reference Sugranyes, Kyriakopoulos, Corrigall, Taylor and Frangou177, Reference Jáni and Kašpárek178]. However, the relationship of these aberrant neural correlates to symptoms or to everyday functioning is still not fully understood and is crucial to our ability to develop neuroscience-based intervention methods and follow-up on their long-term brain effect. Moreover, as social cognitive impairments are transdiagnostic (as discussed in Section Social cognition as transdiagnostic phenotype) the specificity of neural deficits underlying social cognition has to be evaluated as suggested by NIMH Research Domain Criteria (RDoC) initiative [Reference Gur and Gur7]. One area of comparison is with ASD [Reference Sugranyes, Kyriakopoulos, Corrigall, Taylor and Frangou177, Reference Pinkham, Hopfinger, Pelphrey, Piven and Penn179–Reference Hyatt, Wexler, Pittman, Nicholson, Pearlson and Corbera181], but more research is still needed in this area. In addition, future research should further emphasize using ecologically valid tasks to better link differences to everyday social functions [Reference Hildebrandt, Jauk, Lehmann, Maliske and Kanske182] and identifying subgroups of patients with distinctive profiles and thus underlying etiologies [Reference Vaskinn and Horan24]. Defining different etiology levels of social cognitive processes in schizophrenia and transdiagnostically could help identify early markers for neurodevelopmental diagnoses, genetic risk factors, and predictors of risk for later social adaptive difficulties and lead to recommendations for preventive and therapeutic interventions [Reference Bond and Drake148].

Innovative treatments to improve social cognition

Digital society and change in social cognition

Two seminal NIMH workshops in the early 2000s conducted to conceptualize social cognition and its domains catapulted the scientific effort to investigate social cognition in schizophrenia [Reference Green, Penn, Bentall, Carpenter, Gaebel and Gur25, Reference Green, Olivier, Crawley, Penn and Silverstein209]. We are now harvesting the knowledge of more than 15 years of research after those workshops; however, during these years, the world experienced the worldwide COVID-19 pandemic that caused many unprecedented global health challenges [Reference Gruber, Prinstein, Clark, Rottenberg, Abramowitz and Albano210], and with its challenges, the world had to adapt to novel ways of interacting. We are now in a post-COVID-19 era in which social interactions are extremely dependent on technology, especially among the younger generations. As global international researchers that investigate social cognitive mental processes, it is important that we include as part of our current and future research new measures of addressing social cognition in the digital environment, such as social media [Reference Vaskinn and Horan24]. Hence, our way to interact has considerably changed in the last twenty years. We use smartphones to interact, and we communicate through social networks. Therefore, our social skills and our capability to communicate has drastically shifted. These new forms of communication are also broadly used in populations with psychiatric disorders. No studies exist that are investigating these new types of social interactions and their impact on people with social cognitive impairments. This is a new area that we propose needs further understanding. New measures may need to be created to gather clinical vs nonclinical samples to capture their comprehension, perception, understanding, and responses within social media interactions.

Physical exercise and social cognition

Finally, physical exercise has been proposed in schizophrenia as a mean to improve physical health, psychiatric symptoms, neurocognition, social cognition, and ultimately, social functioning [Reference Cui, Liu, Liang and Zhao211–Reference Firth, Stubbs, Rosenbaum, Vancampfort, Malchow and Schuch215]. Most of the studies in this area have focused on examining the effect of different types of exercise modalities (e.g., aerobic exercise, strength training exercise high intensity interval training (HIIT)) and other body-related activities such as yoga in overall improvement of physical health and symptoms [Reference Cui, Liu, Liang and Zhao211, Reference Gallardo-Gómez, Noetel, Álvarez-Barbosa, Alfonso-Rosa, Ramos-Munell and del Pozo Cruz212, Reference Vila-Barrios, Carballeira, Varela-Sanz, Iglesias-Soler and Dopico-Calvo214]. Just a few studies have specifically studied if physical exercise improved social cognition [Reference Firth, Stubbs, Rosenbaum, Vancampfort, Malchow and Schuch215–Reference Behere, Arasappa, Jagannathan, Varambally, Venkatasubramanian and Thirthalli218].

Overall, several meta-analyses concur on the positive effects of physical exercise on health, positive and negative symptoms and cognition and social cognition in individuals with schizophrenia [Reference Gallardo-Gómez, Noetel, Álvarez-Barbosa, Alfonso-Rosa, Ramos-Munell and del Pozo Cruz212–Reference Firth, Stubbs, Rosenbaum, Vancampfort, Malchow and Schuch215]. Specifically, aerobic exercise has been shown to improve cognitive functioning in individuals with schizophrenia [Reference Firth, Stubbs, Rosenbaum, Vancampfort, Malchow and Schuch215] and enhance the impact of cognitive remediation on cognition and social functioning [Reference Nuechterlein, SC, Ventura, Subotnik, Turner and Boucher219]. In terms of specific social cognitive-related changes, of the few studies that have examined exercise related changes in social cognition, most have used the social cognition domain of the MATRICS Consensus Battery (MCCB) [Reference Nuechterlein, Green, Kern, Baade, Barch and Cohen74, Reference Kimhy, Vakhrusheva, Bartels, Armstrong, Ballon and Khan216, Reference Nuechterlein, Ventura, McEwen, Gretchen-Doorly, Vinogradov and Subotnik217]. Kimhy and colleagues [Reference Kimhy, Vakhrusheva, Bartels, Armstrong, Ballon and Khan216] reported a significant correlation between body mass index (BMI) and the MCCB domains of working memory, social cognition, and speed of processing and that found an association between aerobic fitness and the MCCB social cognition domain, which was specially driven by the Social Management Task of the MSCEIT. Similarly, Nuetcherlein and colleagues [Reference Nuechterlein, Ventura, McEwen, Gretchen-Doorly, Vinogradov and Subotnik217] compared the effects of cognitive training alone vs cognitive remediation + physical exercise in individuals with schizophrenia and the latter group showed significantly higher scores of social cognition, using the social cognitive domain of the MATRICS. A systematic review and meta-analysis conducted by Firth and colleagues [Reference Firth, Stubbs, Rosenbaum, Vancampfort, Malchow and Schuch215] revealed a significant improvement of social cognition (g=0.71; p=0.22) with exercise, with greater improvements associated with higher dosages of exercise.

Emotion recognition skills have been found to improve using yoga as an add-on treatment in antipsychotic stabilized individuals with schizophrenia [Reference Behere, Arasappa, Jagannathan, Varambally, Venkatasubramanian and Thirthalli218]. Yoga has been found to improve cognitive skills [Reference Gothe and McAuley220], reduce negative symptoms [Reference Varambally, Gangadhar, Thirthalli, Jagannathan, Kumar and Venkatasubramanian221], and improve quality of life.[222]. After six weeks and 20 sessions of yoga therapy, persons with schizophrenia improved on their social cognition composite scores, negative and positive symptoms, and social disability [Reference Govindaraj, Naik, Mehta, Sharma, Varambally and Gangadhar223]. Despite the optimistic results of these studies, adherence to an exercise regimen presents to be a major challenge for individuals with schizophrenia, with studies showing high drop-out rates, and low engagement in the routines [Reference Gallardo-Gómez, Noetel, Álvarez-Barbosa, Alfonso-Rosa, Ramos-Munell and del Pozo Cruz212, Reference Nuechterlein, SC, Ventura, Subotnik, Turner and Boucher219, Reference Schwaiger, Maurus, Lembeck, Papazova, Greska and Muenz224–Reference Le, Ventura, Ruiz-Yu, McEwen, Subotnik and Nuechterlein226]. Motivation factors may underlie in the engagement in the participation in exercise routines [Reference Vila-Barrios, Carballeira, Varela-Sanz, Iglesias-Soler and Dopico-Calvo214, Reference Schwaiger, Maurus, Lembeck, Papazova, Greska and Muenz224–Reference Le, Ventura, Ruiz-Yu, McEwen, Subotnik and Nuechterlein226], and new digital health technologies may assist with exercise engagement such as using digital text messaging [Reference Ruiz-Yu, Le, Ventura, Arevian, Hellemann and Nuechterlein225]. Schwaiger and colleagues [Reference Schwaiger, Maurus, Lembeck, Papazova, Greska and Muenz224] examined the predictors of exercise adherence in schizophrenia and found that baseline levels of functioning in daily life were the best predictors, underscoring the importance of social functioning for engaging in an exercise regimen.

Notwithstanding the promising results of these initial studies, questions remain to be further investigated such as the differential effects of different single exercise modalities or combinations of them, dose-response outcomes depending on exercise duration and intensity and the enduring effects of exercise at the short and long term.