Overview of Social Cognitive Dysfunctions in Rare Developmental Syndromes With Psychiatric Phenotype

Rare neurodevelopmental syndromes often present social cognitive deficits that may underlie difficulties in social interactions and increase the risk of psychosis or autism spectrum disorders. However, little is known regarding the specificities of social cognitive impairment across syndromes while it remains a major challenge for the care. Our review provides an overview of social cognitive dysfunctions in rare diseases associated with psychiatric symptoms (with a prevalence estimated between 1 in 1,200 and 1 in 25,000 live births: 22q11.2 deletion syndrome, Angelman syndrome, Fragile X syndrome, Klinefelter syndrome, Prader–Willi syndrome, Rett syndrome, Smith–Magenis syndrome, Turner syndrome, and Williams syndrome) and shed some light on the specific mechanisms that may underlie these skills in each clinical presentation. We first detail the different processes included in the generic expression “social cognition” before summarizing the genotype, psychiatric phenotype, and non-social cognitive profile in each syndrome. Then, we offer a systematic review of the social cognitive abilities and the disturbed mechanisms they are likely associated with. We followed the PRISMA process, including the definition of the relevant search terms, the selection of studies based on clear inclusion, and exclusion criteria and the quality appraisal of papers. We finally provide insights that may have considerable influence on the development of adapted therapeutic interventions such as social cognitive training (SCT) therapies specifically designed to target the psychiatric phenotype. The results of this review suggest that social cognition impairments share some similarities across syndromes. We propose that social cognitive impairments are strongly involved in behavioral symptoms regardless of the overall cognitive level measured by intelligence quotient. Better understanding the mechanisms underlying impaired social cognition may lead to adapt therapeutic interventions. The studies targeting social cognition processes offer new thoughts about the development of specific cognitive training programs, as they highlight the importance of connecting neurocognitive and SCT techniques.

Rare neurodevelopmental syndromes often present social cognitive deficits that may underlie difficulties in social interactions and increase the risk of psychosis or autism spectrum disorders. However, little is known regarding the specificities of social cognitive impairment across syndromes while it remains a major challenge for the care. Our review provides an overview of social cognitive dysfunctions in rare diseases associated with psychiatric symptoms (with a prevalence estimated between 1 in 1,200 and 1 in 25,000 live births: 22q11.2 deletion syndrome, Angelman syndrome, Fragile X syndrome, Klinefelter syndrome, Prader-Willi syndrome, Rett syndrome, Smith-Magenis syndrome, Turner syndrome, and Williams syndrome) and shed some light on the specific mechanisms that may underlie these skills in each clinical presentation. We first detail the different processes included in the generic expression "social cognition" before summarizing the genotype, psychiatric phenotype, and non-social cognitive profile in each syndrome. Then, we offer a systematic review of the social cognitive abilities and the disturbed mechanisms they are likely associated with. We followed the PRISMA process, including the definition of the relevant search terms, the selection of studies based on clear inclusion, and exclusion criteria and the quality appraisal of papers. We finally provide insights that may have considerable influence on the development of adapted therapeutic interventions such as social cognitive training (SCT) therapies specifically designed to target the psychiatric phenotype. The results of this review suggest that social cognition impairments share some similarities across syndromes. We propose that social cognitive impairments are strongly involved in behavioral symptoms regardless of the overall cognitive level measured by intelligence quotient. Better understanding the mechanisms underlying impaired social cognition may lead to adapt therapeutic interventions. The studies targeting social cognition processes offer new thoughts about the development of specific cognitive training programs, as they highlight the importance of connecting neurocognitive and SCT techniques. iNTRODUCTiON Neurodevelopmental syndromes are genetic abnormalities fre quently associated with behavioral and/or psychiatric pheno types. Notably, these syndromes increase the difficulties in social interactions and present a high risk to develop psychosis or autism spectrum disorders (ASDs) (1)(2)(3). While social difficulties are generally associated with several comorbidities of genetic condi tions, such as intellectual impairments, facial dysmorphology, speech problems, and psychotic symptoms, the high prevalence of social cognitive disorders recently received a growing interest. The aim here is thus to provide an overview of social cognitive abilities in developmental syndromes associated with psychiatric symptoms and shed some light on the specific mechanisms that may underlie these skills in each clinical presentation. We focus on syndromes with a psychiatric phenotype (psy chosis and/or ASD) and a prevalence estimated between 1 in 1,200 and 1 in 25,000 live births: 22q11.2 deletion syndrome (22q11.2DS) (4)(5)(6), Angelman syndrome (AS), Fragile X syn drome (FXS) (7), Klinefelter syndrome (KS), Prader-Willi syn drome (PWS) (8,9), Rett syndrome (RS), Smith-Magenis syndrome (SMS) (10), Turner syndrome (TS), and Williams syn drome (WS) (11). We first detail the different processes included in the generic expression "social cognition" before summarizing the genotype, psychiatric phenotype, and nonsocial cognitive profile in each syndrome. Then, we offer a systematic review of the social cognitive abilities and the disturbed mechanisms they are likely associated with. We finally provide insights that may have considerable influence on the development of adapted therapeutic interventions such as social cognitive training (SCT) therapies specifically designed to target the psychiatric phenotype.

Social Cognition
Social cognition is defined as the ability to understand, perceive, and interpret information about other people and ourselves in a social context. This includes abilities such as emotion recognition, theory of mind (ToM), attributional style, and social perception and knowledge (Figure 1). Broadly speaking, social cognition includes a wide range of processes that allow people to rapidly, effortlessly, and flexibly perceive and interpret rapidly changing social information, and respond appropriately to social stimuli. Besides, this ability gives meaning to the actions of others. More specifically, social cognition is an "umbrella concept" that includes many heterogeneous cognitive dimensions, such as emotional information processing, social perception and knowledge, ToM, and attributional bias (12).

Emotional Information Processing
Emotional information about others is conveyed by complex signals such as prosody and emotional expressions from the face or the body (13). Regarding emotional expressions, Ekman (14) described six basic emotions (happiness, sadness, fear, disgust, anger, and surprise) that can be easily discriminated from one another depending on different facial patterns (14,15). Such categorical perception requires the processing of fine face changes and specific observation strategies.

Theory of Mind
An individual has a ToM if he imputes mental states to himself and others. A system of inferences of this kind is properly viewed as a theory because such states are not directly observ able, and the system can be used to make predictions about the behavior of others (16)(17)(18). This theory comprises two dimen sions: cognitive ToM (beliefs, intentions, etc.) and affective ToM (emotional state, knowledge of emotion, etc.) (19,20). The ToM is subtended by first and secondorder representations. The firstorder representation is defined as the ability to understand another person's mental state. The secondorder representa tion is the ability to understand what one person thinks about another person's thoughts. These levels are both supported by the primary ability to understand how and why an individual does something (e.g., understanding motor action and eye direction) (16).

Attributional Style
Attribution refers to the way people explain their own and other people's behaviors (21), but individuals are not objective perceiv ers and sometimes attribution does not properly match reality. Some individuals suffer from attribution bias, i.e., they systemati cally over and underuse the available information, which leads to a misinterpretation of the world they live in.

Social Perception and Knowledge
Social perception refers to the understanding of social roles and rules that typically appear in social situations. Social knowl edge is the awareness of the behavior expected in different social contexts and interactions (12). This knowledge consists of social experience, education, and ritualized practices that are not necessarily explicitly communicated. These processes are the basis of an accurate analysis of another person's inten tions, desires, or emotional states; more generally, it is linked to ToM.

Search Strategy
A search of the literature was conducted using the electronic database PubMed and Google Scholar, covering the period between first January 1990 and December 2016 and focusing on articles published in English. Broad search terms were used, including "social cognition, " "theory of mind, " "mind reading, " "mentalizing, " "mentalising, " "emotion recognition, " "emotion perception, " "emotion processing, " "affect recognition, " "affect perception, " "affect processing, " "social knowledge, " "attributional style, " "facial emotion, " "auditory emotion, " "gaze processing, " "body posture, " "social cues" in combination with the diagnostic terms "22q11.2 deletion syndrome" (22q11.2DS), "Angelman syndrome, " "Fragile X syndrome" (FXS), "Klinefelter syndrome" (KS), "PraderWilli syndrome" (PWS), Rett syndrome" (RS), "SmithMagenis syndrome" (SMS), "Turner syndrome" (TS) and "Williams syndrome" (WS). Reference lists of the retrieved articles were also manually searched for relevant publications. All articles investigating the perception and/or recognition of affective stimuli (facial expression, prosody, and body language), ToM, social knowledge and attributional style in patients with rare neurodevelopmental syndromes were evaluated. Only stud ies that involved a behavioral task were included. We excluded from our evaluation the studies that only included neuroimaging data, the studies with nonhumans and the reviews, and studies that did not directly assessed social cognition such as studies using only a questionnaire assessing social behavior. A total of 256 articles were initially identified as potential papers for inclusion. After a comprehensive reading by the first author (Aurore Morel), 136 articles were excluded from the analysis (e.g., studies not focused on behavior or duplicates). A total of 120 papers met the inclusion criteria and were eligible for quality appraisal (40 duplicates).
All titles and abstracts of the identified studies were indepen dently assessed by two authors (Elodie Peyroux and Caroline Demily) for inclusion in or exclusion from the systematic review. Papers were excluded when inclusion criteria were not met (disa greements were resolved by discussion).

Data Analysis
We extracted relevant data from the 132 studies. A metaanalysis of the studies was not feasible given the variations among study designs, social cognitive assessments used, and populations.
A thematic analysis was performed to qualitatively synthesize the data, aiming at identifying specific syndrome patterns. Data were presented in a narrative form. The analytic process was car ried on through discussion between the authors (Aurore Morel, Arnaud Leleu, Elodie Peyroux, Emilie Favre, Nicolas Franck, and Caroline Demily).

Patients
We proposed a review analysis, reporting the number of subjects included, the nature of the control population, the statistical method used, and the results obtained for all current published studies (Tables 2-5).

ReSULTS
The large majority of studies assessed either emotion recognition or ToM whereas fewer studies documented attributional style and social knowledge. Moreover, in some area of social cogni tion, very few studies or no study at all were available for some syndromes, especially AS, RS, and SMS.

Recognition of Facial Emotion
Basic emotions defined by Ekman (14), namely joy, sadness, anger, fear, and surprise, are the first emotions recognized in faces by children with typical development (117). No obvious deficits in basic emotion perception were reported in females with FXS (37,79,81,84). However, males with FXS looked significantly longer at disgusted faces compared to neutral faces (75), with more difficulties to recognize sadness and fear (83). The same pattern of results was revealed in children with ASD. By contrast, individuals with RS spend less time than controls exploring key facial features (eye, nose, and mouth) for all emotions (118) and have difficulties to recognize emotional expressions (happiness, fear, and sadness) in general. These results may be due to increased attentional or informative value of negative over neutral (and positive) expressions (75).
People with 22q11.2DS have difficulties to discriminate facial emotions when they are moderately expressed (66), which can explain major deficits in social interactions, especially during childhood. Moreover, they have more difficulties to perceive negative emotions such as anger, fear, and sadness and to label emotions (64,69,71) while happiness, sadness, and surprise recognition seems efficient (59,67,68). A recent study using the CANTAB ERT test (morphed images with fast cover up times to avoid ceiling effects) showed that 22q11.2DS patients are significantly less accurate in detecting sadness and happiness (71). The same results are observed in males with FXS (73,74,78,80,83,119).
Patients with WS showed emotion recognition levels similar to controls with the same mental age or individuals with devel opmental disabilities (99-101, 104, 106, 107, 110, 114). They are more accurate in perceiving happiness, sadness, anger, and surprise (107) but perform more poorly on auditive stimuli for all emotions except for happiness, which seems relatively surprising considering the classical hyperacusis associated with the syn drome (103). Abnormalities in auditory processing in WS seem to be restricted to the perception of negative affective vocalizations, such as scream or gasp (96). However, children with WS had no difficulty to recognize vocal emotions compared to controls (94,95). Emotional information can be transmitted via both visual (e.g., facial expressions and gestures) and auditory (e.g., affective prosody) channels. In social interactions, both channels are com peting and require the integration of relevant multisensory emo tional information. When identifying facial affects, patients with WS are as accurate as controls in both congruent and incongruent conditions. But they are less efficient when identifying nonsocial affects (97). These results confirmed that emotionally evocative music facilitates the ability of patients with WS to recognize emo tional faces. They spend more time staring at the salient features of the face-particularly the eyes (106,108). Interestingly, the ability to discriminate and match faces with emotions was assessed in children with WS syndrome compared to PWS syndrome, or a nonspecific intellectual deficiency. The three matched groups performed at a similar level (115) without significant correlation between chronological age (CA) and performance on face percep tion tasks (107). Overall, neurocognitive deficits do not seem to be the only explanation of expression recognition impairments across syndromes, as illustrated in TS (88).
Adults with WS demonstrated a normal level of performance in the identification of emotions in unimodal conditions with face stimuli alone and in congruent facevoice multisensory contexts. However, they are less accurate to identify emotions in unimodal conditions with vocal emotions alone and even worse in multisensory incongruent stimuli conditions (95,98). This difference seems to be highly specific and central to understand the hypersociability of WS ( Table 1). Patients with WS may be specifically attracted by facial information, decreasing their attention to vocal information. Interestingly, when adults with WS are asked to identify complex mental states (such as worry, disinterest, etc.) from the whole dynamic face, their performance is similar to controls matched on CA (57). However, they have difficulties with "relief " and "distrust, " which may be related to hypersociability.
The deficit is more pronounced in PWS and WS, with difficul ties to recognize all basic emotions (8,99,101). A global emotion identification impairment was documented in 22q11.2DS, FXS, PWS, and WS but the recognition of happiness seems preserved in these syndromes (101). Regarding the recognition of complex emotions, patients with PWS are less efficient than intelligence quotient (IQ)matched individuals with WS. This result is con sistent with the distinct behavioral symptoms associated to these syndromes (complex psychological disorder with maladaptive behaviors vs. frequent hypersociability).
During a task assessing the recognition of anger, fear, disgust, or sadness, individuals with KS were less accurate in recognizing angry faces, unlike other facial expressions (120). Children and women with TS show a specific impairment for the recognition of fearful and angry faces, both when the stimulus is a whole face or the upper face only (85-89, 91, 92). More specifically, in TS, fear recognition is significantly weaker compared to other emotions, including anger (88), with difficulties to correctly identify affect tones (91). In a study using dynamic facial expressions of different emotional intensities, patients with TS, 22q11.2DS and controls did not show any deficit in recognizing fear (62,90). But patients with 22q11.2 are slower to recognize emotions (62) with a normal level of emotion recognition in auditory stimuli (70). By contrast, dynamic stimuli in WS do not change the accuracy of emotion recognition (114).
During an emotion recognition task using photographs, patients with 22q11.2 (59) spent less time than controls observing the eye region, as observed in FXS and ASD. They spent more time looking at the mouth (63,68), which may explain why they have less difficulty to recognize happiness, surprise and, to a lesser extent, disgust. This result is corre lated with the high prevalence of ASD in 22q11.2DS and FXS. By contrast, women with TS stare longer at the mouth, but only when looking at fearful faces. This result suggests a possible mechanism for selective fear deficit (89) across syndromes. Facial emotion recognition impairment may be related to an atypical observation of faces (121). Fewer fixations on the eyes region are related to difficulties to recognize specific emotions. This deficit is a core symptom in ASD controls who stare longer at the mouth for joy, fear, and disgust; at the eye for sadness; and at the eye and nose for anger (122).

Ability to Infer False-Belief in Others
An important mechanism of ToM is the ability to understand that other people's beliefs or representations about the world may differ from reality.
Children with 22q11.2 did not show major difficulties in ToM tasks (60). When it comes to PWS and WS, children had difficulties attributing mental states and understanding first order beliefs (98,116,123). These results were confirmed in a nonverbal picturesequencing task, but with a preservation of the understanding of intention, social script knowledge, and physical causeandeffect reasoning (105). These results suggest that language abilities do not influence performance on ToM in WS (116), considering the heterogeneity of the process in the syndrome (105).  ANCOVA The TS group had difficulty to recognize "anger" (p = 0.005), but no "happy," "sad," "fear," "surprise," and "disgust" ANOVA

T-test
The TD group outperforming the WS group (p = 0.004). The interaction effect was due to the WS group exhibiting lower recognition accuracy for negative stimuli (p = 0.001), while no between-group differences were in evidence for the processing of the positive stimuli (p =0.73)   Patients with FXS are less accurate in ToM tasks (76,77) and perform at a similar level than individuals with Down syndrome (DS) (73,74) or ASD (77), with a limited impact of autistic symptoms (124). Altogether, these findings suggest that a ToM disorder in FXS may not result solely from the high comorbidity of ASD (77).
Children with WS, PWS, or a nonspecific intellectual deficiency, perform at a similar level in standard ToM tasks (115,125). However, children with PWS had specific difficulties to attribute secondorder beliefs (123,126).
Interestingly, 22q11.2DS seems specifically associated with a poor performance on falsebelief tasks (60,61), this deficit being highly correlated with psychotic symptoms.

Ability to Infer Mental States in Others
Children with 22q11.2 seem to have difficulties to link contex tual and social relevant cues when the scene is visually complex (61) or based on circumstances (61). This deficit may be related to attentional and visuospatial disorders and requires families and therapists to clearly verbalize emotions during social inter actions. Moreover, the comprehension of misunderstanding, persuasion, pretending, sarcasm, and white lie seems impaired in 22q11.2DS (25,60,71). These findings suggest that a cogni tive ToM disorder causes a delay rather than a deficit (60). From this perspective, impairments in the spontaneous attribution of seducing, mocking, playing with one another, getting frightened, or elated mental states to abstract visual stimuli were reported in 22q11.2DS (65), suggesting that gaze direction is also a weak ness in the syndrome (60).
As expected, patients with WS were proficient during a verbal task (111,126) but their performance decreased when the ToM task was presented on a visual medium (111) with difficulties inferring mental states from visual cues, especially for negative intentions (93) and gaze direction (109). These results suggest that they may present a positive bias in the interpretation of social cues. An impaired socialperceptual ability may play a role in increased approachability and a deficit in the processing of basic social cues may have possible repercussions on ToM. The ability to observe and recognize other people's actions contributes to understand goals and intentions (127) and may help people with WS improve their recognition of action process (113). Indeed, in the presence of context cues, children with WS were as accurate as controls (112,113).
Overall, patients with 22q11.2DS, FXS, or WS had diffi culties determining whether someone is looking at them or elsewhere (60,82,102,128) but individuals with PWS seem to yield scores within the normal range (129). When patients with FXS were asked to judge the direction of eye gazes (direct or averted), they performed at the same level than patients with similar general cognitive abilities and autism symptoms (72), and patients with a developmental delay (82). Complex emotion recognition requires knowledge and analysis of interpersonal relationships, unlike basic emotion recognition. Thus, further research on the potential dissociation between the recognition of complex mental states and basic emo tions, and the influence of the nature of the stimuli (static or dynamic) is needed.

DiSCUSSiON interventions and Perspectives
This systematic review shows that social cognition deficits are present in neurodevelopmental disorders at different levels and remain strongly correlated with psychiatric phenotype. Emotion recognition and ToM skills impairments seem to be a core deficit in rare developmental syndromes, otherwise associated with psychosis and ASD. Although more research needs to be done to assess social cognition phenotypes in these syndromes, it is crucial to develop and evaluate appropriate therapeutic interventions. In this section, we briefly discuss therapeutic approaches and present future keys to improve social deficits in that indication.
In daily life, facial expressions are typically embedded in a rich context with many distractors (e.g., noises and visual stimuli) and social information from multiple sensory channels (e.g., prosody, gestures, and change in facial expression). Therefore, social interactions necessitate fast and efficient cognitive skills. They include the ability to identify visual and spatial relation ships in emotion recognition, to integrate and manipulate such perceptions, to select relevant information, to inhibit irrelevant stimuli and alternate between several sensitive channels. Emotion recognition is also a complex process involving visual attention (130), visuospatial abilities (130), working memory (131), divided attention and executive functions (132). As a consequence, improvement of these functions may have a positive impact on social cognition and behavior. Visuospatial and visuoperceptual skills play a key role in everyday life. Visual information and complex visual stimuli are analyzed with a complete unaware ness of the visuoperceptual process or the complexities of the stimuli involved. This process becomes conscious in a context of learning. Repetition and familiarity enable a more spontaneous approach and turn the conscious and effortful process into an automatic one. If this ability is impaired, many types of deficits can occur, ranking from a failure to process the basic elements of a visual stimulus (i.e., colors, lines, and orientation) to more complex and integrative features, such as object identification, faces, or familiar scenes. These deficits can include social cogni tion defects, especially in the area of facial emotion recognition, acquired slowly during childhood and reach adult levels in late adolescence (133).
Currently, two programs, named "Cognitus & Moi" (Cognitus & Me) and "VisàVis" have considered the link between neuropsychological functions and social cognition (facial emo tion recognition). "Cognitus & Moi" is a cognitive remediation program with SCT, designed for 5 to 13yearold children with developmental disorders. The program is specifically focused on attentional, visuospatial functions and emotion recognition (134). It involves a variety of exercises in a paper and/or pencil (n = 30) or a computerized format (n = 29) and a strategy coaching approach. "VisàVis" is a computerized training program based on a trial anderror approach. It can be performed at home with parents and targets social cognition difficulties and working memory (135).
The ability to explore face efficiently develops during child hood. The eyes, nose, and mouth are the preferred attentional targets in facial exploration (121). These facial features play an important role in perception and recognition of emotion. However, the visual exploration developed by individuals with neurodevelopmental syndrome is mainly different, with a lack of eye contact according to a specific pattern previously observed in ASD. Individuals with neurodevelopmental syndrome may ben efit from educational solutions targeting this impairment. In that regard, an important point is that dynamic information facilitates the perception of facial expressions (136,137). Individuals with autism appear to benefit from slow dynamic information when categorizing emotional expressions (138). Considering the pos sibility of overlap between children with neurogenetic disorders and children with typical ASD (139), the question of a similar benefit in rare diseases should also be addressed in the future. More generally, the beneficial effect of motion may play a role in therapeutic intervention. Some current treatment plans which are based on phenotype rather than etiology may be adapted for individual with neurode velopmental syndrome.
For example, considering the wellestablished association between 22q11.2DS and schizophrenia, SCT, a program based on the cognitive enhancement therapy developed for individuals with schizophrenia (140) may be of interest in 22q11.2DS. This program targets general and highlevel social skills such as the ability to practice calming techniques, to identify the main idea in a conversation, to determine expected behaviors in different situations (social perception and social knowledge), or to clearly express one's own thoughts (ToM) (140).
It is necessary to define strengths and weaknesses of each participant and the exact nature of deficits. Indeed this review highlights the implication of different cognitive processes involved in social interaction deficits, which appear to have different effects across syndromes (Tables 6 and 7). Here, we show that social cognition impairments are not uniquely caused by syndrome comorbidities such as IQ. Understanding the differences in the social cognitive abilities may become a  (141). With a training focused on emotion, dis crimination of facial emotion (joy, sadness, and anger) may be improved in RS (142). Moreover, although patients do not develop functional verbal skills, the eyetracking technology has increased the possibilities to understand patients (143) and facilitated the access to augmentative and alternative communication (144)(145)(146). Face recognition and ToM have an important role in social problem solving (14,147,148). These capacities allow to under stand and interpret other's emotion (149). In the social interaction, recognition of expressions leads to adapt behavior (150,151).

Limitations
This systematic review highlights some limits. Some rare syn dromes received only little interest, particularly AS, KS, RS, and SMS. The lack of results regarding some areas of social cognition in these syndromes must be interpreted with caution, as the number of published studies is limited. Furthermore, very few studies have directly compared patients considering anxiety levels. Moreover, psychiatric diagnosis has not been considered and very few studies have proposed to establish a link between social cognition and behavior in rare neurodevelopmental syndromes. In fact, the psychiatric phenotype has not yet been extensively studied. Many data are available for facial emotion recognition and ToM, but a lack of research in social perception/ knowledge and attributional style has been highlighted. Thus, it remains unclear whether individuals with a rare developmental syndrome show similar impairments in these areas. Moreover, impairments may be specific to tasks and emotions, rather than all social cognitive processes. The use of relatively consistent methodology across study designs would help compare these impairments.

Summary and Conclusion
To sum up, some social trainings, such as the use of the eye tracking technology in RS or the focus on visuospatial functions in 22q11.2DS, are created among the specific profile of individu als with a neurodevelopmental syndrome. Understanding the differences in the social cognitive abilities of people with rare neurodevelopmental disorders may lead to a more personalized medicine. This review suggests that cognitive remediation therapy and SCT may be a viable way to improve the social and functional outcomes, although many questions on its effectiveness still need to be answered.
The studies targeting social cognition processes offer new thoughts about the development of specific cognitive train ing programs, as they highlight the importance of connecting neurocognitive and social cognitive training techniques. They point out that individuals with social cognitive impairments may need more help in terms of social interactions. They provide keys for caregivers, who may better adjust their communication with explicit social cues by mentioning what they are thinking or feeling and why. Understanding and adjusting attitudes to individuals with social cognition impairments are required to prevent frustration and maladaptive behaviors.

AvAiLAbiLiTY OF DATA AND SUPPORTiNg MATeRiALS SeCTiON
Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.

AUTHOR CONTRibUTiONS
AM elaborated the PRISMA review. AM and CD wrote the paper and collected the data. EP, AL, EF, and NF collected the data. All authors approved the final version of the manuscript.