Keywords

Introduction

Social cognition refers to a wide range of cognitive abilities that allow individuals to understand themselves and others and also communicate in social interaction contexts (Adolphs, 2001). According to Adolphs (2009), social cognition deals with psychological processes that allow us to make inferences about what is happening inside other people—their intentions, feelings, and thoughts. Although the term can be defined in many ways, it is clear that it must be safeguarded for the mental operations underlying social interactions.

The most investigated cognitive processes of social cognition are emotion recognition and theory of mind (ToM), given that a whole range of socio-affective and interpersonal skills, such as empathy, derive from them (Mitchell & Phillips, 2015). Theory of mind is an intuitive ability to attribute thoughts and feelings to other people, and this ability usually matures in children in preschool age (Wellman, 1990), whereas emotional recognition refers to an individual’s ability to identify others’ emotions and affective states, usually based on their facial or vocal expressions, it is a critical skill that develops early and supports the development of other social skills (Mitchell & Phillips, 2015).

Adverse Childhood Experiences (ACEs), such as parental neglect or physical, sexual or psychological abuse, especially in the early stages of development, can have particularly harmful long-term consequences for the consolidation of cognitive, affective, and emotional skills (Herzog & Schmahl, 2018). A systematic review on the associations between early social environment, early-life adversity, and social cognition in major psychiatric disorders found that emotional and physical abuse, neglect, and avoidant attachment styles were the strongest predictors of ToM and emotion recognition deficits, as well as emotional dysregulation (Rokita et al., 2018). Prolonged exposure to events of this nature can lead to brain changes, particularly in circuits involved in regulating responses to stress, configuring the concept of toxic stress (Shonkoff & Garner, 2012).

In clinical or psychopathological contexts, social and affective impairments worsen social disadvantages that many patients face. As an example, in Autism Spectrum Disorder (ASD), ToM seems to be the most impaired social cognition domain (Baron-Cohen et al., 1985), whereas in schizophrenia, social impairments appear as negative symptoms and tend to predict patients’ lower mental capacity. Several evidences indicate that core social difficulties in ASD are best explained by deficits in controlled processes, such as ToM, rather than automatic ones, such as those dependent on emotional contagion (Hamilton, 2013). This discrepancy between the two seems to be in line with the hypothesis of empathy imbalance (Smith, 2009). As an example, the avoidance of eye or physical contact frequently observed in individuals diagnosed with this disorder can be explained by an exacerbated affective empathy, which is dependent on emotional contagion. On the other hand, a deficit of cognitive empathy, being associated with ToM, could justify the low performance in tasks with greater demand for mental state assignment.

Social cognition deficits are expressed, for instance, in difficulties to interpret social cues and regulate behavior accordingly, which turns engagement in social relationships especially challenging. In ADHD, there are also failures to recognize emotions, especially anger and fear (Bora & Pantelis, 2016). Contrary to what is observed in autism, however, the difficulties tend to be lighter and get better with age. Thus, deficits in social cognition are early and prominent features of many neuropsychiatric, neurodevelopment, and neurodegenerative disorders (Agnew-Blais & Seidman, 2013). In the most recent edition of the American Psychiatric Association’s Diagnostic and Statistical Manual for Mental Disorders (DSM-5, 2013), social cognition has been added as one of the six main components of neurocognitive function, alongside memory and executive control. In addition, the DSM-5 points out that impairments in social cognition often arise as deficient ToM, reduced affective empathy, impaired social perception, or abnormal social behavior.

In this chapter, we discuss social cognition processes in typical and atypical development and its relevance for the understanding of socio-affective disorders in childhood and adolescence. We expect therefore that it will be useful for clinicians, teachers, and perhaps also for parents.

Social Interactions and Theory of Mind in Childhood and Adolescence

Social interactions depend essentially on the exchange of social and affective cues, which can be verbal and non-verbal (Frith & Frith, 2007). Verbal cues include vocalizations, tone of voice, and speech content, while the most important non-verbal cues that humans use are facial expression, body posture, and eye monitoring. Most of these cues are processed automatically and unconsciously (Frith & Frith, 2007). In this way, regardless of an unambiguous interpretation, we automatically decipher people’s emotions through their facial expressions, while also looking for something interesting in the environment from clues such as the direction of others’ gaze. Moreover, curiously, we tend to imitate the behavior of people with whom we have a good relationship (Frith & Frith, 2007). In an eloquent way, social cues guide our behavior through an ambiguous and complex world.

In childhood (especially in the first two years of life), human beings have limited attention and limited working memory. Given these restrictions, one possible question to ask is if babies are able to process complex social information. Sensitivity to social clues seems to develop early, considering that even in early childhood, it is already possible to locate its clues from the orientation of attention in human babies (Michel et al., 2017). In Wu and Kirkham’s study (2010), 8-month-old babies were presented with two identical audiovisual events simultaneously in two different locations on one computer screen. The babies’ attention was focused on a stimulus that contained a social suggestion (a face saying “Hi, baby, look at this!” and that turned to a target event) or a non-social suggestion (a red square surrounding the target event). Both stimuli directed attention equally, as measured by the time the babies stayed looking at the events. However, only babies exposed to social cues predicted the location of the signaled events, suggesting that social attention hints shape the likelihood and content of learning about events during childhood.

Reid and Striano (2007) propose that for a baby to react successfully to a social situation, four stages of cognitive processing of the task must occur: (1) the detection of socially relevant organisms; (2) the identification of socially relevant organisms; (3) the evaluation of the place of attention and direction of the individual’s gaze observed in relation to the child; and (4) the detection of any attention directed to objects or involvement of objects by the observed individual. As a consequence, if the previous stage were successful, then another stage is reached; in the fifth stage, the baby is able to infer the observed objective and/or prepare an appropriate response (e.g., establishing contact).

Reid and Striano (2007) suggest that the detection of biological movement, an early cognitive ability in babies, plays a key role in the detection of co-specifics and, therefore, in identifying the specifics of social interaction in stage 1. Stage 2 is possible because babies can identify idiosyncrasies in the observed organisms, for example, discriminating between a familiar and unfamiliar individual. In relation to stage 3, these abilities are supported because human babies are sensitive to the elements of the human face, especially to follow the gaze of the observed subject. In addition, they are able to distinguish whether an adult interacts in a salient manner, providing contingent feedback on social interaction, such as smiles and vocalizations, or interacts in an irregular manner, with a delay in social return (Striano et al., 2005).

Finally, babies are competent in discerning a relationship between a person and an object (Stage 4). In another study, Reid and Striano (2005) found that the direction of a person’s gaze affects the coding of new objects in 4-month-old babies. In this study, in a first experimental condition, the babies saw the face of a woman with new toys being presented on her right and left sides. The woman then directed her gaze to one side, thus capturing a certain object, and as a consequence, she averted her gaze from the other toy. In a second experimental condition, the 4-month-old babies were presented to the two toys again and looked more at the toy that had not been looked at in the first condition, probably because of the little attention paid to it, and only after that to the toy looked at in the first condition. This suggests that 4-month-old babies not only follow the adult’s gaze in the first condition but also acquire more information about the object that was the focus of adult attention. It is quite convincing that human babies respond early to social cues, however, an idea valued for social and affective neuroscience is that in order to adequately manage the complex levels of social interaction that characterize our social life, human beings need to develop specific social-cognitive mechanisms, such as Theory of Mind.

Since Premack and Woodruff’s (1978) seminal article “Does the chimpanzees have a theory of the mind?” raised the question of chimpanzees’ ability to attribute states of mind to themselves and others, the subject of ToM has become part of human development studies. One of the first studies of ToM development as a child was the work of Wimmer and Perner (1983), who pointed to the understanding of false belief as an indicator of preschoolers’ ToM. One way to test this understanding is to place the child in a task where he or she has to predict the behavior of a character who has a belief that does not correspond to reality. An example would be to present the child with a common box of chocolates, then ask him/her to open the box to check its contents; After checking, to their surprise, that the box doesn’t have chocolates, but colored pencils, some questions are asked to the children, such as if you show a friend the same box of chocolates, and ask him/her what is the content, what do you think he/she would say? Why would he/she say that?

It is expected that children between the ages of 3 and 5 will be able to respond correctly to this type of task, thus revealing an understanding of beliefs about reality which, being personal representations of reality, maybe true or false. However, the ability to understand false beliefs seems to be more the result of an ongoing process of developing skills to assign mental and affective states to others. Therefore, some studies analyzing children’s discourse have found that references to desires precede references to cognition (Bartsch & Wellman, 1995; Peterson & Slaughter, 2006).

According to Wellman et al. (2011), the theory of mind develops progressively and sequentially from the child’s ability to understand different levels of representation of mental states: (1) various desires (people may have different desires for the same thing), (2) various beliefs (people may have different beliefs about the same situation), (3) access to knowledge (something may be true, but someone may not know it), (4) false belief (something may be true, but someone may believe in something different), and (5) hidden emotion (someone may feel one way, but show different emotions).

ToM’s reasoning required by different social situations may involve the assimilation of complex levels of intentionality. For example, Ygor believes (ToM of first order of intentionality) that Larissa thinks (ToM of second order) that her aunt Marcia wants (ToM of third order) that Ygor supposes (ToM of fourth order) that Larissa wants (ToM of fifth order) that her aunt Marcia believes (ToM of fifth order). This ability of first- and third-order ToM seems to improve with age (Dumontheil et al., 2010).

Dumontheil et al. (2010) showed that the ability to adopt the point of view of another agent grows from childhood, passes through adolescence, and improves even more in adulthood. Meinhardt-Injac et al. (2020) tested a two-component model of social cognition, social perception, and cognitive social, in a sample of 267 participants between 11 and 25 years of age. In addition, they measured language, reasoning, and inhibitory control as covariables. In the study, adolescents showed a substantial improvement in ToM (social perception and false belief) and covariable measures. An interesting finding is that the social perception component increased with age, while the socio-cognitive component (false belief) increased with age and covariables measures.

The tasks of false belief can be further developed between adolescence and adult life. In the study by Valle et al. (2015), adolescents performed significantly worse than young adults in tasks of false belief involving third-order ToM, but an equal result for second-order ToM. Other components of ToM develop in adolescence as the social knowledge required in tests involving soft lies, forgeries, and strange stories (Maylor et al., 2002; Bosco et al., 2014). Happé (1994) developed a test called The Strange Stories to evaluate advanced mental capacity, which is suitable for adolescents and adults with superior functioning. In the test, participants read short vignettes and were asked to explain why a character said something that is literally not true. Therefore, successful performance requires the assignment of mental states, such as desires, beliefs, or intentions, and sometimes higher-order mental states, such as one character’s beliefs in what another character knows.

Using a subset of Happé test stories (1994), Maylor et al. (2002) investigated individuals aged 16–29 when performing advanced ToM tasks in the first person and noted that participants scored an average of 4 out of an available maximum of 7, without any ceiling effect being achieved in the age group. In the study of Bosco et al. (2014), teen performance improved with age in all ToM scales, which investigate first-person and third-person ToM, first- and second-order allocating ToM, and egocentric third-person ToM. However, age differences were consistent between 11 and 13 years and then tended to stabilize between 13 and 15 years. The findings of these different studies suggest that some, but not all, components of ToM continue to develop into adulthood.

Development of Emotional Recognition and Understanding in Childhood and Adolescence

Children’s emotional knowledge comprises two distinct dimensions: recognition of emotion and knowledge of the emotional situation. Recognizing emotions means that the child can label facial expressions using expressive knowledge of emotions as well as identifying emotions when expressed with verbal labels.

Nine-month-old babies are already able to discriminate between positive and negative emotions (Otte et al., 2015). In the study, 84 babies received emotional vocalizations (fearful or happy) preceded by the same facial expression or a different expression (i.e., fearful vocalization with a happy expression). The data processing of emotional information (event-related potential, or ERPs) revealed that the potentials were distinct for positive and negative emotions, and that the babies dedicated more process capacity to potentially threatening stimuli than to non-threatening ones. Between 18 and 24 months, children are already able to acquire the necessary terms to label basic emotions, both positive and negative (Widen & Russell, 2003, 2010).

According to Pons et al. (2004), children develop emotional understanding from three levels (external, mental, and reflective) and use at least nine components for this. The first level (already found in children of 5 years) is characterized by the understanding of public aspects of emotions, such as situational causality, external expression, and cues that reactivate an emotion. The second level (developed at age 7) is characterized by an understanding of the mental states of emotions, their connection with desires and beliefs, between expressed and felt emotion. The third level (between the ages of 9 and 11) is characterized by the understanding that we can feel different feelings and that they can be contradictory and even morally charged.

Regarding the components, approximately between 3 and 4 years old, children begin to recognize, and name emotions based on expressive clues (recognition component). Thus, most children in this age group can recognize basic emotions (happiness, sadness, fear, and anger) when presented in images. Also at this age, they begin to understand how external causes affect other children’s emotions (external causality component). For example, they can anticipate the sadness someone feels when losing a favorite toy. Already around the age of 3 to 5, they begin to understand that the emotional reactions of people depend on their desires (component of desire). They are able to understand that two people can feel a different emotion about the same situation, because they have different desires.

Still, according to Pons et al. (2004), children between 4 and 6 years old begin to understand that a person’s beliefs—being false or true—will determine their emotional reaction to a situation (component of beliefs). Also, around this age, between 3 and 6 years, children start to understand the relationship between memory and emotion (memory component). For example, children can recognize that the intensity of an emotion decreases with time and that some aspects of the current situation can reactivate past emotions. Still, in this age group, 4 and 6 years old, children already understand that the emotion expressed can be different from the emotion felt (component of the cover-up). In sequence, between the ages of 6 and 7, they are able to use different strategies to regulate emotions (component of regulation), the younger ones use behavioral strategies, and those over 8 use psychological strategies such as denial and distraction. It is also from the age of 8 that they begin to understand that a person can have various and contradictory emotions (mixed component) to a given situation. Finally, 8-year-old children are able to understand negative feelings resulting from a reprehensible moral action (morality component) (like lying, stealing, and hiding), as well as being morally dignified (like a sacrifice, or resisting a temptation, or even confessing a mistake).

Emotional recognition improves with adolescence. Functional brain imaging studies during facial emotion recognition tasks have demonstrated an increase in activation and connectivity of frontal and temporal regions from childhood to adolescence (Cohen Kadosh et al., 2011, 2013). This increase in emotional recognition tends to continue into adulthood. The study by Tousignant et al. (2017), showed that adolescents perform worse than young adults in this area of social cognition.

Social and Affective Dysfunctions

To allow a better understanding of how social cognition impacts psychosocial functioning in children and adolescents, we will focus on the findings of research that has investigated this domain of cognition in some psychopathological conditions with the presence of social and affective dysfunctions, such as schizophrenia, ASD, ADHD, and impulse control disorders.

In the last two decades, the domains of social cognition have been intensively studied in individuals with neurological conditions, with genetic syndromes, and in population groups at risk for developing the first episode of psychosis, such as children of parents with schizophrenia. There is also broad evidence of these neurological conditions such as epilepsy (Besag & Vasey, 2019). Social cognitive deficits are part of the cognitive phenotype of genetic diseases, including deletion syndromes, such as SD22q11.2, and those related to X-Chromosome numerical alterations, such as Turner and Klinefelter (Morel et al., 2018). In a review study, Agnew-Blais and Seidman (2013) found that young people belonging to families at high biological risk for developing schizophrenia, especially siblings and children of individuals with schizophrenia, show deficits in aspects of social cognition, such as ToM and emotion recognition. However, it is not yet known whether deficits in the domain of social cognition follow a pattern of delays over time or are static. A population-based, prospective cohort study evaluated 7-year-old children at high familial risk of developing schizophrenia or bipolar disorder (Christiani et al., 2019). The authors found significant impairments in social responsiveness in children at risk for schizophrenia compared to controls, but not for children at risk for bipolar disorder compared to controls (Christiani et al., 2019). In several population-based cohort studies, children and adolescents who later developed schizophrenia showed premorbid social impairments (Tarbox & Pogue-Geile, 2008; Agnew-Blais & Seidman, 2013). These findings reinforce the view of schizophrenia as a neurodevelopmental disorder and the importance of identifying in clinical practice differences in the domain of social cognition in children and adolescents belonging to groups at risk for developing schizophrenia.

In ADHD, in addition to classical assessment of executive functions, deficits in the domain of social cognition are increasingly investigated (Mohammadzadeh et al., 2016). In a meta-analysis of studies investigating social cognition in ADHD, it was reported that facial and vocal recognition skills and ToM were significantly impaired in ADHD (Bora & Pantelis, 2016). In addition, the authors rated the performance of individuals with ADHD as intermediate between ASD and healthy controls (Bora & Pantelis, 2016). An interesting finding is that deficits in social cognition appear to occur later in ADHD than in ASD and appear to depend on social interactions with family members and peers at school (Bora & Pantelis, 2016). Not surprisingly, social dysfunction is one of the most impactful aspects for the psychosocial development of children and adolescents with ADHD, as individuals with ADHD often report significant interpersonal problems, including conflict with parents, siblings, peers, and teachers (Ros & Graziano, 2018). Social dysfunction in ADHD appears to depend on a number of factors, including social skills, ability to process information and modulate social responses (social cognition), and contexts of social interaction with peers (Ros & Graziano, 2018). Although these reviews seemingly converge on a view of deficits in social cognition in children and adolescents with ADHD, most clinicians seem to ignore that ADHD patients tend to experience not only limitations in social skills but also impairments in social information processing. Furthermore, ToM skills are associated with different patterns of prosocial behavior such as helping, cooperating, and comforting (Imuta et al., 2016), and since in ADHD ToM may be significantly impaired, these patients would be less likely to develop positive social interactions with their peers, teachers, and family members.

In impulse control disorders, social cognition is an aspect of patient functioning that has gained relevance. Impulse control disorders (ICDs) are grouped as a heterogeneous group of mental disorders related to the failure to resist impulses to perform dangerous, troublesome, or disturbing behaviors. We can include in this category, pathological gambling (PG), kleptomania, pyromania, trichotillomania, internet gaming disorder (IGD), intermittent explosive disorder (IED), among others. Aspects of the social cognition domain have been investigated in at least these last two clinical conditions (Coccaro et al., 2016). In IED, the biopsychosocial model of impulsive aggression holds that the individual with the disorder usually explodes in response to social threat, and one of the main dysfunctions would be in social information processing (Coccaro et al., 2011). Thus, the individual with IED, when encoding and interpreting social cues, often performs an attribution of hostile intent in social interactions (Coccaro et al., 2011). In a study investigating an aspect of social cognition of patients with IED, attribution style, demonstrated that participants with IED have higher hostile attribution in ambiguous social situations than healthy controls and patients with other psychiatric disorders, and also that attribution style was directly related to negative emotional response (Coccaro et al., 2016).

In the case of IGD, patients are observed to have significant social impairments, with a predilection for social interactions only in the online environment than in real life (Caplan, 2010). In the cognitive model of IGD proposed by Caplan (2010), pathological internet use could be defined by two cognitive features: preference for online social interaction and worry. Preference for online social interaction can be defined as an individual’s tendency to develop beliefs that online interactions and relationships are better, safer, and more comfortable than face to face (Caplan, 2010). Worry or cognitive salience is defined as obsessive thought patterns about internet use (Caplan, 2010). Some studies have investigated that a component of social cognition, emotion recognition, is related to problematic internet use in adolescents (Spada & Marino, 2017; Yavuz et al., 2019). Specifically, in relation to IGD, it is suggested that adolescents engage in gaming as a strategy to alleviate affective dysfunction caused by poor skill in recognizing negative emotions (Yavuz et al., 2019). Another study revealed that low negative emotion recognition skills were able to predict cognitive salience, tolerance, and relapse associated with IGD in adolescents (Aydın et al., 2020). These findings analyzed together allow us to think that negative emotion recognition, attribution style, and consequently the strategies used for cognitive-affective regulation used by adolescents aroused by negative affect may increase their risk of developing impulse control mental disorders, specifically IED and IGD.

Treatment

Behavioral interventions focusing on the stimulation of socio-emotional skills have been developed and their effectiveness tested in a more remedial or preventive perspective. For the follow-up of children and adolescents with a history of exposure to maltreatment, there are several approaches such as cognitive-behavioral therapy (CBT), Eye Movement Desensitization and Reprocessing (EMDR); therapies based on artistic activities or contact with animals, and family-focused interventions (e.g., systemic family therapy) (Macdonald et al., 2016).

In clinical conditions, such as autism, the most well-known interventions are those based on behavior analysis models. For children with high functioning autism, virtual reality-based training has proven to be a complementary strategy that can stimulate engagement in the therapeutic process (Didehbani et al., 2016). Simmons et al. (2019) proposed an integrative model for intervention programs considering social cognition and executive functions simultaneously. In fact, impairments in these cognitive domains have been associated with the symptomatology of both ASD and ADHD (Van der Meer et al., 2012). According to the model, for a target social and affective problem, intervention strategies should consider social cognitive (e.g. practices to recognition of emotional cues) or executive (behavioral rehearsal) components.

Other programs have a more preventive character, generally organized for implementation in schools. Emphasis on programs of this nature has been strengthened by evidence of the importance of developing socio-emotional skills, such as understanding one’s point of view or self-regulation, for academic performance (Blair, 2002; Fantuzzo et al., 2007). An example is Social and Emotional Learning, or SEL (Durlak et al., 2011). The program aims to stimulate the management of emotions and the development of empathy in order to strengthen positive relationships and more responsible decision-making. In a meta-analysis, Durlak et al. (2011) reviewed 82 school-based SEL interventions implemented inside and outside the United States, covering 97,406 kindergarten to high-school students. The programs were selected according to recommended practice criteria including a clear alignment of goals and curriculum, involvement of students in all steps of implementation, attention to community needs, and reflective activities such as class discussions. Results provided evidence of positive effects on students’ attitudes toward themselves and toward school and learning, on social behavior, and academic performance, independently of students’ race or socioeconomic background.

Conclusion

In this chapter, we describe how social cognitive processes, such as emotion recognition and ToM, develop. These processes are of paramount importance for the understanding of normal and pathological processes in childhood and adolescence. The continuous processing of social information and the perception of emotional states combine to build up a child’s emotional knowledge and to develop social competence.

The process by which we understand our own emotions and those of others, and assign intentions and desires to others, helps to explain the different challenges we face throughout human development. The abilities to recognize emotions and assign mental states, important domains of social cognition, are crucial to assessing someone’s immediate social environment, providing valuable information about the inner emotional state of others, and influencing adaptive social behavior and social interactions. The observation of social and affective dysfunction in children and adolescents raises the question of whether these deficits are risk factors for neuropsychiatric disorders, the untoward consequences of neuropsychiatric disorders, or both. Socio-affective neuroscience offers ways to elucidate this issue. In addition, in health sciences, there has been a focus on the early identification of developmental conditions in which a fast and rational intervention can favorably reach positive outcomes. In most neuropsychiatric disorders, difficulties in social functioning are evident, and if we can focus on the identification and remediation of social and affective dysfunction at an earlier age, the better.

We address how these aspects are related to healthy development and also in neuropsychiatric disorders. Since the skills involved in social cognition are extremely important for future interactions, diligent investigations of this phenomenon are useful for clinical and applied neuroscience. Bearing in mind that the understanding of the social and affective factors involved in children’s development is fundamental for the understanding of the adults that will emerge from it.