ORIGINAL RESEARCH published: 01 November 2021 doi: 10.3389/fpsyg.2021.721584 Self- but Not Other-Dimensions of Mentalizing Moderate the Impairment Associated With Social Anxiety in Adolescents From the General Population Sergi Ballespí 1* , Jaume Vives 2 , Jacqueline Nonweiler 1 , Ariadna Perez-Domingo 1,3 and Neus Barrantes-Vidal 1,3,4 1 Department of Clinical and Health Psychology, Universitat Autònoma de Barcelona, Barcelona, Spain, 2 Department of Psychobiology and Methodology of Health Sciences, Universitat Autònoma de Barcelona, Barcelona, Spain, 3 Department of Mental Health, Fundació Sanitària Sant Pere Claver, Barcelona, Spain, 4 Center for Biomedical Research Network on Mental Health (CIBERSAM), Instituto de Salud Carlos III, Madrid, Spain Mentalizing, or social cognition, refers to the brain’s higher order capacity that allows humans to be aware of one’s own and others’ mental states (e.g., emotions, feelings, Edited by: intentions). While cognition in social anxiety has been broadly analyzed, there is a Jose P. Espada, Miguel Hernández University of Elche, paucity of research regarding the role of social cognition. Moreover, mentalizing or social Spain cognition research is traditionally focused on the understanding of others’ mental states, Reviewed by: rather than self-mentalizing. Finally, most studies analyze the role of social cognition in Igor Marchetti, University of Trieste, Italy the development or maintenance of social anxiety, yet no study to date has analyzed Je-Yeon Yun, whether social cognition moderates functional impairment associated with it. This study Seoul National University Hospital, analyzes whether self- and other-mentalizing moderate the relationship between social South Korea anxiety and impairment in social and self-functioning. A sample of 262 adolescents *Correspondence: Sergi Ballespí from the non-clinical population was assessed on measures of social anxiety, self- and sergi.ballespi@uab.cat other- mentalization, indicators of social functioning (social competence and sociometric status), and indicators of self-functioning (depression and self-esteem). Multiple linear Specialty section: This article was submitted to regressions were conducted to test possible moderation effects of self-mentalizing Psychology for Clinical Settings, and other-mentalizing on the relationships between social anxiety and social and a section of the journal Frontiers in Psychology self-functioning. Results revealed that other-mentalizing does not moderate social- Received: 07 June 2021 nor self-functioning, while self-mentalizing moderates the impairment of all of them. Accepted: 13 October 2021 While impairment in social functioning is buffered by one dimension of self-mentalizing Published: 01 November 2021 (emotional clarity; b = 0.003, p = 0.043 and b = 0.016, p = 0.008 for social competence Citation: and sociometric status, respectively), impairment in self-functioning is strengthened by Ballespí S, Vives J, Nonweiler J, Perez-Domingo A and the other dimension (attention to emotions; b = −0.007, p = 0.008 and b = 0.009, Barrantes-Vidal N (2021) Self- but Not p = 0.047 for self-esteem and depression, respectively). Probing the moderation at Other-Dimensions of Mentalizing Moderate the Impairment Associated the 16th, 50th, and 84th percentiles revealed that the negative imbalance between With Social Anxiety in Adolescents dimensions (i.e., high attention and low clarity) tended to exacerbate impairment most From the General Population. on all indicators, while the positive imbalance (i.e., low attention and high clarity) Front. Psychol. 12:721584. doi: 10.3389/fpsyg.2021.721584 was usually the most buffering condition. This supports that “low-flying” or implicit Frontiers in Psychology | www.frontiersin.org 1 November 2021 | Volume 12 | Article 721584 Ballespí et al. Mentalizing Moderates Social Anxiety Impairment mentalizing provides more resilience than explicit mentalizing (i.e., high attention and high clarity). Findings suggest that the work on emotional self-awareness should be stressed in the intervention of the social anxiety spectrum conditions in order to improve prevention, functioning, and ultimately, treatments, of people impaired by symptoms of social anxiety. Keywords: social anxiety, self-other mentalizing, social cognition, emotional knowledge, self-other functioning, impairment, resiliency, prevention INTRODUCTION social behaviors and those of others” (American Psychological Association, 2020). This broad definition entails several Social anxiety is anxiety about social situations—specifically processes and dimensions, ranging from emotion recognition one’s performance and interactions, with a core fear of negative to attributional style or social knowledge (Plana et al., 2014), evaluation and judgment as being, for example, anxious, and has come to be referred to using various terms (social crazy, weak, intimidating or unlikeable (American Psychiatric intelligence, Theory of Mind, mentalization, and more) in the Association, 2013). The social anxiety spectrum encompasses literature interchangeably. This has promoted extensive term myriad phenomena sharing this fear (Schneier et al., 2002), dispersion and overlapping concepts. and ranges from non-clinical levels of shyness or behavioral In this context, the more recent paradigm of mentalizing inhibition to psychopathology (i.e., social anxiety disorder, provides a multidimensional perspective which systematizes avoidant personality disorder) (Stein et al., 2004). the field with an umbrella term, rooted in neuroscience Once clinical, social anxiety is a disorder that typically presents and supported by neurobiology (Frith, 1999; Frith and Frith, first in adolescence (75% of individuals experience first onset 2003; Denny et al., 2012; Luyten and Fonagy, 2015), allowing social anxiety between ages 8 and 15), exhibits prevalence rates researchers to gather related concepts and to reduce term- between 2 and 7% in the Western world, and is difficult to treat dispersion. The mentalization paradigm structures this higher (Faravelli et al., 2000; Fehm et al., 2008; Russell and Shaw, 2009; order cognition in four neuroscientifically-based dimensions or Mayo-Wilson et al., 2014). In spite of the stress and functional polarities (Luyten et al., 2020). Thus, defined as the brain’s impairment associated with social anxiety, only half of those capacity to notice one’s own and other’s mental states (i.e., affected by the disorder ever seek treatment, and those who emotions, feelings, intentions, desires) (Fonagy and Luyten, 2009; do so typically endure 15–20 years afflicted before pursuing Sripada et al., 2009), mentalization can be cognitive or affective, it (American Psychiatric Association, 2013). With the aim of explicit (deliberate), or implicit (automatic), based on external or avoiding this high statistic, the identification of maintenance on internal cues, and referred to one’s own (self-mentalizing) or factors and variables that moderate social anxiety could help to to others’ mental states (other-mentalizing) (Luyten et al., 2020). address social anxiety before it reaches clinical significance (e.g., While mentalizing and social cognition have been used as earlier in its developmental course). synonyms, the literature about social anxiety is primarily based While the role of cognition has been broadly analyzed in on social cognition, and despite reference to both self- and social anxiety (e.g., information processing biases) (Clark and other-behavior in social cognition, or “cognition in which people Mcmanus, 2002; Heimberg et al., 2014), the role of social perceive, think about, interpret, categorize, and judge their own cognition in this area has been less well-researched. Further, social behaviors and those of others” (American Psychological findings regarding social cognition are inconsistent; some studies Association, 2020), measures of social cognition have been have found difficulties with social cognition in individuals with traditionally referred to how we know or interpret others’ mental social anxiety (Banerjee and Henderson, 2001; Pile et al., 2017), states (i.e., others’ intentions or feelings in the social context). while others report the opposite finding (LaBounty et al., 2017), Consequently, there are very few studies analyzing the dimension arguing that social anxiety leads individuals to stop and observe of self-mentalizing (i.e., awareness of one’s own mental states before interacting which provides further development of social in the social context) in social anxiety. Moreover, while most cognition skills. Even still, some studies found no association studies analyze the role of social cognition in the development whatsoever (Batanova and Loukas, 2011; Broeren et al., 2013; and maintenance of social anxiety (Plana et al., 2014; Alvi et al., Colonnesi et al., 2017). More recently, a meta-analysis by Pearcey 2020), to our knowledge, no study has yet analyzed to what extent et al. (2020) revealed a small association between social cognition this higher order cognition moderates functional impairment in and social anxiety (r = −0.15). The low consistency of the findings individuals with social anxiety. beyond a simple low association can be attributed to the disparity Thus, the aim of this study is to analyze whether the separate in measures (experimental vs. ecological), populations (clinical dimensions self and other within mentalization moderate vs. non-clinical; different ages), and definitions both regarding the impairment experienced in social anxiety, which we social cognition and social anxiety phenomena, used in the operationalized in the current research as difficulties with social different studies (Plana et al., 2014; Pearcey et al., 2020). functioning and problems with self-functioning, specifically, level Social cognition is defined as “cognition in which people of self-esteem (usually affected in social anxiety) (Farmer and perceive, think about, interpret, categorize, and judge their own Kashdan, 2014; Iancu et al., 2015) and the level of depressive Frontiers in Psychology | www.frontiersin.org 2 November 2021 | Volume 12 | Article 721584 Ballespí et al. Mentalizing Moderates Social Anxiety Impairment symptomatology, which is often comorbid with social anxiety Attention to emotions and emotional clarity are not separate (Brady and Kendall, 1992; Lewinsohn et al., 1997; Brown et al., and independent processes (Boden and Thompson, 2017), 2001; American Psychiatric Association, 2013). and thus beyond isolated effects, their combined effect would As previously stated, associations between social anxiety be interesting to study. Though research is scant regarding and social cognition are primarily based on other-mentalizing the pairings of attention and clarity (possibly due to the (Battaglia et al., 2009). Given the inconsistent findings in this difficulty to interpret their interaction), some authors have matter (see Pearcey et al., 2020) and the gap with regard to hypothesized about their combined effect based on their self-mentalizing, it is not clear what to expect in terms of individual contribution. In their review, Davis and Nichols moderation of the impairment. While it is likely to assume (2016) conclude that excessive attention to emotions coupled that how we read others’ mental states is involved in social with lack of competency to elaborate them might be deleterious functioning (Ballespí et al., 2021), the insight about one’s for mental health. Further, Gohm et al. (2005) found fewer own mental states has been associated to emotional regulation stress symptoms when emotional clarity and attention were (Fonagy and Target, 2002; Fonagy et al., 2005; Hill and uniformly high or low (i.e., balanced), but higher stress in Updegraff, 2012; Greeson et al., 2014), as well as to other those individuals experiencing intense emotions but lack of processes of self- functioning (Ballespí et al., 2019). As such, emotional understanding. This is consistent to the emotionally we predict that other-mentalizing will more strongly moderate “overwhelmed” type described by Gohm (2003), which refers to the association between social anxiety and social functioning a combination of high affect intensity, intermediate attention to (i.e., the functioning in the social world) than self-mentalizing, emotions, and low clarity. Based on this literature, Boden and which will more intensely moderate impairment experienced Thompson (2017) conclude in their meta-analysis that people in self-functioning. In the current study, social functioning who attend highly to emotions but are unable to understand was operationalized using measures of social competence and them well may be more likely to become overwhelmed and to sociometric status, while self-functioning was operationalized have problems with emotion regulation. Kerns and Berenbaum using measures of self-esteem and internalizing symptoms, in this (2010) found, in five studies, that the overwhelmed type case depression. is associated with worse performance in different tasks. In Furthermore, two subdimensions of emotional self-awareness summary, extant literature suggests that the imbalance between have been determined by factor analysis: attention to emotions dimensions composed by higher attention to emotions than and emotional clarity (Mayer and Gaschke, 1988; Salovey et al., emotional clarity (further referred to as “negative imbalance”) is 1995; Mayer et al., 2016). According to Salovey et al.’s (1995) associated with worse mental health. Accordingly, we hypothesize model of meta-mood experience, Attention to emotions is defined that high values of attention to emotions combined with low as “the individual’s willingness to attend to feelings” or, in values of clarity could also magnify impairments associated other words, the magnitude of one’s attention dedicated toward with social anxiety. noticing emotions, while Emotional clarity refers to the ability to Regarding a possible protective effect, we also wonder which pinpoint and understand one’s own mood; this requires a deeper other combinations could buffer the impairment associated with awareness or understanding of feelings (i.e., discrimination social anxiety. There is no evidence about which combinations between different emotions, and perception and cognizance of attention and clarity lead to protective effects, and theoretical of them). Because self-mentalizing is a complex, higher order predictions are scant. On one hand, Gohm et al. (2005) suggested process, it would be incorrect to assume that simple attention that both high or low levels of attention and clarity (that is, to one’s own emotions equates to clear awareness or deep balance between dimensions) are better than an imbalance. understanding of the emotional states. Conversely, Salovey et al. (1995), and recently De la Barrera While both the attention and clarity dimensions of Salovey et al. (2021) theorized that high emotional clarity combined with et al.’s (1995) conception of emotional self-awareness are moderate attention could be the best option for adjustment and indicative of self-mentalizing, previous findings suggest that regulation. We will refer here to the imbalance composed by emotional clarity is more strongly associated with emotional lower attention than clarity as “positive imbalance.” However, regulation than simple attention to emotions (Extremera and in summary, there is evidence supporting clarity as the active Fernández-Berrocal, 2006; Salguero et al., 2012; Balluerka ingredient, though it lacks evidence about its combined effect et al., 2013; Resurrección et al., 2014; Vine and Aldao, 2014; with different levels of attention, so it is unclear whether clarity Eckland and Berenbaum, 2021). In fact, attention to emotions will be more protective when combined with high or low is occasionally associated with higher emotional dysregulation attention to emotions. (Gross, 2002; Gross and John, 2003; Thompson et al., 2009, 2013; Both social anxiety and social cognition reach high levels Davis and Nichols, 2016), especially when it is not combined in adolescence (Stein and Stein, 2008), a developmental stage with high emotional clarity (Ballespí et al., 2019, 2021), in which with high potential for early intervention or even prevention. case people are more likely to become overwhelmed and face Consequently, and in order to encompass the variability of the issues with emotion regulation (Gohm, 2003; Gohm et al., 2005; social anxiety spectrum, we based the study on an adolescent Kerns and Berenbaum, 2010). Accordingly, regarding the self- sample from the general population. Because both social anxiety dimension, we predict that emotional clarity will moderate the and social cognition show differences by sex and age (Aune association between social anxiety and impairment more strongly and Stiles, 2009) (i.e., girls are more mature in adolescence, than simple attention to emotions. and tend to mentalize better, but are also more prone to Frontiers in Psychology | www.frontiersin.org 3 November 2021 | Volume 12 | Article 721584 Ballespí et al. Mentalizing Moderates Social Anxiety Impairment social anxiety) (Asher et al., 2017), all the analyses will be (I usually have an optimistic outlook, although sometimes I feel controlled by age and sex. sad). The TMMS is evaluated on a 5-point Likert scale from 1 (strongly disagree) to 5 (strongly agree). Validity evaluations show moderate internal consistency (Cronbach’s α range from MATERIALS AND METHODS 0.82 to 0.87) and good convergent and discriminant validity. The Spanish version (Fernandez-Berrocal et al., 2004), utilized in this Participants research shows moderate-good internal consistency (Cronbach’s A sample of 262 adolescents (144 girls, 55%) between the ages α from 0.86 to 0.90), and acceptable test-retest reliability (r of 12 and 18 years (M = 14.6, SD = 1.7) from the general between 0.60 and 0.83). The current sample has excellent internal population agreed to participate in the study. This sample was consistency (α = 0.91 for the total score, α = 0.90 for attention to recruited through schools in the context of a broader project emotions, and α = 0.92 for emotional clarity). about psychopathology, personality and coping strategies in The Adolescent Mentalizing Interview (AMI) (Ballespí and adolescence. The inclusion criterion was to be between 12 and Pérez-Domingo, 2015) is a measure specifically designed to 18 years of age, and the exclusion criterion was presence of evaluate mentalizing in adolescence. It consists of two guided severe mental illness such as psychosis, autism spectrum disorder, exercises: the first one refers to the mental states of the characters or intellectual disability. Recruitment was carried out in the of a picture-based story and it is scored in 3 items; the second one schools to simplify logistics. Ten schools of similar characteristics asks about mentalizing in the relationship with two very-close (urbanicity, similar size, family SES, educational orientation, and others (family or close friends) (Bartholomew and Horowitz, methodologies, geographically close to each other) were invited 1991), using demand questions inspired in those used by Fonagy to participate in the project according to their proximity to the et al. (1998) in the Reflective Function Scale and scored through research center. Five of these schools agreed to collaborate, and 4 additional items. All 7 items are scored on a 4-point Likert scale n = 266 families signed the informed consent to participate in the from 0 (no mentalizing) to 4 (sophisticated mentalizing). The study. The principal reasons for refusal were low interest in the AMI provides a total score ranged from 0 to 28, based on one project, being too busy, discomfort in giving data about mental dimension which explains 64% of total variance and has excellent health or, in the case of some immigrant families, the inability to internal consistency (Cronbach’s α = 0.90) (Ballespí and Pérez- understand at least one of the two languages of the questionnaires Domingo, 2015). Concurrent validity is supported by correlations (i.e., Spanish or Catalan). It was possible to obtain self-reported with other measures evaluating mentalization (ranging from 0.21 data from adolescents in the 98% of cases (n = 262), and from to 0.47) and inter-rater reliability boasts independent interview parents and teachers in 95% (n = 254) and 84% (n = 223) of cases, correlations from 0.79 to 0.88 (ICC = 0.91 for the total score). respectively. Approximately 71% of the adolescents came from The internal consistency in this sample is good (α = 0.91). families with middle socio-economic level (11.6% low; 17.7% Achenbach’s System for Empirically Based Assessment (ASEBA) high) and approximately 87% were Caucasian (White-European), is a common dimensional and empirically derived assessment 9% Arabic, 2% Asian, and 2% Latino. of psychopathology and functioning that has good psychometric properties (Achenbach, 2021). The Spanish adaptations of the Instruments ASEBA show good internal consistency [α ranges from 0.78 to The Social Anxiety Scale for Adolescents (SAS-A) (La Greca and 0.97 for the Child Behavior Checklist (CBCL), for which parents Lopez, 1998) is a measure of 22 items–18 items which refer to are respondents] and adequate test-retest reliability (ICC from social anxiety and four filler items. Questions include items such 0.85 to 0.90) (Achenbach and Rescorla, 2001). The CBCL/6-18 as “I feel shy around people I don’t know” and “I’m quiet when outlines competence in three areas—activities, social and school– I’m with a group of people.” Youths self-report how much each along with a total competence score which comprises a sum of the questionnaire item is characteristic of themselves on a 5-point three former scores (Achenbach, 2018). The social competence scale. There are three subscales, which are all structured such scale, scored by parents (n = 254), is used in the present study as that a higher score indicates greater social anxiety. These three an indicator of social functioning. subscales are summed to comprise a total score. The Spanish Sociometric Index (SI) is a brief measure designed to evaluate adaptation (Olivares et al., 2005) of the SAS-A shows adequate sociometric status in the adolescent population (Ballespí, 2013). internal consistency (Cronbach’s α between 0.76 and 0.91), good It consists of four items scored on a Likert scale from 1 to 9, test-retest reliability (r ranging from 0.75 to 0.86) over a 10 which yields a total score between 4 and 36. This study utilizes day period, and evidence for convergent validity. Cronbach’s the responses of both parents and teachers combined as a multi- alpha in the current sample shows excellent internal consistency informant measure. Respondents were prompted regarding (α = 0.90). adolescents’ number of friends, acceptance by peers, leadership, The Trait Meta-Mood Scale (TMMS) (Salovey et al., 1995) is and popularity. The SI has evidence of convergent validity with a short self-report measure that is designed to assess individual’s related measures, with correlations ranging between 0.2 and beliefs about their identification, understanding, and regulation 0.5. Parent and teacher versions both have good to excellent of emotions. This self-mentalizing measure consists of 24 items internal consistency (α = 0.87 and 0.90, respectively). Principal which evaluates three aspects of meta-cognition—attention (I pay component analysis was utilized to create this multi-informant a lot of attention to my feelings), clarity (I can sometimes say measure. The standardized factor scores of the first component which emotions I am experiencing), and beliefs about regulation were used as a sociometric measure. Factor loadings ranged Frontiers in Psychology | www.frontiersin.org 4 November 2021 | Volume 12 | Article 721584 Ballespí et al. Mentalizing Moderates Social Anxiety Impairment between 0.6 and 0.9, while the factor explained 55% of variability. Internal consistency of the current sample was good for parents and excellent for teachers (α = 0.83, α = 0.94, respectively). Rosenberg’s Self-Esteem Scale (RSES) (Rosenberg, 1965) is a widely used measure to assess self-esteem that consists of 10 items ranked 1–5 in accordance with the degree of agreement with each statement. Items include statements such as “I certainly feel useless at times” and “I am able to do things as well as most other people.” The Spanish adaptation of the RSES has adequate psychometric properties (Martín-Albo et al., 2007). Excellent FIGURE 1 | Depiction of analysis using Hayes’ (2017) PROCESS model 2. internal consistency (α = 0.90) exists in the current sample. Beck’s Depression Inventory (BDI-2) (Beck et al., 1996) contains 21 items for self-evaluation, with three symptom choices that of impairment, conditioned to different values of attention to reflect the respondent’s experience over the course of 7 days. The emotions and emotional clarity. This allowed us to probe how Spanish adaptation (Sanz et al., 2003) has good psychometric the association between social anxiety and impairment indicators properties (e.g., Cronbach’s α = 0.87). Reliability in the current changed at different levels of attention to emotions (low, average, sample was excellent (α = 0.90). high) combined with different levels of emotional clarity (low, average, high), and provides information about the combined Procedure influence of both moderators without the complications of a After obtaining ethical approval in accordance with the 3-way interaction. All models tested met the assumptions of Declaration of Helsinki and evaluation by the Ethics Committee normality, independent errors, homoscedasticity, and absence at the Universitat Autònoma de Barcelona (CEEAH 2603, of multicollinearity. Results are presented as linear regression Spain), participants provided written informed consent for coefficients (b), reporting 95% confidence intervals (95% CI), and a broader project entitled “Personality, psychopathology, and P-values (p). Statistical significance threshold was set at p = 0.05. coping strategies in adolescence.” A letter distributed by the school was utilized for the purposes of informing families about objectives, relevance, and implications of the research. Next, data RESULTS were recruited within the school setting. Adolescents, parents and their teachers received sealed envelopes with the questionnaires Descriptive statistics, correlations and sex differences of all inside with an alphanumeric code that was utilized for identity variables involved are detailed in Table 1. All significant encryption. Teachers were asked to complete all questionnaires correlations were in the expected direction. Age was correlated for their students who agreed to participate in the research. with all mentalizing dimensions and with the indicators of self- Once the deadline for returning questionnaire forms had passed, function impairment (self-esteem and depression). There were families were contacted in the case that there were missing or out- sex differences in all variables but two: social competence and of-range values present in their responses. The AMIs took place sociometric status. in private rooms at the schools. Data collection took place over Models with both self-mentalizing moderators (i.e., attention the course of approximately 5 weeks in each of five schools. and clarity) and the other-mentalizing moderator were first tested for the four response variables–social competence, sociometric Statistical Analysis status, self-esteem, and depression. Other-mentalizing showed Sample size was calculated using G∗ Power 3.1.9 (Faul et al., no statistically significant moderator effect on social competence 2007). For a small size effect (f 2 = 0.05), α = 0.05, power (1- (b = 0.002; p = 0.245; 95% CI: −0.002 to 0.007), sociometric status β) = 0.8, three exposure variables and two control variables, the (b = −0.018; p = 0.055; 95% CI: −0.036 to 0.001), self-esteem sample size required was 223. Linear regressions were conducted (b = −0.001; p = 0.848; 95% CI: −0.008 to 0.007) nor depression using IBM SPSS Statistics v25.0 to test the moderation effects (b = 0.005; p = 0.397; 95% CI: −0.007 to 0.018) response variables, of self-mentalizing and mentalizing regarding others on the so the moderation and the conditional effect of other-mentalizing relationship between social anxiety and both social functioning were removed from all models. and self-functioning variables. Age and sex have been shown Therefore, results are primarily devoted to the moderator to introduce differences in the variables involved; mentalization effects of self-mentalizing variables on the relationship between and functioning have been described by sex and age across this social anxiety and indicators of impairment. These results are developmental stage (e.g., Asher et al., 2017), and thus age and summarized in Tables 2, 3 (conditional effects on social- and self- sex were controlled for in all the analyses. function, respectively), which show the effect sizes (b) of social Moderation analyses were conducted using PROCESS version anxiety (conditional to moderators being set at their mean values) 3.5, model 2 (see Figure 1; Hayes, 2017). The combined influence along with those of self-mentalizing moderators (attention to of both moderators was tested by probing the moderation at low, emotions and emotional clarity). These results are graphically average and high values of both moderating variables, determined depicted in Figures 2–5, which also display the expected values by 16th, 50th, and 84th percentile according to Hayes (2017). This of the response variables (social competence, sociometric status, showed the effects of social anxiety on each one of the indicators self-esteem, and depression) for low (16th percentile) average Frontiers in Psychology | www.frontiersin.org 5 November 2021 | Volume 12 | Article 721584 Ballespí et al. Mentalizing Moderates Social Anxiety Impairment TABLE 1 | Descriptive statistics, correlations, and sex effects. Descriptives Correlations Sex comparisons M (SD) Male M (SD) Female M (SD) T (p) 1 2 3 4 5 6 7 8 1-Social anx 45,05 (13.03) 43.75 (12.83) 46.30 (13.13) −2.11 (0.035) 1 2-Other M 14.45 (4.86) 13.42 (4.84) 15.35 (4.70) −3.24 (0.001) −0.086 1 3- Self M-Att 23 (7.05) 21.54 (6.71) 24.22 (7.12) −3.12 (0.002) 0.220** 0.173** 1 4- Self M-Cla 24.81 (7.38) 27.50 (7.04) 22.58 (6.92) 5.70 (< 0.000) −0.279** −0.014 0.213** 1 5- Social com. 7.66 (2.24) 7.78 (2.15) 7.56 (2.31) 0.78 (0.438) −0.082 0.122 0.108 0.059 1 6- Sociometric 47.35 (10.53) 47.26 (10.09) 47.43 (10.94) −0.13 (0.900) −0.247** 0.192** 0.122 0.109 0.481** 1 7- Self-esteem 21.19 (5.47) 23.03 (4.92) 19.66 (5.44) 5.24 (< 0.000) −0.565** 0.035 −0.203** 0.428** 0.129* 0.229** 1 8- Depression 9.09 (8.28) 7.03 (7.39) 10.81 (8.62) −3.84 (< 0.000) 0.396** 0.063 0.266** −0.351** −0.023 −0.108 −0.675** 1 9- Age 14.64 (1.71) 14.40 (1.66) 14.84 (1.74) −2.08 (0.039) 0.016 0.267** 0.206** −0.142* −0.026 −0.040 −0.256** 0.281** M, Mean; SD, Standard deviation; T, Student-Fisher’s t-test; p, p-value or significance degree; 1-Social anxiety; 2-Other Mentalizing; 3-Self Mentalizing-Attention; 4-Self Mentalizing Clarity; 5-Social competence; 6-Sociometric status; 7-Self-esteem (Rosenberg’s scale); 8-Depression (BDI-II). Pearson’s correlations are significant at the *0.05 level or at the **0.01 level (2-tailed). Bold indicates that the value met statistical significance. TABLE 2 | Social anxiety (conditional) effects and self-mentalizing moderation effects on social function. Social functioning Social competence (n = 254) Sociometric status (n = 223) b (p) 95% CI R2 b (p) 95% CI R2 Social anxiety −0.019 (0.097) −0.042 to 0.004 −0.231 (<0.000) −0.335 to -0.127 M: SA × Attention −0.001 (0.326) −0.004 to 0.001 0.004 −0.009 (0.145) −0.023 to 0.003 0.009 M: SA × Clarity 0.003 (0.043) 0.0001–0.005 0.016 0.016 (0.008) 0.004–0.027 0.028 Results are adjusted for age and sex in all cases. R2 , R2 change; M, Moderation; SA, Social Anxiety; Social competence model R2 = 0.046; Sociometric status model R2 = 0.146. Bold indicates that the value met statistical significance. TABLE 3 | Social anxiety (conditional) effects and self-mentalizing moderation effects on self-function. Self-function Self-esteem (n = 262) Depression (n = 262) b (p) 95% CI R2 b (p) 95% CI R2 Social anxiety −0.174 (<0.000) −0.214 to−0.134 0.146 (<0.000) 0.079–0.214 M: SA × Attention −0.007 (0.008) −0.012 to−0.002 0.015 0.009 (0.047) 0.0001–0.018 0.011 M: SA × Clarity 0.002 (0.32) −0.002 to 0.007 0.002 −0.006 (0.085) −0.014 to 0.001 0.007 Results are adjusted for age and sex in all cases. R2 , R2 change; M, Moderation; SA, Social Anxiety; Self-esteem model R2 = 0.471; Depression model R2 = 0.333. Bold indicates that the value met statistical significance. (50th percentile), and high (84th percentile) values of social significant for sociometric status) as well as self-esteem, while anxiety, attention to emotions and emotional clarity. The values there was a positive association between social anxiety and of these three percentiles allowed us to explore the moderation depression. While all the interaction terms certainly had low across the range of measurement without overcrowding the values of R2 , Tables 2, 3 demonstrate that the association between graphics. In line with prevalence rates and previous research, social anxiety and functioning was moderated by facets of self- the results were controlled for sex and age. The combination awareness. Regarding social functioning (Table 2), the negative of values of attention and clarity where the association between association between social anxiety and both social competence social anxiety and the response variable is statically significant are and sociometric status was moderated by emotional clarity, such highlighted in the Figures 2–5. that the higher emotional clarity, the more dampened association Overall, Tables 2, 3 show a negative association between is between social anxiety and social impairment according to both social anxiety and both indicators of social functioning (i.e., social functioning variables. As such, emotional clarity attenuated social competence and sociometric status, though it was only the association between social anxiety symptoms and outcome Frontiers in Psychology | www.frontiersin.org 6 November 2021 | Volume 12 | Article 721584 Ballespí et al. Mentalizing Moderates Social Anxiety Impairment FIGURE 2 | Moderator effects of self mentalizing dimensions on social competence. Conditional effects of social anxiety on social competence at different levels of attention to emotions and emotional clarity. measures. Regarding self-functioning (Table 3), the relationships was a negative imbalance. Detailed results for each outcome between social anxiety and both self-esteem and depression were are provided below. moderated by attention to emotions, such that more attention to emotions strengthened the associations. Therefore, attention to Social Functioning: Social Competence emotion increased the impairment according to both indicators As Table 2 shows, social anxiety was associated with all response of self-function (i.e., the more self-esteem decreases, and the variables except social competence, but the fact that emotional more depression increases). clarity still had a significant moderating role (b = 0.003; More detailed results show the conditional effects of social p = 0.043; 95% CI: 0.0001–0.005) implies that social anxiety has anxiety on the different outcomes at different levels of the an influential effect on social competence for certain values of two self-mentalizing variables (Table 4). This information is emotional clarity and attention to emotions. Moreover, when summarized graphically in Figures 2–5. Overall, this table shows emotional clarity was low and attention to emotions was high or that attention to emotions tended to exacerbate impairments average, there was a negative association between social anxiety while emotional clarity attenuated them. When evaluating and social competence (b = −0.051, p = 0.01; b = −0.038, the combination of values of both moderating factors (i.e., p = 0.014, respectively) (Table 4 and Figure 2). This indicates that how the association between social anxiety and impairment with low emotional clarity, social competence becomes impaired changed at different combinations of low/average/high values unless attention to emotions is also low. Further, when attention of attention and clarity), impairments were lower when clarity to emotions was high and emotional clarity is average (i.e., not was higher and emotional attention was lower. Accordingly, high), social competence was also impaired (b = −0.33, p = 0.043, the lowest impairment appeared when attention was low respectively) (Table 4, see also Figure 2 where highlighted lines and clarity was high–that is, when there was a positive indicate the significant moderating effects). imbalance between these dimensions–while the highest impairment in all cases (i.e., the most intense association Social Functioning: Sociometric Status between social anxiety and each indicator) appeared when Social anxiety was negatively associated with sociometric status attention was high and clarity was low, that is, when there (b = −0.231; p < 0.000; 95% CI: -0.335 to -0.127) and this Frontiers in Psychology | www.frontiersin.org 7 November 2021 | Volume 12 | Article 721584 Ballespí et al. Mentalizing Moderates Social Anxiety Impairment FIGURE 3 | Moderator effects of self mentalizing dimensions on sociometric status. Conditional effects of social anxiety on sociometric status at different levels of attention to emotions and emotional clarity. association was moderated by emotional clarity (b = 0.016; depression increased as social anxiety does) (b = 0.146; p < 0.000; p = 0.008; 95% CI: 0.004–0.027) (Table 2). The positive value 95% CI: 0.079–0.214), and this was strengthened by attention to of the moderator indicates that the association between social emotions (b = 0.009; p = 0.047; 95% CI: 0.0001–0.018) (Table 4 anxiety and sociometric status became less negative as emotional for values, graphically depicted in Figure 5). clarity increased; in other words, the slopes that represent the relationship between social anxiety and sociometric status were flattened as clarity improved. This can be clearly seen in DISCUSSION Figure 3, which shows the effect of each moderator at different values of the other moderator. The effect of social anxiety on The aim of this study was to analyze whether the self- and other- sociometric status is significant in all possible combinations, but dimensions of mentalizing moderate the association between was attenuated with increased emotional clarity and decreased social anxiety and different indicators of impairment and emotional attention (see the specific values in Table 4). well-being. According to previous research, it was predicted that other-mentalizing would be more implicated in moderating the Self-Functioning: Self-Esteem association between social anxiety and social functioning, while Table 3 shows that social anxiety was negatively associated with self-mentalizing would be more strongly involved in moderating self-esteem (b = −0.174; p < 0.000; 95% CI: −0.214 to −0.134), self-functioning impairments. Curiously, no moderation but in this case the moderation effect came from attention to was found for other-mentalizing, though self-mentalizing emotions (b = −0.007; p = 0.008; 95% CI: −0,012 to −0.002). subdomains moderated the association between social anxiety The negative value of the moderation effect entails that the more and the indicators of social functioning (social competence attention was paid to emotions, the lower expected self-esteem as and sociometric status), and self-functioning (self-esteem social anxiety increased (Figure 4). and depression). Self-Functioning: Depression Other-Mentalizing: Lack of Moderation Finally, there was a statistically significant association between The present findings are intriguing for two reasons. First, the social anxiety and depression, but in this case a positive one (i.e., psychopathological core of social anxiety is the fear of scrutiny Frontiers in Psychology | www.frontiersin.org 8 November 2021 | Volume 12 | Article 721584 Ballespí et al. Mentalizing Moderates Social Anxiety Impairment FIGURE 4 | Moderator effects of self mentalizing dimensions on self-esteem. Conditional effects of social anxiety on self-esteem at different levels of attention to emotions and emotional clarity. and negative evaluation from others (American Psychiatric social functioning is in fact moderated by self-mentalizing, Association, 2013), which is considered the foundation of social precisely because it is directly associated with emotional avoidance and has been conceptualized as hypermentalizing– regulation (Fonagy et al., 2005). In other words, once social essentially excessive Theory of Mind (Sharp and Vanwoerden, anxiety is present, it is less debilitating if those who experience 2015). In social anxiety, this would present as an over- social anxiety are aware of (and therefore more able to tendency to assume that others’ intentions are toward negative cope with) their experience, compared to those with less evaluation of them. Because this is a clear other-mentalizing awareness and regulation, who may become inundated by error, it was reasonable to expect that other-mentalizing would their incomprehensible feelings. From this point of view, the moderate social impairment. Second, previous evidence suggests finding that self- but not other-mentalizing moderates all a differential association between self- vs. other-mentalizing, evaluated functional consequences of social anxiety supports and self- vs. other-function (Ballespí et al., 2021). As such, a well-established evidence that social anxiety is an internalizing possible explanation to our unexpected result is that other- (self) problem (American Psychiatric Association, 2013), and mentalizing may intervene in how social anxiety is developed, is consistent with previous evidence that supports insight as an as an endophenotypical mechanism involved in its appearance active ingredient promoting mental health (Stefan and Cheie, (Tibi-Elhanany, 2011), but not necessarily in moderating the 2020). This suggests that the extent to which people with social consequences of social anxiety once present. anxiety are aware of and understand their socially anxious experiences is involved in their impairment, for both self- and social-function. Self-Mentalization: Moderator of Social Functioning While it is logical that self-mentalizing moderates self- Subdimensions of Self-Mentalizing: functioning, the finding that self-mentalizing moderated Attention vs. Clarity impairment in social functioning requires some reflection. Closer analysis of how self-mentalizing moderates this Viewing the development of social anxiety chronologically, impairment showed differences regarding the two subdimensions it is possible that hypermentalizing leads to social anxiety, of self-mentalizing, attention and clarity. While our hypothesis but the mechanism through which social anxiety impairs that the association between social anxiety and impairment Frontiers in Psychology | www.frontiersin.org 9 November 2021 | Volume 12 | Article 721584 Ballespí et al. Mentalizing Moderates Social Anxiety Impairment FIGURE 5 | Moderator effects of self mentalizing dimensions on depression. Conditional effects of social anxiety on depression at different levels of attention to emotions and emotional clarity. would be more strongly moderated by clarity than attention In fact, emotional dysregulation in social anxiety involves was supported for social functioning, it was in fact attention to attentional biases to the physiological signs of anxiety, which the emotions that moderated the effect of social anxiety symptoms individual expects and fears are perceived and negatively judged on self-functioning. by others. Results suggest that clarity about this process buffers Interestingly, clarity does not simply moderate “more” than the impairment on social functioning. Nonetheless, for the same attention, but was the only significant moderator of social reason one could question why attention to emotions does not functioning, while the opposite finding was found for self- exacerbate the impairment on social functioning as it seems to functioning, whereby attention moderated but clarity did not. with indicators of self-function. This demonstrates that two different dimensions (attention Literature suggests that excessive attention to one’s own and clarity) provide opposite moderation (strengthening and emotional reactions, particularly if this attention is not followed buffering, respectively) regarding two different domains of by emotional clarity, tends to exacerbate rather than buffer this functioning (self- vs. social- functioning). reaction (Gross, 2002; Gross and John, 2003). This assertion is The incomplete accordance with our hypothesis led us to consistent with classic etiopathogenic models of social anxiety wonder why emotional clarity decreases social functioning (Wong et al., 2014), where attention to–and therefore excessive impairment, while this does not occur regarding self-function. awareness of–the physiological reaction of anxiety is expected The fact that self-mentalization buffers negative effects on social to aggravate emotional dysregulation. However, once more, this functioning is consistent with the view of mentalizing as a places the role of attention in the development of the anxiety resilience factor (Stein, 2006; Fonagy and Campbell, 2017). More reaction, but not moderating the association between already specifically, this is aligned with evidence that supports emotional present social anxiety and its functional consequences. This awareness and insight as an adaptive coping mechanism for explanation, however, is logical for social functioning where emotional distress (Troy and Mauss, 2011; Subic-Wrana et al., results show that attention does not moderate the association, 2014), an ability consistent with–or possibly necessary for–good but not for the indicators of self-function, where attention (not social functioning (Sendzik et al., 2017). clarity) is involved. Frontiers in Psychology | www.frontiersin.org 10 November 2021 | Volume 12 | Article 721584 Ballespí et al. Mentalizing Moderates Social Anxiety Impairment TABLE 4 | Conditional effects of SA at different levels of attention to emotions and robbing individuals of the protective effects of social interaction emotional clarity. (Aderka et al., 2012; Aune et al., 2021). Since attention to Emotional Attention to b P 95% CI one’s own anxiety reaction strengthens the process that causes clarity emotions lower–upper said social avoidance (Jakymin and Harris, 2012), and clarity Social Low 0.012 0.529 −0.026 to 0.050 is obstructed under excessive emotional arousal (Luyten et al., competence 2012), this could justify the idea that impairment in self-function, High Average 0.004 0.819 −0.027 to 0.034 which is associated with consistently high rates of social anxiety, High −0.009 0.610 −0.045 to 0.027 could be worsened by the same excessive attention to emotions Low −0.011 0.474 −0.042 to 0.020 that contribute to increasing social anxiety and impeding the Average Average −0.020 0.09 −0.043 to 0.003 protective effect of clarity in the first place. Yet another possible High −0.033* 0.043 −0.064 to −0.001 explanation is that attention and clarity should not be analyzed Low −0.029 0.105 −0.065 to 0.006 separately because they are interdependent dimensions of the Low Average −0.038* 0.014 −0.068 to −0.008 same process. Therefore, how they interact with each other High −0.051* 0.01 −0.089 to −0.012 and moderate the relationship, could shed light on this result. Given that a 3-way interaction would not be interpretable Sociometric Low −0.024 0.798 −0.210 to 0.161 status because attention and clarity moderate in different direction, the High Average −0.093 0.207 −0.237 to 0.051 influence of their imbalances was tested by probing moderation High −0.182* 0.038 −0.353 to −0.010 at low, average and high levels of both moderating variables, Low −0.166* 0.033 −0.317 to −0.014 according to Hayes (2017). Average Average −0.234* <0.000 −0.340 to −0.129 High 0.323* <0.000 −0.468 to −0.178 Low −0.278* <0.002 −0.450 to −0.106 Impairment Varies According to Low Average −0.347* <0.000 −0.484 to −0.210 Imbalances Between Attention and High −0.436* <0.000 −0.609 to −0.262 Clarity Self esteem Low −0.106* 0.002 −0.174 to −0.038 When evaluating moderation at different levels of both High Average −0.152* <0.000 −0.206 to −0.098 moderators, or put simply, analyzing the conditional effect of High −0.212* <0.000 −0.277 to −0.148 social anxiety on impairment at different values of attention Low −0.126* <0.000 −0.182 to −0.071 combined with different values of clarity, the analysis reveals a Average Average −0.172* <0.000 −0.213 to −0.132 High −0.232* <0.000 −0.289 to −0.176 very consistent result within the current study and with those Low −0.142* <0.000 −0.206 to −0.078 previously reported in the literature: the higher one’s emotional Low Average −0.188* <0.000 −0.241 to −0.134 clarity and the lower one’s emotional attention is, the better the High −0.248* <0.000 −0.316 to −0.180 outcome. In fact, this is a combination of the two factors found Depression Low 0.031 0.592 −0.083 to 0.146 to foster mental health when analyzed separately: high clarity High Average 0.089 0.056 −0.002 to 0.180 and low attention. High 0.165* 0.003 0.056–0.274 In the case of social functioning, where clarity was a clear Low 0.089 0.063 −0.005 to 0.183 moderator that buffered impairment experienced with social Average Average 0.147* <0.000 0.078 to 0.215 anxiety symptoms, the values of the effect of social anxiety High 0.222* <0.000 0.127 to0.318 on the indicators (social competence and sociometric status) Low 0.134* 0.016 0.025 to0.242 diminish in a near-linear fashion (Table 4) as clarity increases Low Average 0.192* <0.000 0.101 to0.282 and attention decreases. Beyond the exception commented above, High 0.267* <0.000 0.152 to0.382 this result is incredibly consistent. The extreme polarities between Low, 16th percentile; Average, 50th percentile; High, 84th percentile according to attention and clarity show extreme changes in the effect of social Hayes (2017). *Denotes statistically significant effects (p < 0.05). anxiety regarding all outcomes. Accordingly, high attention and low clarity constitute the worst combination in all cases, and therefore the most impairment (i.e., the highest strengthening Self-Functioning: Why Doesn’t Clarity of the association between social anxiety and impairment on Moderate? all four outcome variables), while the opposite, low attention While the expected outcome was revealed for attention on self- and high clarity, shows the highest buffering effect. This finding function in the present research (i.e., it worsens it), the fact that extends previous evidence that high attention with low clarity emotional clarity did not attenuate the association between social is associated with mental health detriments (Gross, 2002; Gross anxiety and self-functioning is an intriguing result. One possible and John, 2003; Boden and Thompson, 2017). The second explanation is that self-functioning impairment (i.e., in this finding, however, is not as well-supported by previous research, case, decreasing self-esteem and increasing depression) appears and introduces an important research question: is the positive as social anxiety tends to increase, signifying frequent and imbalance of self-mentalizing subdimensions (low attention and intense fear that causes individuals to avoid social interaction, high clarity) better than balanced high self-awareness (high thereby depriving them of the support and rewards that social attention and clarity) in terms of mental health? According to relationships offer. This, in turn, would exacerbate symptoms by evidence supporting insight as a factor promoting mental health Frontiers in Psychology | www.frontiersin.org 11 November 2021 | Volume 12 | Article 721584 Ballespí et al. Mentalizing Moderates Social Anxiety Impairment (David, 2004; Jennissen et al., 2018), it seems that high level of light of the low values of R2 of the interaction. Particularly due to the both attention and clarity should provide better emotional its cross-sectional study design, causation cannot be ascertained, awareness and yield the most protective effect. However, in terms and thus the directionality and timeframe of when mentalization of what recent literature points to regarding attention and clarity, becomes preventative in the presence and development of social where clarity consistently appears as the active ingredient and the anxiety remains to be clear. Finally, although well-established negative imbalance (high attention and low clarity) as the most psychometric measures are acceptably used, especially to assess impairing combination (Boden and Thompson, 2017), it seems large samples, the lack of a better measure of self-mentalizing that the opposite, that is, low level of attention (which seems to be than the self-report used in this study, is also a limitation. Given harmful) and high level of clarity (which seems to be beneficial) the importance of self-mentalizing as a potential general resilient should reasonably be most protective for mental health outcomes. factor, innovation in the assessment of this higher order cognition Moreover, the debate as to whether explicit (high attention is a hot topic and deserves attention. and high clarity) or implicit mentalizing (low attention but high clarity) is more advantageous for mental health naturally presents itself given this result. The current results suggest that CONCLUSION AND CLINICAL implicit mentalizing is clearly better in terms of moderating IMPLICATIONS the impairment of symptoms on functioning, as clarity with low attention does not demonstrate significantly worse effects of Overall, the current study demonstrated that self- but not social anxiety on the four indicators of impairment, while clarity other- mentalizing moderates the association between social with high attention does (Table 4 and Figures 2–5). This is an anxiety and different measures of impairment. This highlights interesting and novel finding. the significance of mentalizing self-processes when treating social On one hand, explicit mentalizing is encouraged by anxiety symptoms in the non-clinical range of the spectrum, mentalization-based treatments (MBT) (Bateman and Fonagy, though this likely also extends to the clinical population. Thus, 2004) to repair mentalizing errors that appear when emotional while classic multidimensional treatments include elements such arousal switches off explicit mentalizing and pre- mentalizing social skill training or exposure to the social world, focused modes appear (Fonagy and Allison, 2014; Luyten and Fonagy, on interaction, the current results stress the extent to which 2015). Pre-mentalizing modes are automatic and therefore comprehension of one’s own emotional experience is crucial implicit forms of “failed mentalizing.” This suggests that explicit to buffer the social impairments associated with social anxiety. or “full mentalizing” could be more advantageous. These implications, which currently refer to the sub- and the non- However, despite the importance of explicit mentalizing, clinical range of the social anxiety spectrum, could possibly be automatic processes also denote proficiency or expertise, extended to the clinical range given the dimensional perspective precisely because automatization reduces resource load and and continuity between clinical and non-clinical social anxiety allows one to utilize them for other cognitive processes (Van that the notion of a continuum introduces (e.g., Schneier et al., Merrienboer and Sweller, 2005). In light of current results, 2002; Van Os et al., 2009; Barrantes-Vidal et al., 2015; Thapar it is possible that this “low-flying” mentalizing, which seems and Riglin, 2020). Further, these results showed that attention is to occur with low attention, is a more sophisticated form of particularly harmful when it is not paired with emotional clarity, mentalization which yields emotional clarity while releasing which leads us to suggest that future research evaluate how the attentional resources to attend to processes outside of the self, subdimensions of self-mentalizing interact with each other in such as those that take place in the social world. Future studies the complex interplay between symptoms, mentalizing, function, should further examine these combinations to better understand and resilience. Given this result, therapies and strategies that which is best for mental health resilience. encourage emotional self-consciousness (e.g., mindfulness-based interventions, acceptance and commitment therapy, MBT) may Strengths and Limitations be particularly helpful to reduce the impairment in social anxiety, A primary benefit of the present research is its novelty; to and possibly further still in other psychopathologic spectra. our knowledge, no research before has evaluated the moderator role of self- and other- mentalizing polarities, let alone on the impairment associated with social anxiety. This study is DATA AVAILABILITY STATEMENT also the first to analyze how the association between social anxiety and impairment changes in accordance with the balance The datasets presented in this study can be found in online or imbalance of two self-mentalizing dimensions, attention to repositories. The names of the repository/repositories emotions and emotional clarity. Further, in the attempt to and accession number(s) can be found in the article/ capture and understand mental health issues earlier in their Supplementary Material. developmental course, this research focused on a non-clinical, adolescent sample. The spectrum-based perspective that this adopts allows researchers to better understand the mechanisms ETHICS STATEMENT involved in mental health development from a dimensional perspective. Nevertheless, as a novel result, the present research The studies involving human participants were reviewed and should be replicated before conclusions are drawn, especially in approved by the Ethics Committee at the Universitat Autònoma Frontiers in Psychology | www.frontiersin.org 12 November 2021 | Volume 12 | Article 721584 Ballespí et al. Mentalizing Moderates Social Anxiety Impairment de Barcelona (CEEAH 2603). Written informed consent to Avançats) from the Government of Catalonia. The funders had participate in this study was provided by the participants’ legal no contributing role in study design, data collection or analysis, guardian/next of kin. publication decision, or manuscript preparation. AUTHOR CONTRIBUTIONS ACKNOWLEDGMENTS SB conceptualized and conducted research and wrote the first We thank to all families who contributed to the study, as well draft. SB and JV validated the data, contributed to data curation as the schools Escola Pia Sarrià-Calassanç, Institut Lluís Vives, and visualization, defined the methodology, and performed Institut Besós, Institut Corbera, and Institut Angeleta Ferrer i formal analysis. SB, JN, and JV wrote the manuscript. SB, NB-V, Serrat from Barcelona, as well as Institut Francesc Ribalta and and JV acquired funding. All authors provided resources and IES-Escola Arrels from the center of Catalonia, for their generous critically reviewed the manuscript. collaboration. We gratitude to Judit López-Berga, Ángel Lorite, Ruth Ocón, and Cristian Lago for their contribution to data collection. We also thank to Marina Cotanda for help in editing FUNDING and visualization of graphics. This work was supported by the Catalan Government (grant no. 2017SGR1612, Suport als Grups de Recerca), by the Spanish SUPPLEMENTARY MATERIAL Ministry of Economy and Competitiveness [grant nos. PSI2014- 56303-REDT (PROMOSAM), PSI2018-100675-B-100, PSI2017- The Supplementary Material for this article can be found 88416-R, and PSI2017-87512-C2-00], and the ICREA Academia online at: https://www.frontiersin.org/articles/10.3389/fpsyg. Research Award (Institució Catalana de Recerca i Estudis 2021.721584/full#supplementary-material REFERENCES Ballespí, S., Vives, J., Alonso, N., Sharp, C., Ramírez, M. S., Fonagy, P., et al. (2019). To know or not to know? Mentalization as protection from Achenbach, T. M. (2018). ASEBA CBCL/6-18 Competence Scale Scores. Burlington, somatic complaints. PLoS One 14:e0215308. doi: 10.1371/journal.pone.021 VT: University of Vermont. 5308 Achenbach, T. M. (2021). Achenbach System of Empirically Based Assessment Ballespí, S., Vives, J., Sharp, C., Chanes, L., and Barrantes-Vidal, N. (ASEBA). Available online at: https://aseba.org/ (accessed March 30, 2019). (2021). Self and other mentalizing polarities and dimensions of Achenbach, T. M., and Rescorla, L. A. (2001). Inventario Del Comportamiento mental health: association with types of symptoms, functioning and De Niños(as) De 6–18 Años para Padres (CBCL/6-18) y Para Maestros (TRF/6- well-being. Front. Psychol. 12:566254. doi: 10.3389/fpsyg.2021.56 18) [Child Behavior Checklist 6-18, Parents’ form (CBCL/6-18) and -Teachers’ 6254 Report form (TRF)] (Unitat D’epidemiologia i De Diagnòstic En Psicopatologi. Balluerka, N., Aritzeta, A., Gorostiaga, A., Gartzia, L., and Soroa, G. (2013). Burlington, VT: University of Vermont. Emotional intelligence and depressed mood in adolescence: a multilevel Aderka, I. M., Hofmann, S. G., Nickerson, A., Hermesh, H., Gilboa-Schechtman, approach. Int. J. Clin. Health Psychol. 13, 110–117. doi: 10.1016/S1697- E., and Marom, S. (2012). Functional impairment in social anxiety disorder. 2600(13)70014-0 J. Anxiety Disord. 26, 393–400. doi: 10.1016/j.janxdis.2012.01.003 Banerjee, R., and Henderson, L. (2001). Social-cognitive factors in childhood social Alvi, T., Kouros, C. D., Lee, J., Fulford, D., and Tabak, B. A. (2020). Social anxiety anxiety: a preliminary investigation. Soc. Dev. 10, 558–572. doi: 10.1111/1467- is negatively associated with theory of mind and empathic accuracy. J. Abnorm. 9507.00180 Psychol. 129, 108–113. doi: 10.1037/abn0000493 Barrantes-Vidal, N., Grant, P., and Kwapil, T. R. (2015). The role of schizotypy in American Psychiatric Association (2013). DSM-5 Diagnostic Classification. the study of the etiology of schizophrenia spectrum disorders. Schizophr. Bull. Diagnostic and Statistical Manual of Mental Disorders. Washington, DC: 41(Suppl. 2), S408–S416. doi: 10.1093/schbul/sbu191 American Psychiatric Association. doi: 10.1176/appi.books.9780890425596. Bartholomew, K., and Horowitz, L. M. (1991). Attachment styles among young x00DiagnosticClassification adults: a test of a four-category model. J. Pers. Soc. Psychol. 61, 226–244. doi: American Psychological Association (2020). Social Cognition. Dictionary of 10.1037/0022-3514.61.2.226 Psychology. Washington, DC: American Psychological Association. Batanova, M. D., and Loukas, A. (2011). Social anxiety and aggression in Asher, M., Asnaani, A., and Aderka, I. M. (2017). Gender differences in social early adolescents: examining the moderating roles of empathic concern and anxiety disorder: a review. Clin. Psychol. Rev. 56, 1–12. doi: 10.1016/j.cpr.2017. perspective taking. J. Youth Adolesc. 40, 1534–1543. doi: 10.1007/s10964-011- 05.004 9634-x Aune, T., Juul, E. M. L., Beidel, D. C., Nordahl, H. M., and Dvorak, R. D. (2021). Bateman, A. W., and Fonagy, P. (2004). Mentalization-based treatment of BPD. Mitigating adolescent social anxiety symptoms: the effects of social support J. Pers. Disord. 18, 36–51. doi: 10.1521/pedi.18.1.36.32772 and social self-efficacy in findings from the Young-HUNT 3 study. Eur. Child Battaglia, M., Pesenti-Gritti, P., Medland, S. E., Ogliari, A., Tambs, K., and Spatola, Adolesc. Psychiatry 30, 441–449. doi: 10.1007/s00787-020-01529-0 C. A. M. (2009). A genetically informed study of the association between Aune, T., and Stiles, T. C. (2009). The effects of depression and stressful life childhood separation anxiety, sensitivity to CO(2), panic disorder, and the effect events on the development and maintenance of syndromal social anxiety: sex of childhood parental loss. Arch. Gen. Psychiatry 66, 64–71. doi: 10.1001/ and age differences. J. Clin. Child Adolesc. Psychol. 38, 501–512. doi: 10.1080/ archgenpsychiatry.2008.513 15374410902976304 Beck, A. T., Steer, R. A., and Brown, G. K. (1996). Beck Depression Inventory– Ballespí, S. (2013). Brief Sociometric Status Index for Adolescents. Barcelona: Second Edition (BDI-II). San Antonio, TX: The Psychological Corporation. Universitat Autonoma de Barcelona. doi: 10.1037/t00742-000 Ballespí, S., and Pérez-Domingo, A. (2015). Reliability and validity of a brief Boden, M. T., and Thompson, R. J. (2017). Meta-analysis of the association between interview for assessing reflective function in the adolescence: the adolescent emotional clarity and attention to emotions. Emot. Rev. 9, 79–85. doi: 10.1177/ mentalization interview (AMI). Eur. Child Adolesc. Psychiatry 24, S213–S213. 1754073915610640 Frontiers in Psychology | www.frontiersin.org 13 November 2021 | Volume 12 | Article 721584 Ballespí et al. Mentalizing Moderates Social Anxiety Impairment Brady, E. U., and Kendall, P. C. (1992). Comorbidity of anxiety and depression in Frith, C. D. (1999). Interacting Minds–A Biological Basis. Science 286, 1692–1695. children and adolescents. Psychol. Bull. 111, 244–255. doi: 10.1037/0033-2909. doi: 10.1126/science.286.5445.1692 111.2.244 Frith, U., and Frith, C. (2003). Development and neurophysiology of mentalizing. Broeren, S., Muris, P., Diamantopoulou, S., and Baker, J. R. (2013). The course Philos. Trans. R. Soc. Lond. B Biol. Sci. 358, 459–473. doi: 10.1098/rstb.2002. of childhood anxiety symptoms: developmental trajectories and child-related 1218 factors in normal children. J. Abnorm. Child Psychol. 41, 81–95. doi: 10.1007/ Gohm, C. L. (2003). Mood regulation and emotional intelligence: individual s10802-012-9669-9 differences. J. Pers. Soc. Psychol. 84, 594–607. doi: 10.1037/0022-3514.84.3.594 Brown, T. A., Campbell, L. A., Lehman, C. L., Grisham, J. R., and Mancill, R. B. Gohm, C. L., Corser, G. C., and Dalsky, D. J. (2005). Emotional intelligence under (2001). Current and lifetime comorbidity of the DSM-IV anxiety and mood stress: Useful, unnecessary, or irrelevant? Pers. Individ. Dif. 39, 1017–1028. disorders in a large clinical sample. J. Abnorm. Psychol. 110, 585–599. doi: doi: 10.1016/j.paid.2005.03.018 10.1037/0021-843X.110.4.585 Greeson, J., Garland, E. L., and Black, D. (2014). “Mindfulness: a transtherapeutic Clark, D. M., and Mcmanus, F. (2002). Information processing in social Phobia. approach for transdiagnostic mental processes,” in The Wiley Blackwell Biol. Psychiatry 51, 92–100. doi: 10.1016/S0006-3223(01)01296-3 Handbook of Mindfulness, eds A. le, C. T. Ngnoumen, and E. J. Langer Colonnesi, C., Nikolić, M., de Vente, W., and Bögels, S. M. (2017). Social anxiety (Hoboken, NJ: Wiley Blackwell), 533–562. doi: 10.1002/9781118294895.ch28 symptoms in young children: investigating the interplay of theory of mind and Gross, J. J. (2002). Emotion regulation: affective, cognitive, and social expressions of shyness. J. Abnorm. Child Psychol. 45, 997–1011. doi: 10.1007/ consequences. Psychophysiology 39, 281–291. doi: 10.1017/S0048577201393198 s10802-016-0206-0 Gross, J. J., and John, O. P. (2003). Individual differences in two emotion regulation David, A. S. (2004). “The clinical importance of insight: an overview,” in Insight processes: implications for affect, relationships, and well-being. J. Pers. Soc. and Psychosis: Awareness of Illness in Schizophrenia and Related Disorders, 2nd Psychol. 85, 348–362. doi: 10.1037/0022-3514.85.2.348 Edn, eds X. F. Amador and A. S. David (Oxford: Oxford University Press), Hayes, A. F. (2017). “Introduction to mediation, moderation, and conditional doi: 10.1093/med/9780198525684.001.0001 process analysis: a regression-based approach,” in Methodology in the Social Davis, S. K., and Nichols, R. (2016). Does emotional intelligence have a “dark” side? Sciences Series, ed. T. D. Little (New York, NY: Guilford publications). A review of the literature. Front. Psychol. 7:1316. doi: 10.3389/fpsyg.2016.01316 Heimberg, R. G., Brozovich, F. A., and Rapee, R. M. (2014). “A cognitive-behavioral De la Barrera, U., Villanueva, L., Montoya-Castilla, I., and Prado-Gascó, V. (2021). model of social anxiety disorder,” in Social Anxiety, eds S. G. Hofmann and P. M. How much emotional attention is appropriate? The influence of emotional DiBartolo (Amsterdam: Elsevier), 705–728. doi: 10.1016/B978-0-12-394427-6. intelligence and subjective well-being on adolescents’ stress. Curr. Psychol. 1–13. 00024-8 Denny, B. T., Kober, H., Wager, T. D., and Ochsner, K. N. (2012). A meta-analysis Hill, C. L. M., and Updegraff, J. A. (2012). Mindfulness and its relationship to of functional neuroimaging studies of self-and other judgments reveals a spatial emotional regulation. Emotion 12, 81–90. doi: 10.1037/a0026355 gradient for mentalizing in medial prefrontal cortex. J. Cogn. Neurosci. 24, Iancu, I., Bodner, E., and Ben-Zion, I. Z. (2015). Self esteem, dependency, self- 1742–1752. doi: 10.1162/jocn_a_00233 efficacy and self-criticism in social anxiety disorder. Compr. Psychiatry 58, Eckland, N. S., and Berenbaum, H. (2021). Emotional awareness in daily life: 165–171. doi: 10.1016/j.comppsych.2014.11.018 exploring its potential role in repetitive thinking and healthy coping. Behav. Jakymin, A. K., and Harris, L. M. (2012). Self-focused attention and social anxiety. Ther. 52, 338–349. doi: 10.1016/j.beth.2020.04.010 Aust. J. Psychol. 64, 61–67. doi: 10.1111/j.1742-9536.2011.00027.x Extremera, N., and Fernández-Berrocal, P. (2006). Emotional intelligence as Jennissen, S., Huber, J., Ehrenthal, J. C., Schauenburg, H., and Dinger, U. (2018). predictor of mental, social, and physical health in university students. Span. J. Association between insight and outcome of psychotherapy: systematic review Psychol. 9, 45–51. doi: 10.1017/S1138741600005965 and meta-analysis. Am. J. Psychiatry 175, 961–969. doi: 10.1176/appi.ajp.2018. Faravelli, C., Zucchi, T., Viviani, B., Salmoria, R., Perone, A., Paionni, A., et al. 17080847 (2000). Epidemiology of social phobia: a clinical approach. Eur. Psychiatry 15, Kerns, J. G., and Berenbaum, H. (2010). Affective processing in overwhelmed 17–24. doi: 10.1016/S0924-9338(00)00215-7 individuals: strategic and task considerations. Cogn. Emot. 24, 638–660. doi: Farmer, A. S., and Kashdan, T. B. (2014). Affective and self-esteem instability in the 10.1080/02699930902927664 daily lives of people with generalized social anxiety disorder. Clin. Psychol. Sci. La Greca, A., and Lopez, N. (1998). Social anxiety among adolescents: linkages 2, 187–201. doi: 10.1177/2167702613495200 with peer relations and friendships. J. Abnorm. Child Psychol. 26, 83–94. doi: Faul, F., Erdfelder, E., Lang, A.-G., and Buchner, A. (2007). G∗ Power 3: a flexible 10.1023/A:1022684520514 statistical power analysis program for the social, behavioral, and biomedical LaBounty, J., Bosse, L., Savicki, S., King, J., and Eisenstat, S. (2017). Relationship sciences. Behav. Res. Methods 39, 175–191. doi: 10.3758/BF03193146 between social cognition and temperament in preschool-aged children. Infant Fehm, L., Beesdo, K., Jacobi, F., and Fiedler, A. (2008). Social anxiety disorder above Child Dev. 26:e1981. doi: 10.1002/icd.1981 and below the diagnostic threshold: prevalence, comorbidity and impairment Lewinsohn, P. M., Zinbarg, R., Seeley, J. R., Lewinsohn, M., and Sack, W. H. (1997). in the general population. Soc. Psychiatry Psychiatr. Epidemiol. 43, 257–265. Lifetime comorbidity among anxiety disorders and between anxiety disorders doi: 10.1007/s00127-007-0299-4 and other mental disorders in adolescents. J. Anxiety Disord. 11, 377–394. Fernandez-Berrocal, P., Extremera, N., and Ramos, N. (2004). Validity and doi: 10.1016/S0887-6185(97)00017-0 reliability of the Spanish modified version of the Trait Meta-Mood Scale. Luyten, P., Campbell, C., Allison, E., and Fonagy, P. (2020). The mentalizing Psychol. Rep. 94, 751–755. doi: 10.2466/pr0.94.3.751-755 approach to psychopathology: state of the art and future directions. Annu. Rev. Fonagy, P., and Allison, E. (2014). The role of mentalizing and epistemic trust in Clin. Psychol. 16, 297–625. doi: 10.1146/annurev-clinpsy-071919-015355 the therapeutic relationship. Psychotherapy 51, 372–380. doi: 10.1037/a0036505 Luyten, P., and Fonagy, P. (2015). The Neurobiology of Mentalizing. Pers. Disord. Fonagy, P., and Campbell, C. (2017). Mentalizing, attachment and epistemic 6, 366–379. doi: 10.1037/per0000117 trust: how psychotherapy can promote resilience. Psychiatr. Hung. 32, Luyten, P., Fonagy, P., Lowyck, B., and Vermote, R. (2012). “PPT - The assessment 283–287. of mentalization,” in Handbook of Mentalizing in Mental Health Practice, Fonagy, P., Gergely, G., Jurist, E., and Target, M. (2005). Affect Regulation, eds A. W. Bateman and P. Fonagy (Washington, DC: American Psychiatric Mentalization, and the Development of the Self. New York, NY: Other Press, Publishing, Inc). LLC. Martín-Albo, J., Núñez, J. L., Navarro, J. G., and Grijalvo, F. (2007). The Rosenberg Fonagy, P., and Luyten, P. (2009). A developmental, mentalization-based approach self-esteem scale: Translation and validation in university students. Span. J. to the understanding and treatment of borderline personality disorder. Dev. Psychol. 10, 458–467. doi: 10.1017/S1138741600006727 Psychopathol. 21, 1355–1381. doi: 10.1017/S0954579409990198 Mayer, J. D., Caruso, D. R., and Salovey, P. (2016). The ability model of emotional Fonagy, P., and Target, M. (2002). Early intervention and the development of intelligence: principles and updates. Emot. Rev. 8, 290–300. doi: 10.1177/ self-regulation. Psychoanal. Inq. 22, 307–335. doi: 10.1080/07351692209348990 1754073916639667 Fonagy, P., Target, M., Steele, H., and Steele, M. (1998). Reflective-Functioning Mayer, J. D., and Gaschke, Y. N. (1988). The experience and meta-experience Manual, Version 5.0, for Application to Adult Attachment Interviews. London: of mood. J. Pers. Soc. Psychol. 55, 102–111. doi: 10.1037/0022-3514.55. University College London. doi: 10.1037/t03490-000 1.102 Frontiers in Psychology | www.frontiersin.org 14 November 2021 | Volume 12 | Article 721584 Ballespí et al. Mentalizing Moderates Social Anxiety Impairment Mayo-Wilson, E., Dias, S., Mavranezouli, I., Kew, K., Clark, D. M., Ades, A. E., Stein, H. (2006). “Does mentalizing promote resilience?” in Handbook of et al. (2014). Psychological and pharmacological interventions for social anxiety Mentalization-Based Treatment, eds J. G. Allen and P. Fonagy (Chichester: John disorder in adults: a systematic review and network meta-analysis. Lancet Wiley & Sons, Ltd), 307–326. doi: 10.1002/9780470712986.ch16 Psychiatry 1, 368–376. doi: 10.1016/S2215-0366(14)70329-3 Stein, M. B., and Stein, D. J. (2008). Social anxiety disorder. Lancet 371, 1115–1125. Olivares, J., Ruiz, J., Hidalgo, M. D., Garcia-lopez, L. J., Rosa, A. I., and Piqueras, doi: 10.1016/S0140-6736(08)60488-2 J. A. (2005). Social Anxiety Scale for Adolescents (SAS-A): psychometric Subic-Wrana, C., Beutel, M. E., Brähler, E., Stöbel-Richter, Y., Knebel, A., Lane, properties in a Spanish-speaking population. Int. J. Clin. Health Psychol. 5, R. D., et al. (2014). How is emotional awareness related to emotion regulation 85–97. strategies and self-reported negative affect in the general population? PLoS One Pearcey, S., Gordon, K., Chakrabarti, B., Dodd, H., Halldorsson, B., and Creswell, 9:e91846. doi: 10.1371/journal.pone.0091846 C. (2020). Research Review: the relationship between social anxiety and social Thapar, A., and Riglin, L. (2020). The importance of a developmental perspective cognition in children and adolescents: a systematic review and meta-analysis. in Psychiatry: What do recent genetic-epidemiological findings show? Mol. J. Child Psychol. Psychiatry 62, 805–821. doi: 10.1111/jcpp.13310 Psychiatry 25, 1631–1639. doi: 10.1038/s41380-020-0648-1 Pile, V., Haller, S. P. W., Hiu, C. F., and Lau, J. Y. F. (2017). Young people Thompson, R. J., Dizén, M., and Berenbaum, H. (2009). The unique relations with higher social anxiety are less likely to adopt the perspective of another: between emotional awareness and facets of affective instability. J. Res. Pers. 43, Data from the director task. J. Behav. Ther. Exp. Psychiatry 55, 41–48. doi: 875–879. doi: 10.1016/j.jrp.2009.07.006 10.1016/j.jbtep.2016.11.002 Thompson, R. J., Mata, J., Jaeggi, S. M., Buschkuehl, M., Jonides, J., and Gotlib, Plana, I., Lavoie, M.-A., Battaglia, M., and Achim, A. M. (2014). A meta- I. H. (2013). The role of attention to emotion in recovery from major depressive analysis and scoping review of social cognition performance in social phobia, disorder. Depress. Res. Treat. 2013:540726. doi: 10.1155/2013/540726 posttraumatic stress disorder and other anxiety disorders. J. Anxiety Disord. 28, Tibi-Elhanany, Y. (2011). Social cognition in social anxiety: first evidence for 169–177. doi: 10.1016/j.janxdis.2013.09.005 increased empathic abilities. Isr. J. Psychiatry Relat. Sci. 48, 98–106. Resurrección, D. M., Salguero, J. M., and Ruiz-Aranda, D. (2014). Emotional Troy, A. S., and Mauss, I. B. (2011). “Resilience in the face of stress: intelligence and psychological maladjustment in adolescence: a systematic emotion regulation as a protective factor,” in Resilience and Mental Health: review. J. Adolesc. 37, 461–472. doi: 10.1016/j.adolescence.2014.03.012 Challenges Across the Lifespan, eds S. Southwick, B. Litz, D. Charney, and Rosenberg, M. (1965). Society and the Adolescent Self-Image. Princeton, NJ: M. Friedman (Cambridge: Cambridge University Press), 30–44. doi: 10.1017/ Princeton University Press. doi: 10.1515/9781400876136 CBO9780511994791.004 Russell, G., and Shaw, S. (2009). A study to investigate the prevalence of social Van Merrienboer, J. J. G., and Sweller, J. (2005). Cognitive load theory and complex anxiety in a sample of higher education students in the United Kingdom. learning: recent developments and future directions. Educ. Psychol. Rev. 17, J. Ment. Health 18, 198–206. doi: 10.1080/09638230802522494 147–177. doi: 10.1007/s10648-005-3951-0 Salguero, J. M., Palomera, R., and Fernández-Berrocal, P. (2012). Perceived Van Os, J., Linscott, R. J., Myin-Germeys, I., Delespaul, P., and emotional intelligence as predictor of psychological adjustment in adolescents: a Krabbendam, L. (2009). A systematic review and meta-analysis of the 1-year prospective study. Eur. J. Psychol. Educ. 27, 21–34. doi: 10.1007/s10212- psychosis continuum: evidence for a psychosis proneness–persistence– 011-0063-8 impairment model of psychotic disorder. Psychol. Med. 39, 179–195. Salovey, P., Mayer, J. D., Goldman, S. L., Turvey, C., and Palfai, T. P. (1995). doi: 10.1017/S0033291708003814 “Emotional attention, clarity, and repair: exploring emotional intelligence using Vine, V., and Aldao, A. (2014). Impaired emotional clarity and psychopathology: the Trait Meta-Mood Scale,” in Emotion, Disclosure, & Health, ed. J. W. a transdiagnostic deficit with symptom-specific pathways through emotion Pennebaker (Washington, DC: American Psychiatric Publishing, Inc), 125–154. regulation. J. Soc. Clin. Psychol. 33, 319–342. doi: 10.1521/jscp.2014.33. doi: 10.1037/10182-006 4.319 Sanz, J., Perdigón, A. L., and Vázquez, C. (2003). Adaptación española del Wong, J., Gordon, E. A., and Heimberg, R. G. (2014). “Cognitive-behavioral Inventario para la Depresión de Beck-II (BDI-II): 2. Propiedades psicométricas models of social anxiety disorder,” in The Wiley Blackwell Handbook of Social en población general. Clin. Salud 14, 249–280. Anxiety Disorder, ed. J. W. Weeks (Hoboken, NJ: Wiley), 1–23. doi: 10.1002/ Schneier, F. R., Blanco, C., Antia, S. X., and Liebowitz, M. R. (2002). The social 9781118653920.ch1 anxiety spectrum. Psychiatr. Clin. North Am. 25, 757–774. doi: 10.1016/S0193- 953X(02)00018-7 Conflict of Interest: The authors declare that the research was conducted in the Sendzik, L., Schäfer, J. Ö., Samson, A. C., Naumann, E., and Tuschen-Caffier, B. absence of any commercial or financial relationships that could be construed as a (2017). Emotional awareness in depressive and anxiety symptoms in youth: a potential conflict of interest. meta-analytic review. J. Youth Adolesc. 46, 687–700. doi: 10.1007/s10964-017- 0629-0 Publisher’s Note: All claims expressed in this article are solely those of the authors Sharp, C., and Vanwoerden, S. (2015). Hypermentalizing in borderline personality and do not necessarily represent those of their affiliated organizations, or those of disorder: a model and data. J. Infant Child Adolesc. Psychother. 14, 33–45. the publisher, the editors and the reviewers. Any product that may be evaluated in doi: 10.1080/15289168.2015.1004890 this article, or claim that may be made by its manufacturer, is not guaranteed or Sripada, C. S., Angstadt, M., Banks, S., Nathan, P. J., Liberzon, I., and Phan, endorsed by the publisher. K. L. (2009). Functional neuroimaging of mentalizing during the trust game in social anxiety disorder. Neuroreport 20, 984–989. doi: 10.1097/WNR. Copyright © 2021 Ballespí, Vives, Nonweiler, Perez-Domingo and Barrantes-Vidal. 0b013e32832d0a67 This is an open-access article distributed under the terms of the Creative Commons Stefan, C. A., and Cheie, L. (2020). Self-compassion and social anxiety in late Attribution License (CC BY). The use, distribution or reproduction in other forums adolescence: contributions of self-reflection and insight. Self Identity. 1–13. is permitted, provided the original author(s) and the copyright owner(s) are credited doi: 10.1080/15298868.2020.1861082 and that the original publication in this journal is cited, in accordance with accepted Stein, D. J., Ono, Y., Tajima, O., and Muller, J. E. (2004). The social anxiety academic practice. No use, distribution or reproduction is permitted which does not disorders spectrum. J. Clin. Psychiatry 65(Suppl. 14), 27–33. comply with these terms. Frontiers in Psychology | www.frontiersin.org 15 November 2021 | Volume 12 | Article 721584