ORIGINAL RESEARCH published: 09 July 2019 doi: 10.3389/fpsyg.2019.01501 Hypermentalizing in Social Anxiety: Evidence for a Context-Dependent Relationship Sergi Ballespí 1* , Jaume Vives 2 , Carla Sharp 3 , Andrea Tobar 1 and Neus Barrantes-Vidal 1,4,5 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 Psychology, University of Houston, Houston, TX, United States, 4 Department of Mental Health, Fundació Sanitària Sant Pere Claver, Barcelona, Spain, 5 Centre for Biomedical Research Network on Mental Health (CIBERSAM), Instituto de Salud Carlos III, Madrid, Spain Social anxiety (SA) means fear of scrutiny and of others’ negative evaluation, thus indicating that hypermentalizing (HMZ) (i.e., the over-attribution of intentions and thoughts to others) might be the most common error of social cognition in SA. However, evidence for this is weak. One explanation is that HMZ is not stable in SA, but rather context-dependent. The first aim of the current study was testing this hypothesis. The second aim was analyzing whether the association between SA and HMZ is moderated Edited by: Mattie Tops, by a negative self-image. One-hundred and thirteen young adults (85.8% females; Vrije Universiteit Amsterdam, M = 21.1 years old; SD = 2.7) were assessed on measures of SA, HMZ, and self-image. Netherlands Given the over-representation of females, conclusions may not be safely extrapolated Reviewed by: Sebastian Ocklenburg, to males. Results revealed that HMZ is associated with SA only in the self-referential Ruhr University Bochum, Germany social situation [B = 2.68 (95% CI: 0.72–4.65), p = 0.007]. This supports that HMZ is Alexander Strobel, not global in SA (i.e., a stable cognitive style), but rather is active only in some contexts. Dresden University of Technology, Germany Implications for the conceptualization and treatment of SA are discussed. Contrary to *Correspondence: predictions, neither self-esteem, nor positive or negative self-schema moderated the Sergi Ballespí association between SA and self-referential HMZ. This contradicts findings in the field of sergi.ballespi@uab.cat paranoid delusion and requires replication, including measures of implicit self-esteem. Specialty section: Keywords: social anxiety, social cognition, mentalization, hypermentalizing, context-dependency, self-image, This article was submitted to self-esteem Cognition, a section of the journal Frontiers in Psychology INTRODUCTION Received: 22 March 2019 Accepted: 13 June 2019 Social Anxiety Disorder (SAD) is characterized by marked fear of scrutiny and negative social Published: 09 July 2019 evaluation (American Psychiatric Association [APA], 2013). This disorder represents the clinical Citation: extreme of the Social Anxiety (SA) Spectrum (Schneier et al., 2002), affects 2 to 7% of adults Ballespí S, Vives J, Sharp C, annually, and is frequently associated with depression (Beesdo et al., 2007). Despite the highly Tobar A and Barrantes-Vidal N (2019) Hypermentalizing in Social Anxiety: impairing and chronic nature of SAD, only half of people with the disorder ever seek treatment Evidence for a Context-Dependent Relationship. Abbreviations: BPD, borderline personality disorder; HMZ, hypermentalizing; MASC, movie for the assessment of social Front. Psychol. 10:1501. cognition; MIMS, method for inducing mentalization in a self-referential situation; MZ, mentalizing; SA, social anxiety; SAD, doi: 10.3389/fpsyg.2019.01501 social anxiety disorder; ToM, theory of mind. Frontiers in Psychology | www.frontiersin.org 1 July 2019 | Volume 10 | Article 1501 Ballespí et al. Hypermentalizing in Social Anxiety Is Context-Dependent and most of them do so after 15–20 years of experiencing However, despite the social dimension of SA, fewer studies symptoms (American Psychiatric Association [APA], 2013). analyzed social cognition. To analyze SA from social cognition The challenge in recovery from SAD highlights the implies an additional step as it moves the interpretation bias from importance of preventing clinical levels of SA, and emphasizes the scenario (i.e., the interpretation of the situation as a potential the need to identify some of the underlying mechanisms along threat, or the excessive attention to negative consequences), to the the SA Spectrum (Schneier et al., 2002). Given the acknowledged mental states (i.e., the interpretation or attribution of intentions, importance of cognitive aspects in SA, the role of attention and feelings and judgments to others). memory biases has been widely analyzed (Heeren et al., 2015). Interestingly, among the very few studies that analyzed social However, little is known about the role of social cognition, cognition errors in individuals with SA, only two of them especially regarding aspects such as social perception and found an association between HMZ and SA. Hezel and McNally mentalizing (Plana et al., 2014). (2014) referenced greater prevalence of “excessive ToM” in Mentalizing (MZ) is defined as the higher-order cognition people with SAD compared with non-SAD individuals, albeit that allows the identification of mental states (intentions, demonstrating a small effect size. Washburn et al. (2016) desires, feelings) that underpin human behavior (Fonagy and reported higher rates of “excessive ToM” in SAD groups Bateman, 2016). MZ is a multidimensional process (Choi- (i.e., SAD and SAD+Depression) compared to people with Kain and Gunderson, 2008; Kim, 2015), usually referred to major depression, but not compared to controls. Thus, against as social cognition or Theory of Mind (ToM), the latter theoretical predictions, evidence is limited and implies a modest specifically referring to the capacity to infer or imagine what association between HMZ and SA. A possible explanation is that is happening in another’s mind. This process is hypothesized HMZ is not a general and stable attribute of SA. to play a role in SA Spectrum because at the core of SA is In disorders like BPD, HMZ is the underlying social-cognitive the fear of negative evaluation, meaning that people with SA problem (Sharp et al., 2013), and is attributed to absence of proper may tend to avoid social interaction because of the frightening attachment relationships in childhood (Fonagy and Bateman, mental states that they attribute to other people (i.e., the 2008; Fonagy and Luyten, 2009; Fonagy et al., 2011), which are intention of scrutiny and negative evaluation). Put differently, the necessary for the correct development of MZ capacity (Fonagy interpretations of what others think or intend are the core of fear and Target, 1997). However, there is no reason to suppose that in people with SA. MZ problems in SA are similarly rooted in disrupted attachment While some classic ToM tasks have been employed in children relationships in childhood. with SA (Banerjee, 2008), a relatively new approach to studying Etiological models of SA, as in Rapee and Spence (2004), MZ involves the assessment of excessive ToM (Dziobek et al., have established that SA is heavily influenced by genetic risk 2006), or hypermentalizing (HMZ) (Sharp et al., 2011). HMZ factors (with heritability estimates around 0.65), which can be refers to the over-attribution of mental states to others (such as expressed as a highly reactive or an inhibited temperament. Such the intention of negative evaluation), that goes far beyond the temperaments make children more susceptible to environmental relevant context. The concept of HMZ was originally associated influences – such as modeling by anxious parents (American with paranoid ideation (Langdon and Coltheart, 1999), and Psychiatric Association [APA], 2013). Thus, although early social has been found in Borderline Personality Disorder (BPD; environment plays a role in moderating genetic dispositions Sharp et al., 2013). (Nachmias et al., 1996), attachment-related trauma is not The essential component of SA (American Psychiatric usually underlying the development of SA as it is in BPD. Association [APA], 2013), suits the definition of HMZ because Therefore, it is possible that most people with SA had good the fear to scrutiny is excessive since it is unlikely or paranoid enough attachment relationships and developed normal or (without basis in context), and the individual is convinced even improved MZ capacity (Mink et al., 2014), at least in his/her attribution of negative evaluation rather merely in non-distressing social situations (e.g., those with close considering it possible. This makes people with SA exhibit friends or family). MZ reasoning more similar to those in the paranoid spectrum In contrast, since arousal affects MZ capacity, as noted by (over-attributing intentions to others), than those in the autism Fonagy and Bateman (2016), Frith and Frith (2003), and Luyten spectrum (Baron-Cohen et al., 1985). “Excessive theory of mind” and Fonagy (2015), it is possible that those situations that become or HMZ should be the most common error of MZ in individuals distressing to people with SA – i.e. uncertain social situations in with SA. This is consistent with the interpretation bias found which they feel involved – elicit MZ errors because they activate by most studies about cognition (Clark and McManus, 2002; self-referential MZ. This may partly explain why prior studies Miers et al., 2008). were unable to show that general HMZ accounted for most of Cognition is an important element in SA, and distortions the MZ errors associated with SA. in attention, interpretation, and memory biases have been The aim of the current study was to evaluate this hypothesis. thoroughly analyzed (Heinrichs and Hofmann, 2001). The It is assumed that if HMZ was generalized in people with SA, it interpretation bias is one of the most extensively analyzed should appear in all MZ activity, even in non-frightening social from the information processing paradigm and refers to the situations. By contrast, if HMZ in SA is context-dependent, then tendency to misinterpret the social scenario by exaggerating cost it should be especially present in frightening social situations, estimations of negative outcome, or the negative consequences of operationalized here as uncertain situations that involve the the social situation (Chen et al., 2018). individual and therefore foster self-referential MZ. Frontiers in Psychology | www.frontiersin.org 2 July 2019 | Volume 10 | Article 1501 Ballespí et al. Hypermentalizing in Social Anxiety Is Context-Dependent Thus, we hypothesize that non-self-referential HMZ – may moderate the association between SA and self-referential measured with a computerized procedure in which participants HMZ, because a negative idea of one’s self might lead people to mentalize about the characters of a film – will be less strongly fear that others judge them as they do; that is, to fear that others associated to SA than self-referential HMZ, as measured through consider (or “discover”), they are as they believe themselves to be a social situation that directly involves the participant. (i.e., according to a poor self-image: “bad” or defective in some To analyze this hypothesis in a sample from general way). This hypothesis assumes that people with SA tend to HMZ population, we adopt the criterion of the Harvard’s group (Kagan because they assume as certain what they fear: that others will et al., 1987) to operationalize as “presence” of SA those cases with consider them in the same way they do. The over-attribution to scores above percentile 85th in measures of SA (Kagan et al., others of one’s own mental state (i.e., negative self-judgment), not 1998). This criterion is psycho-physiologically based (Kagan only suits the definition of HMZ but it also suggests that HMZ et al., 1988) and widely used in literature about SA spectrum (see in SA arises from negative self-image. According to the literature Ballespí et al., 2013 for a review). (Smith et al., 2006; Kesting and Lincoln, 2013) both affective Assuming support for our hypothesis, we additionally aimed and cognitive dimensions of self-image were considered, so the to explore why those with SA tend to HMZ in self-referential moderating role of self-esteem and self-concept was analyzed. social situations. Clark and McManus (2002) suggest that SA is maintained through a vicious cycle in which people engage in a series of biased pre- and post-interaction cognitive processes that MATERIALS AND METHODS act to generate anxious emotion and impair social performance. From this point of view, context-dependent HMZ may appear as Participants an intermediate process between trait-SA (i.e., tendency to SA), Participants included 113 individuals (97 females, or 85.8%; 85% and symptoms (expression of this disposition). However, it is still White-European, 14% Latino, 1% Arabic) with a mean age of unknown what leads people with SA to project the intention of a 21.1 years old (SD = 2.7), who participated in the Phase 2 of a negative judgment onto others – a judgment which leads to fear study about MZ and SA. In the Phase 1 of this study, the students and social avoidance. Similar to those with SA, individuals with of 1st and 2nd year of Psychology Degree were encouraged to persecutory ideation are by definition frequently misreading the report levels of SA through the SPAI-B (Garcia-Lopez et al., 2008; intentions of other people and over attributing mental states to Piqueras et al., 2012; see Section “Instruments”) and the Brief others (i.e., HMZ). Version of the Fear of Negative Evaluation Scale (BFNE) (Leary, In the field of psychosis, self-image has been identified as 1983; see also Gallego et al., 2007). To operationalize “presence” a psychological mechanism associated with persecutory beliefs of SA basing on a spectrum perspective (Schneier et al., 2002), and attributional biases (Valiente et al., 2011, 2014). Bentall the criterion of the Harvard’s group was used (Kagan et al., and colleagues’ “defensive model” (see Thewissen et al., 2008; 1987, 1988). This criterion considers that the extremes of 15% Udachina et al., 2009; Melo and Bentall, 2013) of paranoia in a distribution of SA are qualitatively different than middle- establishes that those with persecutory ideation attribute to others levels positions (Kagan et al., 1998). However, in contrast to the responsibility, actions, and intentions of negative events in classic studies which compare two extreme groups (those with order to protect the self from low self-esteem and negative self- scores above percentile 85th in a SA measure, with those below beliefs. Freeman and Garety’s investigations (Freeman et al., percentile 15th, i.e.: the extremes of 15% of the distribution of 1998; Smith et al., 2006; Freeman, 2007), also show that low that measure), in order to avoid the extremes discontinuity, we mood, low self-esteem, and negative schematic beliefs contribute aimed to compare those with SA (i.e., cases above percentile to the development of psychotic symptoms but, in this case, 85th in measures of SA) with a random selection of those below in a direct and non-defensive way. In contrast to Bentall’s percentile 85% (see Figure 1). suggestion, these authors suggest that low self-esteem in paranoia Thus, from the approximately 750 students invited to study, may not depend on psychotic, but on neurotic, processes 244 provided written informed consent and completed the (Freeman and Garety, 2003). screening measures. From those 244, the 65 participants with Beyond differences, both the aforementioned models support high levels of SA in the screening measures (i.e., scores above the idea that self-concept and self-esteem play an important the 85th percentile in the SPAI-B or in the BFNE), along with role in the development of psychotic delusion (Bentall et al., a randomized sample of 90 of the 177 “controls” (i.e., those 2001; Freeman, 2007). The excessive, “paranoid” fear to others’ with SA scores below the 85th percentile in SPAI-B and BFNE) scrutiny and negative evaluation can be considered a lower-level were invited to participate in Phase 2. From the 155 potential delusion from a dimensional point of view, since the mental participants for Phase 2, 122 agreed to participate and 113, or states attributed to others tend to be held as true instead of only 73%, provided written informed consent for this phase of the possible. Consequently, it is also possible that self-image may play study and came to the lab to complete additional measures. The a role in the development of “neurotic” over-attributions (i.e., 42 individuals who did not participate were those who did not the suspicion of scrutiny and negative evaluation), that people answer the phone calls from the research team (10 participants), with SA do when HMZ. and those who cited not being interested or having logistic Thus, our second aim was to evaluate if self-referential HMZ difficulties (i.e., conflicts with work schedule, distance to the depends on low self-esteem. Specifically, we hypothesized that lab). It is of course possible that SA could be a reason for some poor self-image (i.e., low self-esteem or negative self-schema), participants to avoid this part of the project with “in-person” Frontiers in Psychology | www.frontiersin.org 3 July 2019 | Volume 10 | Article 1501 Ballespí et al. Hypermentalizing in Social Anxiety Is Context-Dependent Recruitment Eligible sample in Phase 1 (n= 750) Declined to participate in Phase 1 (n= 506) Total recruited (n= 244) Failed to fill in one or more scales involving study variables (n= 2) Candidates to Phase 2 (n=242) Cases with scores Cases with scores above Pc 85th in below Pc 85th in SA (n=65) SA (n=177) Excluded for Phase 2 (n=87) All of them invited Random selection to participate in invited to Phase 2 Phase 2 (n=65) (n=90) Declined to participate in Phase 2 Agreed to Agreed to participate participate (n=36) (n=86) Failed to provide all data (n=9) Analyses Provided all data Provided all data (n=33) with high (n=80) with low (n=113) levels of SA levels of SA FIGURE 1 | Sample recruitment flow chart. assessments. The final sample was composed of 80 participants Social phobia and anxiety inventory – brief form (SPAI-B) with low levels of SA, and 33 participants considered to have (Garcia-Lopez et al., 2008; Piqueras et al., 2012) high level SA (i.e., with scores above the 85th percentile on the This is a cost-effective 16-item version of the SPAI (Turner BFNE or the SPAI-B). et al., 1989), particularly useful as a screening measure of the three-response system approach (i.e., motor, psychophysiological Measures and cognitive), for SA, and consistent with the original scale. Social Anxiety There is evidence for construct and concurrent validity both in According to literature, two of the most widely used instruments samples of adolescents and adults, and it possesses good test– for the screening of SA were used in phase 1 to select people with retest reliability (r = 0.60 in 6 months), and excellent internal high (i.e., scores > 85th percentile on at least in one measure), consistency (Cronbach’s α = 0.92). The internal consistency in the and low (i.e., scores < 85th percentile in both measures), for current sample is α = 0.93. phase 2. The SPAI-B (Garcia-Lopez et al., 2008) was selected because it emphasizes the measuring of psychophysiological signs Brief Version of the Fear of Negative Evaluation Scale (Leary, of SA and of social avoidance. The BFNE (Leary, 1983) was 1983) selected to emphasize the cognitive component of SA. Both This scale consists of 12 items scored on a 5-point Likert instruments are described below. scale and it assesses a defining characteristic of SA and SAD: Frontiers in Psychology | www.frontiersin.org 4 July 2019 | Volume 10 | Article 1501 Ballespí et al. Hypermentalizing in Social Anxiety Is Context-Dependent the degree of fear to negative evaluation (American Psychiatric The administration of the MIMS includes three steps (before, Association [APA], 2013). There is evidence for convergent during, and after the interview). and concurrent validity of this scale, and it shows an excellent internal consistency (Cronbach’s α = 0.90), both in the original Before the interview. Participants were informed that after a Spanish version (Gallego et al., 2007; Pitarch, 2010) and in sequence of lab situations, such as Rosenberg’s Self-Esteem Scale the current study. (RSES) (Rosenberg, 1965), a MZ interview, and the MASC, they would finish with an important and definitive interview. Hypermentalizing The interview. The content of the interview consisted of personal In order to test if this MZ error is context-dependent in SA, questions, inspired by the initial questions of the SCID II (First we obtained a measure of HMZ in two different contexts. et al., 1997, 1999; i.e., How would you describe yourself? How do First, we used the movie for the assessment of social cognition you think that others would describe you? What kinds of things (MASC) (Dziobek et al., 2006), described below. To score the have you done that other people found disgusting or annoying? MASC, participants have to think about the mental states of the If you could change your personality, how would you like to be characters of a film. Given that participants are not personally different?), but also including other questions (i.e., What do you involved in the social situation, and they individually score the want to do in life? or If you met a genie who granted you two MASC alone in a lab, HMZ errors will not be influenced by SA in wishes, what would you ask for?). It was expected that personal this case; thus, we consider that the MASC scale of HMZ is a good questions would help to increase arousal especially in people operationalization of non-self-referential HMZ. By contrast, in with SA. Neither the actual questions nor the answers were our second procedure [the method for inducing mentalization important for the purpose of the assessment. The focal point in a self-referential situation (MIMS), see below], we expose was the standardized sequence of “neutral social signs” that the participants to a social situation in which they are involved, so interviewer introduced in the interview in order to activate MZ their MZ activity will be self-referential. Given that, people prone in the participants. For instance, after the instruction (i.e., “Now to SA are expected to experience SA in this situation, because they I am going to ask you some questions about the kind of person that have to talk about themselves with a non-familiar interviewer. you are, that is, how you feel and behave in general”), the first Movie for the assessment of social cognition (Dziobek et al., question is asked: “How would you describe yourself?” At the first 2006) pause in the participant’s response the interviewer clears his/her This is a widely used and well-established method for the throat (item 1). If the participant takes more than 5 s to resume assessment of the ToM reasoning style. Administration of the answering, the interviewer clears his/her throat again. When the MASC takes about 45 min. It requires that participants watch participant finishes, the interviewer again clears his/her throat a 15 min film about four characters getting together for a and simulates writing down the answer. Then the interviewer dinner party. The video is paused 46 times and multiple- asks the second question: How do you think that others would choice questions concerning characters’ feelings, thoughts, and describe you? When participants are carrying on their response intentions are asked. The participants answer the questions that the interviewer yawns (item 2). When participants are asked follow every one of the 46 video segments by choosing one about their defects and virtues, the interviewer hesitates and of the four options of response regarding the mental states of expresses surprise, respectively (items 3 and 4). And so on with the characters. In all cases, four response options are offered: other “neutral social signs” (e.g., to stare, to knit the brow, to adequate MZ, hypomentalizing (under-mentalizing), no MZ, and be silent for a while, to hesitate before writing a response, to HMZ. The MASC provides a score for every one of these four pretend to improvise a question out of the interview), that are scales. The Spanish version of the MASC (Lahera et al., 2014) expected not to be negatively interpreted by controls but indeed shows psychometric properties as good as those of the original by SA participants. version, including good internal consistency (α = 0.86). The computerized administration of the measure (as it only provides After the interview. Immediately after the interview, participants final scores), does not allow estimation of the internal consistency were placed alone in a room with a computer in order to in the current sample. allow them to rate from 1 (“I do not agree at all”), to 7 (“Completely agree”), the different possible attributions of every Method for inducing mentalization in a self-referential “neutral social sign.” For instance, the first question was: “During situation the interview, it is possible that the interviewer coughed or This method was created to assess MZ through a real social self- cleared his/her throat. Why do you think he/she did this?,” and referential situation. It is inspired by the MASC (Dziobek et al., the four responses that the participant had to rate from 1 to 2006; see the description above), in the sense that participants are 7 were: “(a) He/she needed to clear his/her throat” (Correct, asked about different moments of a social interaction. However, because it is likely; refers to a specific attributed need); “(b) in the MIMS the social situation is not fictitious like in the MASC I possibly said something wrong or negative and he/she tried (i.e., a movie watched in a computer), but it is real and involves to cloak his/her reaction” (HMZ, because this attribution goes the participant (i.e., it consists of a standardized interview of the beyond the context); “(c) It is probably a tic; some people have participant). In this context, the mental states about which the them” (under-mentalizing, because it refers to a global tendency participants will be asked are not those of characters, but those of and tends to place the cause out of mind); “(d) It is not the interviewer, a person with whom participants are interacting. worth wondering why people clear their throat” (no MZ, because Frontiers in Psychology | www.frontiersin.org 5 July 2019 | Volume 10 | Article 1501 Ballespí et al. Hypermentalizing in Social Anxiety Is Context-Dependent the response lack any reference to mental states in order to (Support to Research Groups, 2014SGR1070) and by the explain behavior). Spanish Ministry of Economy and Competitiveness, PSI2014- So, this method makes it possible to ask participants about the 56303-REDT and PSI2014-52962-P). The authors declare no mental states beyond the behavior of a person (the interviewer), conflict of interest. with whom they have been interacting in a real (and self- The study consisted on 2 phases. In phase 1, participants rated referential) situation, instead of the characters of a video several on-line questionnaires (including SPAI-B and BFNE) after watched alone in a lab. providing written informed consent. These measures involved Principal Components Analysis revealed that the MIMS suited the SA-screening measures used in this study to design the groups a 4-factor solution according to the scree-plot and criteria of with high and low SA. The design of the groups was based parsimony, interpretability, and explained variance (40%; weights on the Harvard’s group criterion to operationalize “presence” ranged from 0.3 to 0.8). Varimax rotation highlighted a first factor of SA (Kagan et al., 1987, 1988, 1998). Following this criterion, grouping items of HMZ versus other 3 factors grouping items of among the 242 who agreed to participate and provided full data correct-, hypo- and no-MZ. Basing on the 6 items of HMZ that: in Phase 1 (see Figure 1), those above percentile 85th in the (a) were more clearly referred to HMZ according to a peer-review SPAI-B or BFNE were selected as “cases with high SA” (n = 65), process, and (b) unequivocally weighted only in the factor of and a random selection of those below percentile 85th (90 of HMZ (weights ranged 0.55–0.79), a total score of self-referential 177) were selected as cases with “low SA.” In Phase 2, after HMZ was calculated. providing further written informed consent for this second phase, The score of self-referential HMZ was the sum of the 33 participants with high SA and 80 participants with low SA took scores of participants on those items, ranging from 6 to part in different laboratory tasks (necessary for other studies), 42, and showed an acceptable internal consistency according including the measures here used: the MASC, the MIMS, and self- to Cronbach’s α = 0.75. The HMZ scale of the MIMS was image measures Supplementary Material. The participation in positively correlated with theoretically associated constructs, the phase 2 was compensated with 20 euros. such as the scale of Fear to Negative Evaluation of the SASA (r = 0.31, p = 0.001), and with the HMZ scale of Statistical Analysis the MASC (r = 0.25, p = 0.009). This is evidence for T-test and χ2 (Fisher’s exact test was used in case of cells with convergent validity. Moreover, it was negatively correlated expected count less than 5), were performed to test for statistical to Rosenberg’s Scale of Self-Esteem (r = –0.35, p = 0.002) independence between exposure and demographic variables. and showed no association with the other scales of the Linear regressions were performed to investigate the association MASC (correct-, hypo- and no-MZ), which is evidence for between SA and self- and non-self-referential HMZ, as well as discriminant validity. to appraise the moderating role of self-esteem (measured with Rosenberg’s Scale), and self-concept (i.e., positive and negative Self-Image self-schemas measured with BCSS), on the relation between SA According to literature, measures of self-esteem (SE) and and self-referential HMZ. self-concept were used to assess the affective and the cognitive We conducted post hoc power analyses using G∗ Power v.3.1.9 dimensions of self-image (Smith et al., 2006). The RSES (Faul et al., 2007) for a small to medium effect (f 2 = 0.1), (Rosenberg, 1965) is one of the most widely used instruments α = 0.05, and the smallest sample size used in our analyses to assess self-esteem. It consists of 10 items rated through a (n = 109), two explanatory variables, two control variables, the 4-point Likert-type scale and provides a single global score of power (1 – β) was 0.84. SE. The Spanish version shows good psychometric properties All data analyzes were using IBM SPSS Statistics v19.0 package (Martín-Albo et al., 2007) and the internal consistency in (IBM Corp., 2010). Assumptions of independent errors (Durwin– the current sample was excellent (Cronbach’s α = 0.91). Watson test), homoscedasticity (Plot of standardized predicted The Brief Core Schema Scales (BCSS) (Fowler et al., 2006) values against standardized residuals), normality of residuals is a 24-item self-report that provides measures of self and (P–P plot), absence of multicollinearity (VIF and tolerance), others’ positive and negative schemas. The scales of positive and of influential cases (Cook’s distance), were checked. All and negative self-schema were used in the present study as models tested met the assumptions. The results of the association an operationalization of self-concept. These scales showed between anxiety (high/low) and each response variable, as well excellent internal consistency (Cronbach’s α ≥ 0.90) in as the moderation effects are presented as linear regression the current sample. coefficients (B), reporting 95% confidence intervals (95% CI), and p-values. Statistical significance was set at p < 0.05. Procedure The study meets ethical standards according to Declaration of Helsinki and was approved by the Ethics Committee of the RESULTS Universitat Autònoma de Barcelona (CEEAH, Spain; CEEAH no. 4061). We report how we determined and obtained our sample, Demographic Characteristics and all data exclusions, all manipulations, and all measures in the Descriptive Statistics of Study Variables study. Data used in the analyses are provided as Supplementary There were no statistical differences in the distribution of age Material. The study was funded by the Catalan Government [t(111) = 0.62, p = 0.54], sex (Fisher’s exact test, p = 0.39), Frontiers in Psychology | www.frontiersin.org 6 July 2019 | Volume 10 | Article 1501 Ballespí et al. Hypermentalizing in Social Anxiety Is Context-Dependent ethnic group (Fisher’s exact test, p = 0.39), or socioeconomic TABLE 2 | Moderation of self-image on the effect of SA on self-referential HMZ. status (Fisher’s exact test, p = 0.30) between the group with Self-referential HMZ low level of SA (N = 80; 84% of females, 83% Caucasian, 16% Latino, 1% Arabic; aged M = 21.3 years old, SD = 2.9), and B (95% bCI) p the group with high level of SA (N = 33; 91% of females, 91% Moderation of Self-Esteem (n = 112) Caucasian, 9% Latino; aged M = 20.8 years old, SD = 2.1). SA (High vs. Low) 0.99 (−1.70 to 3.63) 0.37 The overall distribution of socioeconomic status was: 8% Low, Self-esteem −0.25 (−0.42 to 0.06) 0.07 33% Medium–Low, 23% Medium, 31% Medium–High, 5% High. SA × Self-esteem −0.16 (−0.58 to 0.29) 0.36 Table 1 presents the descriptive statistics of the study variables for Moderation of Negative Self-Schema (n = 109) each one of the comparison groups. SA (High vs. Low) 1.77 (−0.57 to 4.15) 0.09 Negative Self-Schema −0.04 (−0.08 to 0.16) 0.50 Effects of SA on Self- and SA × Negative Self-Schema −0.17 (−0.17 to 0.51) 0.21 Non-self-referential HMZ Moderation of Positive Self-Schema (n = 109) To analyze the relationship between SA and HMZ, linear SA (High vs. Low) 2.10 (−0.31 to 4.78) 0.45 regression models were tested with the self- and non-self- Positive Self-Schema −0.03 (−0.25 to 0.17) 0.81 referential HMZ as response variables. Final models show that the SA × Positive Self-Schema −0.21 (−0.46 to 0.79) 0.43 association between SA and self-referential HMZ was statistically Social anxiety (SA), Linear regression coefficients (B) of mean-centered predictors, significant [B = 2.68 (95% CI: 0.72–4.65), p = 0.007], but not the 95% bootstrap confidence intervals (95% bCI), and p values (p). association with Non-self-referential HMZ [B = 0.58 (95% CI: – 0.34–1.49), p = 0.22]. This indicates context-dependency in the relationship between SA and HMZ. mental states that go so far beyond observable data that the average observer will struggle to see how they are justified”; Sharp et al., Moderators of the Effect of SA on 2013, p. 4) than to that of people with autism spectrum disorders Self-Referential HMZ (i.e., deficit of ToM, hypo-mentalizing or mind-blindness; Baron- To analyze the moderating role of self-image on the relationship Cohen et al., 1985). However, previous studies did not find a between SA and HMZ, linear regression models were tested with clear relationship between SA and HMZ (Hezel and McNally, self-referential HMZ as the response variable. Table 2 shows 2014; Washburn et al., 2016). This leads us to hypothesize that the final models where no moderation effect was found to be HMZ might not be a “stable” MZ error in SA, but it might be statistically significant. Therefore, the relationship between SA context-dependent. and Self-referential HMZ is not moderated by self-esteem, nor by self-concept (i.e., positive and negative self-schema). The Context-Dependency of Hypermentalizing in Social Anxiety DISCUSSION The current findings point to the presence of HMZ associated with SA in a context-dependent manner. Specifically, it seems The phenomenological definition of SA (American Psychiatric to depend on how arousing the social situation is. Given Association [APA], 2013) highlights the tendency to attribute that the MASC is a procedure that places participants in a intentions of scrutiny and certainty of negative judgments to quiet situation (i.e., they are alone in a lab thinking about others. This is an essential feature of SA. From the perspective of the mental states of the characters of a film, so they are social cognition, this mechanism suits the definition of “excessive not involved in the social situation that they evaluate), it theory of mind” (Dziobek et al., 2006) or HMZ (Sharp et al., is expected that SA will not be present, so it will not 2013), so it was reasonable to expect that people with SA will interfere with MZ activity. By contrast, given that in the exhibit a pattern of social cognition more similar to that of people MIMS participants are directly involved in a self-referential with paranoia (i.e., “to make assumptions about other people’s situation (an interview about themselves with a non-familiar TABLE 1 | Descriptive statistics of variables of the study. Low SA High SA M (95%CI) Min Max M (95%CI) Min Max 1. Non-self-referential HMZ 3.64 (3.13–4.15) 0 9 4,37 (3.49–5.25) 1 11 2. Self-referential HMZ 12 (11.07–12.93) 6 22 14.10 (11.74–16.39) 6 30 3. Self-esteem 23.03 (21.95–24.10) 10 30 18.30 (16.07–20.53) 9 29 4. Negative self-schema 7.83 (5.94–9.72) 0 24 10.3 (7.45–13.15) 0 21 5. Positive self-schema 16.15 (15.25–17.06) 6 24 15.50 (14.08–16.92) 5 22 Social anxiety (SA), 95% confidence intervals (95% CI), mean (M), minimum (Min), maximum (Max). Frontiers in Psychology | www.frontiersin.org 7 July 2019 | Volume 10 | Article 1501 Ballespí et al. Hypermentalizing in Social Anxiety Is Context-Dependent adult), which is considered arousing for socially anxious people, mentalization should be active ingredient in treatment, both SA is expected to be activated and therefore to impair MZ as a method (using adequate MZ to foster recovery), and activity leading to HMZ. The current results support this an objective (to detect, understand and deactivate context- hypothesis. HMZ was associated to SA only when it was dependent HMZ). self-referential, that is, while in the interview, but not while watching a film, as it happens when the MASC is used. This finding is important for several reasons. First, because it Explicit Self-Image Does Not Moderate helps to clarify why previous studies using only the MASC Self-Referential Hypermentalizing (Hezel and McNally, 2014; Washburn et al., 2016) did not find Our second question considered the role of self-image in context- a clear relationship between SA and general HMZ. Second, dependent HMZ. It was hypothesized that a negative self-image if the MZ error most expectably associated with SA is not might be involved in the excessive suspiciousness of people with stable but is context-dependent, then this suggests that people SA, thus leading them to expect the same negative judgment with SA can show adequate MZ capacity, which can be from others that they have toward themselves, and therefore to used in treatment. In light of these results, situational states over-attribute the intention of scrutiny and negative judgment of MZ merit as much attention as the general MZ style. (HMZ). Current results do not support this idea. Nor self- These results suggest that considering both trait- and state- esteem neither self-schemas moderate the association between perspectives might shed new light in the study of MZ and SA and HMZ. This goes against both theoretical framework mental disorders. (Asendorpf et al., 2002; Wilson and Rapee, 2006; Zhao et al., 2012) and clinical experience, suggesting an important role of self-image in SA. One possible explanation is that the role of Implications of Self-Referential self-esteem in SA is different than the relationship analyzed Hypermentalizing for Treatment of Social here. It is possible that self-esteem does not moderate the Anxiety association between SA and HMZ, but it contributes to general Knowing the reluctance of people with SA to seek for help expectances of negative evaluation (i.e., a negative idea of (American Psychiatric Association [APA], 2013), as well as others), or to specific ones (i.e., increasing the fear to possible their expectable difficulties to interact with a strange (the negative evaluation before the interview). Moreover, self-esteem therapist), it is attractive to speculate implications for treatment may interact with specific types or components of SA (e.g., of current results. The situational dependence of MZ errors general fear to negative evaluation), or with other factors highlights the existence of adequate MZ, which can be used (e.g., necessity of recognition or external validation), in the for prevention and treatment. In this sense, the adaptation of prediction of HMZ. Mentalization Based Treatment (MBT) (Bateman and Fonagy, 2008, 2009) to SA could be useful to: (a) focus on this Higher Order Cognition (MZ) involved in SA by definition (American Limitations and Guidelines for Further Psychiatric Association [APA], 2013), (b) specifically increasing Research awareness of the situational tendency to HMZ, and (c) therefore However, it is also possible that implicit but not explicit inoculating against “delusional” beliefs about others’ intentions self-esteem is moderating the association between SA through introducing doubt of those distorted thoughts that have and HMZ. All measures of self-esteem in the current been held to be true (see Sharp et al., 2013). study are based in self-reports. The lack of the implicit Additionally, we should expect not only that HMZ might dimension is a limitation. In the field of psychosis, several intervene in the rejection and delay of treatment of SA, but studies go beyond the analysis of “simple” self-esteem that projections and distortions derived from HMZ should to analyze the discrepancy between implicit and explicit be expected also in the therapeutic relationship. This is an self-esteem and its role in the development of paranoid additional reason to adapt MBT to SA, since MBT allows delusion (Valiente et al., 2011, 2014). Further analyzes may one to mentalize the therapeutic relationship and to address consider whether implicit self-esteem or a combination MZ errors en vivo in an experiential way. Moreover, MZ of implicit-explicit are involved in the activation of can be used even beyond MBT since it is present as a context-dependent HMZ in SA. common active ingredient of treatments known to work (Wells Another limitation of the current study is that the sample, and Purdon, 1999; Ahn and Wampold, 2001; Allen et al., based on volunteers and predominantly including females, is 2008; Wampold, 2015; Hayes and Hofmann, 2017). Beyond not representative of the general population. In fact, given the the specific therapeutic program, MZ activity in treatment over-representation of females (85%), conclusions may not be contributes strengthening of the therapeutic relationship (Allen safely extrapolated to males. Therefore, replications with more and Munich, 2003; Wampold and Imel, 2015), reinforces trust representative samples are required. Future studies should also in treatment and in the clinician (Fonagy and Allison, 2014; analyze the influence of general MZ profile in context-dependent Fonagy et al., 2015), and therefore, fosters adherence that MZ errors, as well as the relationship of self-esteem and MZ further increases probability of success (Sandi-Urena et al., problems with different dimensions of SA (e.g., SA associated 2011; Tschuschke et al., 2015). So, given the social dimension with interaction versus cognitive SA; physiological components of SA, and therefore the importance of social cognition, versus behavioral components). Frontiers in Psychology | www.frontiersin.org 8 July 2019 | Volume 10 | Article 1501 Ballespí et al. Hypermentalizing in Social Anxiety Is Context-Dependent CONCLUSION AUTHOR CONTRIBUTIONS Despite limitations, this study contains an important strength. SB conceptualized and investigated the study and wrote To our knowledge, it is the first to use a measure of self- the manuscript. SB, JV, and AT contributed to data referential MZ activity induced in a live situation, and it is the curation and visualization. JV performed formal analysis. first to show that MZ errors associated with a psychopathological NB-V and SB acquired funding and provided resources. condition can be context-dependent, and not necessarily a stable SB, JV, and CS defined the methodology. CS and NB-V cognitive style. This highlights the importance of analyzing MZ validated the data. CS, NB-V, JV, and SB critically reviewed from both trait- and state-points of view, because beyond the the manuscript. relatively stable social cognitive style, there can also be persistent context-dependent variations of MZ activity. If future studies support current findings, to identify why HMZ is activated in FUNDING specific contexts, and therefore to elucidate which are the specific mechanisms that lead to HMZ might help to improve treatment This work was supported by the National Plan for Research of SA. This study highlights the underestimated role of social and Development of the Spanish Ministry of Economy and cognition in SA, and more specifically the context-dependency Competitiveness (Grant Nos. PSI2014-56303-REDT, PSI2014- of HMZ in SA. Implications for treatment are also important. 52962-P, and PSI2017-87512-C2-00), and by the Agency Since HMZ is present in self-referential social situations and the for Support to the Research Groups (AGAUR) (Grant No. therapeutic context is one of those, HMZ might be present in 2017SGR1612), and the Academia Research Award (ICREA2014) the therapeutic relationship and the clinician might be able to from Catalan Government. observe and directly treat one of the most essential symptoms of the SA spectrum. Future research should analyze whether treating HMZ in the therapeutic relationship improves cost-efficiency of current treatments and long-term benefits for people with SA. ACKNOWLEDGMENTS The authors would like to thank Melania Fernández, Jaume DATA AVAILABILITY Bonet, Maria Capellas, Marina Navarro, Nora Pérez, Àlex Muñoz, Ana Cortijos, and Maria Miralda for their contribution to data All datasets generated for this study are included in the collection. Our gratitude to Kathryn Mettman, Teresa Kearn, and manuscript and/or the Supplementary Files. Naida Alonso for proofreading and formatting the manuscript. Special thanks to Kathryn Mettman for the edition work. ETHICS STATEMENT SUPPLEMENTARY MATERIAL The study meets ethical standards according to Declaration of Helsinki and was approved by the Ethics Committee of the The Supplementary Material for this article can be found Universitat Autònoma de Barcelona (CEEAH, Spain; CEEAH online at: https://www.frontiersin.org/articles/10.3389/fpsyg. no. 4061). We report how we determined and obtained our 2019.01501/full#supplementary-material sample, all data exclusions, all manipulations, and all measures in the study. DATA SHEET S1 | Data used in the analyses. REFERENCES Ballespí, S., Jané, M. C., and Riba, M. D. (2013). Reliability and validity of a brief clinician-report scale for screening behavioral inhibition. J. Psychopathol. Ahn, H. N., and Wampold, B. E. (2001). Where oh where are the specific Behav. Assess. 35, 321–334. doi: 10.1007/s10862-013-9344-7 ingredients? A meta-analysis of component studies in counseling and Banerjee, R. (2008). “Social cognition and anxiety in children,” in Social Cognition psychotherapy. J. Couns. Psychol. 48, 251–257. doi: 10.1037/0022-0167.48.3.251 and Developmental Psychopathology, eds C. Sharp, P. Fonagy, and I. Goodyer Allen, J. G., Bateman, A. W., and Fonagy, P. (2008). Mentalizing (New York, NY: Oxford University Press), 239–269. in Clinical Practice. Washington, DC: American Psychiatric Baron-Cohen, S., Leslie, A. M., and Frith, U. (1985). Does the autistic child have a Publishing. “theory of mind”? Cognition 21, 37–46. doi: 10.1016/0010-0277(85)90022-8 Allen, J. G., and Munich, R. L. (2003). Clinical implications of attachment and Bateman, A., and Fonagy, P. (2008). 8-year follow-up of patients treated mentalization: efforts to preserve the mind in contemporary treatment – Part for borderline personality disorder: mentalization-based treatment versus I – Introduction. Bull. Menninger. Clin. 67, 87–90. doi: 10.1521/bumc.67.2.87. treatment as usual. Am. J. Psychiat. 165, 631–638. doi: 10.1176/appi.ajp.2007. 23441 07040636 American Psychiatric Association [APA] (2013). Diagnostic and Statistical Bateman, A., and Fonagy, P. (2009). Randomized controlled trial of outpatient Manual ofMental Disorders, 5th Edn. Washington DC: American Psychiatric mentalization-based treatment versus structured clinical management for Publishing. borderline personality disorder. Am. J. Psychiat. 166, 1355–1364. doi: 10.1176/ Asendorpf, J. B., Banse, R., and Mucke, D. (2002). Double dissociation between appi.ajp.2009.09040539 implicit and explicit personality self-concept The case of shy behavior. J. Pers. Beesdo, K., Bittner, A., Pine, D. S., Stein, M. B., Höfler, M., Lieb, R., et al. (2007). Soc. Psychol. 83, 380–393. doi: 10.1037//0022-3514.83.2.380 Incidence of social anxiety disorder and the consistent risk for secondary Frontiers in Psychology | www.frontiersin.org 9 July 2019 | Volume 10 | Article 1501 Ballespí et al. Hypermentalizing in Social Anxiety Is Context-Dependent depression in the first three decades of life. Arch. Gen. Psychiat. 64, 903–912. Garcia-Lopez, L. J., Hidalgo, M. D., Beidel, D. C., Olivares, J., and Turner, S. (2008). doi: 10.1001/archpsyc.64.8.903 Brief form of the social phobia and anxiety inventory (SPAI-B) for adolescents. Bentall, R. P., Corcoran, R., Howard, R., Blackwood, N., and Kinderman, P. (2001). Eur. J. Psychol. Assess. 24, 150–156. doi: 10.1027/1015-5759.24.3.150 Persecutory delusions: a review and theoretical integration. Clin. Psychol. Rev. Hayes, S. C., and Hofmann, S. G. (2017). The third wave of cognitive behavioral 21, 1143–1192. doi: 10.1016/S0272-7358(01)00106-4 therapy and the rise of process-based care. World Psychiatry 16, 245–246. doi: Chen, J., Milne, K., Dayman, J., and Kemps, E. (2018). Interpretation bias and 10.1002/wps.20442 social anxiety: does interpretation bias mediate the relationship between trait Heeren, A., Mogoaşe, C., Philippot, P., and McNally, R. J. (2015). Attention bias social anxiety and state anxiety responses? Cogn. Emot. 19, 1–16. doi: 10.1080/ modification for social anxiety: a systematic review and meta-analysis. Clin. 02699931.2018.1476323 Psychol. Rev. 40, 76–90. doi: 10.1016/j.cpr.2015.06.001 Choi-Kain, L. W., and Gunderson, J. G. (2008). Mentalization: ontogeny, Heinrichs, N., and Hofmann, S. G. (2001). Information processing in social phobia: assessment, and application in the treatment of borderline personality disorder. a critical review. Clin. Psychol. Rev. 21, 751–770. doi: 10.1016/s0272-7358(00) Am. J. Psychiat. 165, 1127–1135. doi: 10.1176/appi.ajp.2008.07081360 00067-2 Clark, D. M., and McManus, F. (2002). Information processing in social phobia. Hezel, D. M., and McNally, R. J. (2014). Theory of mind impairments in social Biol. Psychiat. 51, 92–100. doi: 10.1016/S0006-3223(01)01296-3 anxiety disorder. Behav. Ther. 45, 530–540. doi: 10.1016/j.beth.2014.02.010 Dziobek, I., Fleck, S., Kalbe, E., Rogers, K., Hassenstab, J., Brand, M., et al. (2006). IBM Corp. (2010). IBM SPSS Statistics for Windows (Version 19.0) [Computer Introducing MASC: a movie for the assessment of social cognition. J. Autism software]. Armonk, NY: IBM Corp. Dev. Disord. 36, 623–636. doi: 10.1007/s10803-006-0107-0 Kagan, J., Reznick, S. J., and Snidman, N. (1987). The physiology and psychology Faul, F., Erdfelder, E., Lang, A. G., and Buchner, A. (2007). G∗ Power 3: a flexible of behavioral inhibition in children. Child Dev. 58, 1459–1473. doi: 10.2307/ statistical power analysis program for the social, behavioral, and biomedical 1130685 sciences. Behav. Res. Methods 39, 175–191. doi: 10.3758/BF03193146 Kagan, J., Reznick, S. J., and Snidman, N. (1988). Biological bases of childhood First, M. B., Gibbon, M., Spitzer, R. L., Williams, J. B. W., and Benjamin, L. S. shyness. Science 240, 167–171. doi: 10.1126/science.3353713 (1997). Structured Clinical Interview for DSM-IV Axis II Personality Disorders Kagan, J., Snidman, N., and Arcus, D. (1998). “The value of extreme groups,” in (SCID-II). Washington DC: American Psychiatric Publishing. Methods and Models for Studying the Individual, eds R. B. Cairns and L. R. First, M. B., Gibbon, M., Spitzer, R. L., Williams, J. B. W., and Benjamin, L. S. Bergman (Thousand Oaks, CA: Sage), 65–82. (1999). Entrevista Clínica structurada para los Trastornos de la Personalidad del Kesting, M.-L., and Lincoln, T. M. (2013). The relevance of self-esteem and self- Eje II del DSM-IV. Masson: Barcelona. schemas to persecutory delusions: a systematic review. Compr. Psychiat. 54, Fonagy, P., and Allison, E. (2014). The role of mentalizing and epistemic trust in 766–789. doi: 10.1016/j.comppsych.2013.03.002 the therapeutic relationship. Psychotherapy 51, 372–380. doi: 10.1037/a0036505 Kim, S. (2015). The mind in the making: developmental and neurobiological Fonagy, P., and Bateman, A. (2008). The development of borderline origins of mentalizing. Pers. Disord. 6, 356–365. doi: 10.1037/per0000102 personalityDisorder - A mentalizing model. J. Pers. Disord. 22, 4–21. Lahera, G., Boada, L., Pousa, E., Mirapeix, I., Morón-Nozaleda, G., Marinas, L., doi: 10.1521/pedi.2008.22.1.4 et al. (2014). Movie for the assessment of social cognition (MASC): Spanish Fonagy, P., and Bateman, A. W. (2016). Adversity, attachment, and mentalizing. validation. J. Autism Dev. Disord. 44, 1886–1896. doi: 10.1007/s10803-014- Compr. Psychiat. 64, 59–66. doi: 10.1016/j.comppsych.2015.11.006 2061-6 Fonagy, P., and Luyten, P. (2009). A developmental, mentalization-based Langdon, R., and Coltheart, M. (1999). Mentalising, schizotypy, and schizophrenia. approach to the understanding and treatment of borderline personality Cognition 71, 43–71. doi: 10.1016/s0010-0277(99)00018-9 disorder. Dev. Psychopathol. 21, 1355–1381. doi: 10.1017/S095457940999 Leary, M. R. (1983). A brief version of the fear of negative evaluation scale. Pers. 0198 Soc. Psychol. B 9, 371–375. doi: 10.1177/0146167283093007 Fonagy, P., Luyten, P., and Allison, E. (2015). Epistemic petrification and the Luyten, P., and Fonagy, P. (2015). The neurobiology of mentalizing. Pers. Disord. restoration of epistemic trust: a new conceptualization of borderline personality 6, 366–379. doi: 10.1037/per0000117 disorder and its psychosocial treatment. J. Pers. Disord. 29, 575–609. doi: 10. Martín-Albo, J., Núñiez, J. L., Navarro, J. G., and Grijalvo, F. (2007). The Rosenberg 1521/pedi.2015.29.5.575 self-esteem scale: translation and validation in university students. Span. J. Fonagy, P., Luyten, P., and Strathearn, L. (2011). Borderline personality disorder, Psychol. 10, 458–467. doi: 10.1017/s1138741600006727 mentalization, and the neurobiology of attachment. Inf. Mental Hlth. J. 32, Melo, S. S., and Bentall, R. P. (2013). ‘Poor me’ versus ‘Bad me’ paranoia: the 47–69. doi: 10.1002/imhj.20283 association between self-beliefs and the instability of persecutory ideation. Fonagy, P., and Target, M. (1997). Attachment and reflective function: their Psychol. Pscyhother. 86, 146–163. doi: 10.1111/j.2044-8341.2011.02051.x role in self-organization. Dev. Psychopathol. 9, 679–700. doi: 10.1017/ Miers, A. C., Blöte, A. W., Bögels, S. M., and Westenberg, P. M. (2008). s0954579497001399 Interpretation bias and social anxiety in adolescents. J. Anx. Disord. 22, 1462– Fowler, D., Freeman, D., Smith, B., Kuipers, E., Bebbington, P., Bashforth, H., 1471. doi: 10.1016/j.janxdis.2008.02.010 et al. (2006). The Brief Core Schema Scales (BCSS): psychometric properties Mink, D., Henning, A., and Aschersleben, G. (2014). Infant shy temperament and associations with paranoia and grandiosity in non-clinical and psychosis predict preschoolers theory of mind. Infant Behav. Dev. 37, 66–75. doi: 10.1016/ samples. Psychol. Med. 36, 749–759. doi: 10.1017/S0033291706007355 j.infbeh.2013.12.001 Freeman, D. (2007). Suspicious minds: the psychology of persecutory delusions. Nachmias, M., Gunnar, M., Mangelsdorf, S., Parritz, R. H., and Buss, K. (1996). Clin. Psychol. Rev. 27, 425–457. doi: 10.1016/j.cpr.2006.10.004 Behavioral inhibition and stress reactivity: the moderating role of attachment Freeman, D., Garety, P., Fowler, D., Kuipers, E., Dunn, G., Bebbington, P., et al. security. Child Dev. 67, 508–522. doi: 10.2307/1131829 (1998). The London-East Anglia randomized controlled trial of cognitive- Piqueras, J. A., Espinosa-Fernandez, L., Garcia-Lopez, L. J., and Beidel, D. C. behaviour therapy for psychosis IV: self-esteem and persecutory delusions. (2012). Validation of the social phobia and anxiety inventory-brief form (SPAI- Brit. J. Clin. Psychol. 37(Pt 4), 415–430. doi: 10.1111/j.2044-8260.1998.tb0 B) in Spanish young adults. Behav. Psychol. 20, 505–528. 1399.x Pitarch, M. J. G. (2010). Brief version of the fear of negative evaluation Freeman, D., and Garety, P. A. (2003). Connecting neurosis and psychosis: the scale straightforward items (BFNE-S): psychometric properties in a Spanis direct influence of emotion on delusions and hallucinations. Behav. Res. Ther. population. Span. J. Psychol. 13, 981–989. doi: 10.1017/s1138741600002626 41, 923–947. doi: 10.1016/S0005-7967(02)00104-3 Plana, I., Lavoie, M.-A., Battaglia, M., and Achim, A. M. (2014). A meta-analysis an Frith, U., and Frith, C. D. (2003). Development and neurophysiology of scoping review of social cognition performance in social phobia, posttraumatic mentalizing. Philos. Trans. R. Soc. B 358, 459–473. doi: 10.1098/rstb.2002.1218 stress disorder and other anxiety disorders. J. Anx. Disord. 28, 169–177. doi: Gallego, M. J., Botella, C., Quero, S., Baños, R. M., and García-Palacios, A. (2007). 10.1016/j.janxdis.2013.09.005 Propiedades psicométricas de la Escala de Miedo a la Evaluación Negativa Rapee, R. M., and Spence, S. H. (2004). The etiology of social phobia: empirical versión breve (BFNE) en muestra clínica. Rev. Psicopatol. Psicol. Clín. 12, evidence and an initial model. Clin. Psychol. Rev. 24, 737–767. doi: 10.1016/j. 163–176. doi: 10.5944/rppc.vol.12.num.3.2007.4042 cpr.2004.06.004 Frontiers in Psychology | www.frontiersin.org 10 July 2019 | Volume 10 | Article 1501 Ballespí et al. Hypermentalizing in Social Anxiety Is Context-Dependent Rosenberg, M. (1965). Society and the Adolescent Self-iImage. Princeton, NJ: Valiente, C., Cantero, D., Sánchez, A., Provencio, M., and Wickham, S. (2014). Self- Princeton University Press. esteem and evaluative beliefs in paranoia. J. Behav. Ther. Exp. Psychiatry 45, Sandi-Urena, S., Cooper, M. M., and Stevens, R. H. (2011). Enhancement of 297–302. doi: 10.1016/j.jbtep.2014.01.002 metacognition use and awareness by means of a collaborative intervention. Int. Valiente, C., Cantero, D., Vázquez, C., Sanchez, Á, Provencio, M., and Espinosa, J. Sci. Educ. 33, 323–340. doi: 10.1080/09500690903452922 R. (2011). Implicit and explicit self-esteem discrepancies in paranoia and Schneier, F. R., Blanco, C., Antia, S. X., and Liebowitz, M. R. (2002). The social depression. J. Abnorm. Psychol. 120, 691–699. doi: 10.1037/a0022856 anxiety spectrum. Psychiat. Clin. N. Am. 25, 757–774. doi: 10.1016/s0193- Wampold, B. E. (2015). How important are the common factors in 953x(02)00018-7 psychotherapy? An update. World Psychiatry 14, 270–277. doi: 10.1002/wps. Sharp, C., Ha, C., Carbone, C., Kim, S., Perry, K., Williams, L., et al. (2013). 20238 Hypermentalizing in adolescent inpatients: treatment effects and association Wampold, B. E., and Imel, Z. E. (2015). The Great Psychotherapy Debate: The with borderline traits. J. Pers. Disord. 27, 3–18. doi: 10.1521/pedi.2013.27.1.3 Evidence for What Makes Psychotherapy Work, 2nd Edn. New York, NY: Sharp, C., Pane, H., Ha, C., Venta, A., Patel, A. B., Sturek, J., et al. (2011). Theory of Routledge. mind and emotion regulation difficulties in adolescents with borderline traits. Washburn, D., Wilson, G., Roes, M., Rnic, K., and Harkness, K. L. (2016). Theory J. Am. Acad. Child. Psychiatry 50, 563.e1–573.e1. doi: 10.1016/j.jaac.2011.01.017 of mind in social anxiety disorder, depression, and comorbid conditions. J. Anx. Smith, B., Fowler, D. G., Freeman, D., Bebbington, P., Bashforth, H., Garety, P., Disord. 37, 71–77. doi: 10.1016/j.janxdis.2015.11.004 et al. (2006). Emotion and psychosis: links between depression, self esteem, Wells, A., and Purdon, C. (1999). Metacognition and cognitive-behaviour therapy: negative schematic beliefs and delusions and hallucinations. Schizophr. Res. 86, a special issue. Clin. Psychol. Psychother. 6, 71–72. doi: 10.1016/j.comppsych. 181–188. doi: 10.1016/j.schres.2006.06.018 2015.11.006 Thewissen, V., Bentall, R. P., Lecomte, T., van Os, J., and Myin-Germeys, I. (2008). Wilson, J. K., and Rapee, R. M. (2006). Self-concept certainty in social phobia. Fluctuations in self-esteem and paranoia in the context of daily life. J. Abnorm. Behav. Res. Ther. 44, 113–136. doi: 10.1016/j.brat.2005.01.006 Psychol. 117, 143–153. doi: 10.1037/0021-843X.117.1.143 Zhao, J. J., Kong, F., and Wang, Y. H. (2012). Self-esteem and humor style Tschuschke, V., Crameri, A., Koehler, M., Berglar, J., Muth, K., Staczan, P., et al. as mediators of the effects of shyness on loneliness among Chinese college (2015). The role of therapists’ treatment adherence, professional experience, students. Pers. Indiv. Differ. 52, 686–690. doi: 10.1016/j.paid.2011.12.024 therapeutic alliance, and clients’ severity of psychological problems: prediction of treatment outcome in eight different psychotherapy approaches. Preliminary Conflict of Interest Statement: The authors declare that the research was results of a naturalistic study. Psychother. Res. 25, 420–434. doi: 10.1080/ conducted in the absence of any commercial or financial relationships that could 10503307.2014.896055 be construed as a potential conflict of interest. Turner, S. M., Beidel, D. C., Dancu, C. V., and Stanley, M. A. (1989). An empirically derived inventory to measure social fears and anxiety: the social phobia and Copyright © 2019 Ballespí, Vives, Sharp, Tobar and Barrantes-Vidal. This is an anxiety inventory. Psychol. Assess. 1, 35–40. doi: 10.1037/1040-3590.1.1.35 open-access article distributed under the terms of the Creative Commons Attribution Udachina, A., Thewissen, V., Myin-Germeys, I., Fitzpatrick, S., O’Kane, A., License (CC BY). The use, distribution or reproduction in other forums is permitted, and Bentall, R. P. (2009). Understanding the relationships between self- provided the original author(s) and the copyright owner(s) are credited and that the esteem, experiential avoidance, and paranoia: structural equation modelling original publication in this journal is cited, in accordance with accepted academic and experience sampling studies. J. Nerv. Ment. Dis. 197, 661–668. doi: 10.1097/ practice. No use, distribution or reproduction is permitted which does not comply NMD.0b013e3181b3b2ef with these terms. Frontiers in Psychology | www.frontiersin.org 11 July 2019 | Volume 10 | Article 1501