Skip to main content
Download PDF

Personality subtypes in adults with social anxiety disorder - novelty seeking makes the difference

BMC Psychiatry volume 22, Article number: 832 (2022) Cite this article

Abstract

Background

Up to now several subtypes of social anxiety disorder (SAD) have been proposed.

Methods

In the present study, we used a cluster analytic approach to identify qualitatively different subgroups of SAD based on temperament characteristics, that is, harm avoidance (HA) and novelty seeking (NS) dimensions of Cloninger’s Temperament and Character Inventory.

Results

Based on a large, diverse clinical sample (n = 575), we found evidence for two distinct subgroups of SAD: a larger (59%) prototypic, inhibited cluster characterized by high HA and low NS, and a smaller atypic, and comparatively more impulsive cluster characterized by medium to high HA and increased NS. The subgroups differed regarding a variety of sociodemographic and clinical variables. While the prototypic SAD subtype suffered from more severe SAD and depressive symptoms, suicidal ideation, and reduced social functioning, the atypic NS subtype showcased higher reproductive behaviour, self-directedness and -transcendence, comparatively. Additional hierarchical logistic regression highlights the contribution of age and education.

Conclusions

Our results valuably extend previous evidence for the existence of at least two distinct subtypes of SAD. A better knowledge of the characteristic differences in prototypic behaviour, personality, coping strategies and comorbidities between the identified (and further) subtypes can contribute to the development of effective prevention interventions and promotes the conceptualization of tailored treatments.

Peer Review reports

Background

Social anxiety disorder (SAD) is among the most common anxiety disorders [1]. It is characterized by a marked fear and avoidance of social situations that involve the potential for evaluation or rejection by others. SAD is associated with high rates of comorbidity, in particular with avoidant personality disorder (APD), affective and other anxiety disorders [2, 3]. It is further related to high psychosocial impairment and discussed as a risk factor for depressive and substance use disorders [1, 2, 4].

Despite the disorder-defining characteristics that all individuals with SAD share, several subtypes have been proposed and debated [5, 6]. These subtypes show different phenotypes and are proposed to represent different etiological subgroups, including the genetic factors involved [7]. For instance, based on symptomatology one can differentiate a generalized subtype characterized by fear/avoidance of a broad array of social situations from a subtype with the predominant fear of displaying physical signs of anxiety (see DSM-IV; [8]). This distinction, however, is under critical debate, and is not part of the social phobia characterization raised in DSM-5. In individuals with a performance-only anxiety subtype as classified in DSM-5, anxiety is limited to speaking or performing in public situations [9]. SAD subtypes are further differentiated based on comorbid psychiatric disorders such as alcohol use disorder [10, 11] or, with respect to axis-II, APD [12].

Besides above-mentioned distinctions, more recently, a categorization of SAD subtypes based on the Big-5 personality traits was proposed, consisting of three clusters (prototypical, introvert-conscientious, instable-open) [13]. Especially neuroticism and extraversion are relevant for this categorization. Interestingly, two temperament dimensions, harm avoidance (HA) and novelty seeking (NS) [14, 15], which have been connected to the aforementioned personality traits [16, 17], are of particular interest as well [18, 19]. According to the psychobiological model of personality of Cloninger et al. [15], one can distinguish between three character dimensions (cooperativeness, self-directedness, and self-transcendence), and four temperaments (HA, NS, persistence and reward dependence) that are all supposed to have their corresponding biological substrate. The dimensions can be measured via the well-established “Temperament and Character Inventory” (TCI) questionnaire [15, 20].

Heritability estimates for the TCI scales range from 30–60% [21, 22]. During the past decades, the theory driven conceptualisation of the TCI-temperament and -character dimensions has been discussed and evaluated by a variety of phenotypic and genetic studies [21,22,23,24,25,26,27]. The proposed temperament traits have been found to be associated with several psychiatric disorders. For instance, high levels of HA and avoidance tendencies have been linked to major depression disorder (MDD) [28, 29] and several anxiety disorders [18, 19, 29, 30] while NS and behavioural approach in Eysenck’s model were associated with substance use disorders [25, 31], and, inversely, also anxiety disorders [19, 29, 30]

Focussing on HA and NS, both mirror differences in the automatic emotional response to stimuli. HA, most likely regulated by the serotonin system, is associated with inhibited behaviour and the tendency to respond fearfully to novel situations. In contrast, NS, regulated by the dopamine system, describes a tendency to respond actively to novel stimuli and is associated with impulsive and exploratory behaviour [14, 15]. In line with the prototypical SAD phenotype characterized by behavioural inhibition, avoidance of social situations and shyness are associated with increased HA [18, 32,33,34,35,36,37,38,39], and reduced NS [32, 35, 36]; for meta-analytic findings see [19].

However, there is also evidence for a more atypical subgroup of SAD, still affected by social anxiety, but simultaneously characterized by impulsive, exploratory and risky behaviour, reflecting increased NS [7, 36, 40,41,42,43]. Individuals in the high NS subtype are more often affected by comorbid impulse control, bipolar [41] and substance use disorders [7, 44], suffer more often from severe substance use [42] and greater functional impairment, are less educated and of younger age [41]. Furthermore, there is some evidence that they are less likely to seek for, complete, or fare well in treatment [43]. Taking a closer look at lifetime impulsive control problems, Binelli et al. [7] reported (descriptive) differences between the two subtypes with more frequent suicide attempts and self-harm in the impulsive high NS subgroup. They further differentiated the pattern of substance use and reported a greater use of alcohol in the inhibited/low NS subgroup, as compared to a more pronounced use of substances with a higher-sensation-seeking profile in the impulsive/high NS subgroup. However, this more differentiated pattern of substance use seems to be specific for the analysed sample and was not reported in other studies [44]. Furthermore, reverse findings with increased alcohol use and stronger craving in more impulsive SAD patients have also been reported [45]. Cluster research focusing on the Big-5 personality traits also identified an anxious-extravert subtype characterized by high neuroticism, and simultaneously high levels of extraversion, impulsivity, excitement seeking, and monotony avoidance [13].

In sum, the differentiation of these two SAD subtypes – both characterized by high HA but one displaying low and the other high NS – was repeatedly mirrored in cluster analytic approaches [7, 40, 42, 46] and latent class analyses [41, 47]. With respect to the proposed subtype-characteristics, it needs to be noted that the differences reported in previous studies are still inconsistent. They are partly based on rather homogenous, small (e.g. [7]: n = 142; [42]: n = 82; [40]: n = 125; [46]: n = 84), often mere student’s samples [7, 40]. Beyond that, most previous work neglected potential differences in relevant comorbidities such as anxiety disorders other than SAD, and MDD, or former psychotherapeutic treatment. Therefore, further replication and exploration in large, heterogeneous, clinical samples is needed [18].

A profound characterization of SAD subtypes based on distinct personality profiles can contribute to a valid identification and diagnosis of SAD patients and promotes the conceptualization of tailored treatments [43]. With this in mind, the present study utilised a cluster analytic approach to identify and validate qualitatively different subgroups of SAD based on HA and NS in a large (n = 575) sample of SAD patients. Cluster analysis revolved around the TCI temperament scales HA and NS due to meta-analytic evidence [19] and existing research [7, 40, 42, 46]. We expect to find the best fit for a two-cluster solution differentiating two distinct subtypes: a larger subgroup that shows the prototypic inhibiting SAD pattern with high HA and low NS and a smaller, atypic subgroup with impulsive tendencies, that is, high HA and high NS. Previous research suggests that the two subtypes may differ regarding several sociodemographic and clinical variables. However, based on the rather scarce data basis, we refrain from any specific predictions. Additionally, due to potential implications for therapeutic interventions, personal skill training for SAD individuals and the lack of multivariate research, we conduct a hierarchical logistic regression to identify sociodemographic and psychiatric predictors for cluster affiliation.

Methods

Sample

The final sample consisted of n = 575 [nfemale = 337 (59%); age: 18–78 years; M = 41.60, SD = 13.96] German participants with a DSM-IV lifetime diagnosis of SAD as confirmed by the Structured Clinical Interview for DSM-IV axis-I disorders (SCID-I) [48, 49]. All participants showed clinical significant social anxiety as indicated by a score of 19 or above in the Social Phobia Inventory (SPIN) [50, 51]. Recruitment was realised within the context of the German “Social Phobia Research” project which was set up in 2012 as a collaboration between the Department of Psychosomatic Medicine and Psychotherapy and the Institute of Human Genetics at the University of Bonn, Germany. Participants were recruited via various access paths (e.g., clinical services, cooperation with external clinics and outpatient practitioners, internet, reports on regional TV and radio channels, newspaper advertisements and articles, advertisements on urban areas). Exclusion criteria were a comorbid psychosis, inadequate German language skills or somatic and/or mental difficulties in completing inserted questionnaires. The study was approved by the local ethics committee of the University of Bonn and was in accordance with the declaration of Helsinki. All participants gave written informed consent.

Measures

Demographic information was assessed by a standardised questionnaire. This included sex, age, current partnership status and level of education. Furthermore, family psychiatric history and previous treatment for mental disorders were enquired.

Diagnoses of SAD and relevant psychiatric comorbidities, including MDD, anxiety disorders, posttraumatic stress disorder (PTSD), and substance use disorders (alcohol, cannabis, other substances), were assigned by the German version of the SCID-I while APD was assessed by the German version of the SCID-II [48, 49] administered by trained interviewers. Due to economic reasons and the high comorbidity between SAD and APD, only the APD section of the SCID-II was considered. Interrater-reliability for SCID-I ranges from kappa values 0.61 and 0.83 and for SCID-II, a kappa value of 0.83 for APD was achieved [52].

Severity of social anxiety symptoms was measured by the German version of the SPIN [50, 51]. As a valid brief self-report measure, the SPIN captures behavioural, physiological, and cognitive symptoms of social anxiety on 17 items using a four-point Likert-scale ranging from not at all to extremely. The total sum score ranges between 0 and 68. Total scores of 19 and above indicate clinical relevant social anxiety [50].

The German version of the Beck Depression Inventory (BDI) [53, 54] served as a measure of severity of potential comorbid depressive symptoms. On a four-point Likert scale (0–3), 21 items ask participants for a self-report of past weeks depressive symptoms. Sum scores range from 0 to 63. The BDI provides a highly valid and reliable measure of depressive symptoms [20]. As a short measure of suicidality, we further analysed item I of the BDI that explicitly asks participants for any suicidal ideations.

Personality was assessed by the German version of the TCI [15, 55], a valid, psychological and neurobiological based, measure of temperament and character. Using 240 items, the TCI captures four temperaments (HA, NS, persistence, and reward dependence) and three character dimensions (cooperativeness, self-directedness, and self-transcendence) by means of a dichotomous true/false format. T-scores for each scale was utilised, which are based on norm values from healthy subjects.

To verify the validity of the TCI-based SAD-clusters, we further included a short measure of behavioural inhibition and behavioural approach [56,57,58,59]. On a theoretical level, HA and the behavioural inhibition system (BIS) and NS and the behavioural approach system (BAS) share several properties [60, 61]. Accordingly, the atypic-impulsive subtype should be characterized by increased BAS, while the prototypic-inhibited subtype should exhibit increased BIS. The applied short version of the action regulating emotions system scales (ARES scales) measures BIS and BAS sensitivity by means of 20 four-point Likert-scaled items [62]. Reliability (Cronbach’s α) of all questionnaires was good to excellent, with few exceptions among subscales (see Table 1).

Table 1 Reliability of the applied questionnaires in the given data
Full size table

Statistical analyses

To identify personality-based subgroups of SAD, the global T-Scores of HA and NS were submitted to a hierarchical cluster analysis using Ward’s method and squared Euclidean distance. The number of clusters was determined by a dendrogram which served as the basis for a subsequent k-means cluster analysis. The formal evaluation of the adequate number of clusters was based on the formal fit indicators ETA2, PRE, and FMX [63]. Cluster solutions containing two to six clusters were analysed. The best fitting cluster solution was evaluated by stability checks using Cohen’s kappa [64]. Internal validity of the clusters was evaluated with respect to BIS and BAS scales.

Potential differences between the identified clusters regarding sociodemographic-, psychiatric-, and personality variables were examined using t-tests for quantitative variables and chi2-tests or Fisher’s exact test for categorical data. Multiple testing was accounted for by adjusting the p-value utilising the Benjamini–Hochberg procedure with a false discovery rate of 0.05 [65]. Based on 63 significance tests, critical values were calculated for each p-value. After identifying the largest p-value that is still under the critical value (here: p = 0.011, Benjamini–Hochberg critical value = 0.021), all results with lower p-values were considered significant. Cohen’s d and Cramer’s V served as measures of effect size.

A hierarchical logistic regression with cluster type as the dependent variable was further conducted as suggested by one of the reviewers. Sociodemographic variables (age, sex, partnership, education, smoking status, medication) were entered first followed by psychiatric comorbidities (panic disorder, agoraphobia, generalized anxiety disorder, specific phobia, posttraumatic stress disorder, major depression disorder, APD, alcohol- and substance-related disorders). A correlation matrix for all predictors detected no multicollinearity. Goodness-of-fit was evaluated by the Hosmer–Lemeshow-Test while Nagelkerke’s R2 represented the explained variance by the predictors. All analyses were performed using SPSS Version 27 [66].

Results

Results of the hierarchical cluster analysis

The combined analysis of formal fit indicators revealed the best fit for a two-cluster solution which explained 46% of variance (ETA2 = 0.46) and showed best PRE (PRE = 0.46) and FMX (FMX = 404.09) values (for a direct comparison with alternative three- to six-cluster solutions, see supplementary table A). The two-cluster solution was further supported by visual inspection of plotted fit-values with clearest peaks at the two-cluster solution. Stability checks based on Cohen’s kappa revealed an assignment of cases to clusters that was independent of the starting values. That is, the two-cluster solution proofed high stability with all Cohen’s kappa >|.940|.

The first, larger cluster was characterized by very high HA (+ 2 standard deviation (SD)) and low NS (- 1 SD) T-scores. 59% (n = 338) of the sample were assigned to this prototypic cluster. The second, smaller cluster was characterized by high HA (+ 1 SD) and average NS T-scores. 41% (n = 237) of the sample were assigned to this atypic cluster (see Table 2).

Table 2 Comparison of TCI dimensions, sub dimension and BIS/BAS scores between SAD subtype groups
Full size table

The internal validity of the two-cluster solution was further proved by the pattern of differences in BIS and BAS sensitivity. In line with our predictions, the prototypic cluster showed high BIS and lower BAS values, while the atypic cluster was characterized by still high, but compared to the prototypic cluster, reduced BIS and higher BAS values.

Subtype comparison

Total NS and all NS subscale T-scores were greater in the atypic compared to the prototypic cluster. For HA, the pattern of results was reverse, with higher total HA and HA subscale T-scores in the prototypic compared to the atypic cluster (Table 2). Concerning the other TCI subscales, higher T-scores in “self-directedness” and “self-transcendence” were reported in the atypic cluster while the prototypic cluster achieved higher T-scores in “persistence”.

The prototypic cluster contains younger and more SAD individuals with children than the prototypic cluster but there were no significant differences regarding other sociodemographic variables (see Table 3). However, individuals in the prototypic cluster descriptively showed reduced levels of social functioning as indicated by less partnerships and reduced education. They further reported significantly higher symptom severity as measured by the SPIN. While there were no differences regarding former psychotherapeutic treatment of SAD, an increased number of individuals in the prototypic cluster reported a history of former medical treatment.

Table 3 Comparison of empirical clusters based on TCI-NS and TCI-HA dimensions
Full size table

There were no significant differences regarding family psychiatric history (neither in general, nor specifically for SAD). Individuals in both clusters showed a similar amount of current and lifetime comorbid anxiety disorders (panic disorder, agoraphobia, generalized anxiety disorder, specific phobia) and PTSD. However, individuals in the prototypic cluster were more often affected by a comorbid lifetime MDD (nominally significant), reported higher BDI scores, and had higher score on the suicidality-item I of the BDI. Furthermore, lifetime APD was overrepresented in the prototypic cluster. With respect to current and lifetime substance use behaviour, there was a higher number of individuals in the atypic cluster that was affected by comorbid lifetime alcohol abuse and lifetime cannabis abuse, but no significant difference could be observed.

Hierarchical logistic regression

The Hosmer–Lemeshow-Test yielded no significance for this logistic model, representing a good model fit (χ2(8) = 1.57, p = 0.99). Correct classification of SAD individuals based on included predictors lies at 65.6%. Both models overall achieved significance compared to a model without any predictors (Step 1: χ2(6) = 44.81, p =  < 0.001, Step 2: χ2(15) = 53.535, p =  < 0.001), however, if we look at the sole contribution of Step 2, the included predictors did not significantly improve the model (χ2(9) = 8.73, p = 0.46). A similar result can be found for each predictor in Step 2, as none of them can significantly predict the odds of being in the prototypic or atypic cluster (Table 4). In both models, age and a German high school degree reached significance after Benjamini–Hochberg adjustment. With each year, the odds of being in the atypic cluster is increased by 4%. Likewise, having a German high school degree increases the odds for the atypic cluster by over 40%. Altogether, the models explain 12% and 14% of the variance (Nagelkerke’s R2).

Table 4 Hierarchical logistic regression (n = 503)
Full size table

Discussion

In the present study, we used a cluster analytic approach to characterize subtypes of SAD patients based on temperament traits of the TCI (HA and NS) in a large, diverse clinical sample (n = 575). In line with previous findings, we identified two distinct clusters, a larger subgroup of SAD patients showed a quite prototypic pattern with very high HA and low NS. In contrast, a smaller group of patients was characterized by high HA and average NS. These findings support previous reports of an atypic, more NS orientated subtype of SAD [7, 40,41,42, 46]. The validity of the cluster solution was further confirmed by the pattern of differences in BIS and BAS sensitivity. The two subtypes markedly differed with respect to several sociodemographic and clinical variables. These qualitative differences may entail critical implications for treatment approaches and treatment success [36, 43, 46].

Overall, female participants were overrepresented in our sample (59% female). However, this unbalanced sex distribution is well in line with epidemiological findings on the sex ratio of psychiatric diseases and SAD [67, 68]. There were no sex differences between the two clusters. Participants in the prototypic cluster were younger and, in tendency, showed increased daily life impairments as indicated by reduced education and a reduced probability of being in a partnership. They showed higher SAD symptom severity and more likely reported a history of former psychopharmaceutic treatment. Prototypic SAD patients more often suffered from a comorbid lifetime MDD, reported more severe depressive symptoms and suicidal ideation, and were more often affected by a comorbid APD. In contrast, participants in the atypic cluster showcased higher reproductive behaviour, self-directedness and -transcendence.

Thus, our data suggest the existence of two distinct subtypes of SAD: a prototypic subtype with more severe SAD and depressive symptoms that more often suffers from comorbid MDD, APD and increased social impairment. This depressive subtype is younger, and, in line with the increased symptom severity, more likely receives psychopharmacological treatment and reports more suicidal ideation. Conversely, the NS subtype shows less severe SAD and depressive symptoms, and a higher level of social functioning and reproductive behaviour.

In our sample, we found reduced symptom severity in the atypic cluster. This is well in line with previous reports of SAD symptom severity being positively related to HA [18, 19, 33, 37], and related concepts such as neuroticism [13], while negatively associated with NS [19], and related concepts such as openness, extraversion or impulsivity [13]. While increased HA implies a rather cautious, restrained behaviour, increased NS entails an approach tendency towards and increased interest in new, unknown stimuli [14]. This might also include anxiety-associated social situations. In terms of habituation, this NS-initiated self-exposure can serve as a functional self-regulatory approach reducing anxiety/SAD symptoms. Even though initially conceptualized as a stable trait, one might further speculate that this tendency towards an average NS behavioural strategy develops over the course of the SAD impairment, and, therefore, is more present in SAD participants of older age. This speculation would fit with the older age observed in the NS subtype.

Diverging from our results, previous studies did not report any differences in symptom severity [7, 42]. This might be due to rather small sample sizes and reduced variance in previous studies that only included specific SAD-subtypes [42] or patients that reported high symptom severity [7]. In contrast, striving to mirror the entire range of clinically relevant SAD symptoms, we included all participants with a clinically relevant SAD diagnosis in our study. At the same time, however, external validity might not be given for SAD individuals who showcase rather low symptom severity or those, who are only afraid of certain social situations as the distribution of generalized SAD was the same for both clusters.

Together with SAD symptom severity, participants in the atypic cluster less often suffered from comorbid MDD and showed reduced depressive symptom severity as measured by the BDI. These results fit with previous findings that showed reduced HA, as it was observed in the atypic cluster, to be associated with reduced depressive symptoms [37]. In line with that, a reduction of HA in the course of SAD treatment comes along with a significant reduction of depressive symptoms [34]. Even though HA is conceptualized as a stable trait, there is evidence that it might be affected by successful treatment of SAD [34, 36]. The same might be the case for NS. Based on the present findings, a modulation of HA and/or NS might provide a promising approach to be integrated in innovative SAD therapy. Crucially, higher HA has been linked to poorer outcome in different types of psychotherapies [18, 46, 69]. Reducing HA related safety behaviour that, combined with a comorbid APD, likely contributes to the maintenance of the disease [70], might reduce SAD severity. It could further increase responsiveness to cognitive behavioural therapy [36, 46].

Potentially, strengthening individual NS tendencies up to a certain degree could ease social interaction in everyday life, reduce functional impairment and increase the responsiveness for approach related therapeutic concepts. Reduced HA and increased NS behaviour in the atypic subtype may contribute to reduced symptom severity and thereby reduce the probability of burdening comorbidity with MDD and associated suicidal ideation. Then again, the overall reduced level of suffering allows higher levels of social functioning in the NS subtype. In contrast in the prototypic cluster, increased HA, conceptualized as an innate responsiveness to avoid punishment, may foster the development of SAD and secondary depressive symptoms. However, even though the proposed relations seem plausible, they remain rather speculative, as the cross-sectional approach of the present study prohibits any causal interpretation.

Despite potential protective effects of increased NS, this temperament dimension is also associated with impulsive, unpredictable behaviour [14]. Thus, it might increase harmful impulse-related behaviour such as self-harm, suicide attempts, and substance misuse [42]. In line with that, previous studies reported (at least a tendency towards) increased suicidal attempts and substance use in the identified atypic, high NS cluster [7, 42, 44]. However, there are also divergent findings. For instance, Kashdan and Hofmann [42] reported no differences in the occurrence of substance use disorder and the results in this present study also did not showcase increased substance or alcohol use in the atypic cluster. Based on our cross-sectional results, increasing NS tendencies to an average degree, comparable to healthy subjects, do not seemingly increase harmful impulse-related behaviour. Furthermore, considering the same distribution of generalized SAD in both clusters, presumably, NS tendencies do not influence the quantity but rather the handling of social situations in which SAD individuals face insecurities and anxiety.

Concerning substance use, it might function as a rather dysfunctional try of coping by creating safety, sedation or distraction when facing SAD symptoms [7, 42]. Such risky behaviour seems more likely in more impulsive, NS oriented individuals. In terms of alcohol it might be the case, that increased impulsivity promotes abuse and dependence, as it draws attention to the immediate, anxiolytic effects of alcohol instead of focusing potential negative (social) outcomes such as embarrassing behaviour [45]. Furthermore, reduced HA in the atypic cluster may come along with less pronounced protective behavioural strategies, that is, cognitive and behavioural harm-reducing strategies to decrease alcohol use and alcohol problems [4]. As our results do not support this evidence, in future studies, a longitudinal approach in combination with a dimensional, instead of the applied categorical measure of (subclinical) substance-related behaviour, would provide further inside into the proposed relations.

While substance use can be interpreted as a manifestation of more venturesome NS behaviour, our findings of increased suicidal ideation in the prototypic cluster, more likely are a consequence of the overall increased disease burden, that is, increased SAD symptom severity, increased HA, and comorbid depressive symptoms in this cluster. Accordingly, high HA has been associated with suicidal ideation and the total risk of suicide [71]. Furthermore, there is some evidence that suicidal ideation and self-harm attempts are observed more frequently in some personality disorders, among others, APD, which we observed to occur more frequent in the prototypic subtype [72]. However, as we only applied a one-item measure of suicidal ideation, all these speculative interpretations must be treated with caution and need further exploration.

Personality disorders, like APD, have been associated with low self-directedness (SDi) [73], another aspect that is in line with the results in this sample as the prototypic cluster, which contains more SAD individuals with APD, had lower SDi than the atypic cluster. Individuals with high SDi are described as mature, reliable, goal-oriented and self-confident while those with low SDi are frail, irresponsible and poorly integrated without external direction [15]. The former characteristics are reflected in positive outcomes e.g. wellbeing [74, 75], autonomy, self-acceptance [74] and treatment outcome in depressive patients [76] whereas the latter are connected to mood disorders [77] and higher psychopathological distress [78]. Based on the T-scores, however, both clusters are still below the healthy norm group in this dimension, indicating difficulties in goal-oriented behaviour and positive self-perception for both clusters. This is particularly true for the prototypic cluster, as it achieved a low SDi score two SDs below the norm group. Combined with highly increased HA, these characteristics might be connected to increased SAD and depressive symptom severity.

Our multivariate approach further highlights the role of age. Again, with older age might come the willingness to engage and face more novel social situations. Interestingly, SAD individuals with a German high school degree are at greater odds to be in the atypic cluster, which might be related to the lower levels of HA compared to the prototypic cluster rather than the higher NS tendencies [79]. Recent research supports this relation as SAD individuals are, for example, less likely to start or obtain a university degree or obtain postgraduate education compared to healthy and unaffected siblings [80]. Longitudinal results showcased the hampering influence of SAD symptoms on social ties and academic achievement [81]. As the prototypic cluster report higher SAD symptom severity, combined with high HA, the prototypic SAD individuals are at risk of facing difficulties in educational settings. Knowing the relationship between HA and SAD symptom severity might help to tailor appropriate school or university interventions to adequately support and address the ones in need.

On a descriptive level, it is interesting to point out, that NS oriented characteristics (e.g. having children, being in a relationship, smoking, alcohol- and substance-related disorder) and HA oriented characteristics (e.g. higher medication intake, higher comorbidity rate) reached Odds ratio’s that are cluster conform. This might be a sign for future research to highlight and assess more diverse NS and HA oriented characteristics or behaviour to fully grasp the concept and nature of the prototypic and atypic cluster of SAD.

One further important, practically highly relevant clinical implication of the proposed categorisation of SAD patients lies in a more in-depth characterization of the atypic subtype. In case of a rather superficial exploration, the characteristics of the atypic SAD patients may bear the risk of missing out a diagnosis [43]. This risk seems particularly important, as these patients might be especially open minded and approachable to therapeutic intervention. First, due to the increased approach tendency, they might be more open towards even starting a therapy. Additionally, they might more likely take an active part in confrontational therapeutic interventions, which should increase treatment efficacy (for relevance of confrontational approaches in cognitive behavioural therapy see e.g. [82]). Besides that, early intervention seems especially important to prevent the development of dysfunctional coping strategies in terms of substance misuse in atypic SAD patients.

Conclusion

The present study provides further evidence for the heterogeneity of SAD [5, 6]. It contributes to a better understanding of SAD characteristics, and therewith can provide several important implications for the clinical routine. Our data support the existence of distinct subtypes of SAD patients. Besides a prototypic SAD subtype suffering from more severe SAD symptoms, comorbid depression, and functional impairment, there is a more NS oriented subtype with less severe symptoms, a higher level of social functioning and self-directedness. A better knowledge of the characteristic differences in prototypic behaviour, personality, coping strategies and comorbidities between the identified (and, perhaps further) subtypes can contribute to the development of effective prevention interventions [83], and promotes the conceptualization of tailored treatments.

Availability of data and materials

Data are available from the corresponding author on reasonable request.

References

  1. Stein MB, Stein DJ. Social anxiety disorder. The Lancet. 2008;371:1115–25. https://doi.org/10.1016/S0140-6736(08)60488-2.

    Article  Google Scholar 

  2. Filho AS, Hetem LAB, Ferrari MCF, Trzesniak C, Martín-Santos R, Borduqui T, et al. Social anxiety disorder: what are we losing with the current diagnostic criteria? Acta Psychiatr Scand. 2010;121:216–26. https://doi.org/10.1111/j.1600-0447.2009.01459.x.

    Article  CAS  Google Scholar 

  3. Grant BF, Hasin DS, Blanco C, Stinson FS, Chou SP, Goldstein RB, et al. The Epidemiology of Social Anxiety Disorder in the United States: Results From the National Epidemiologic Survey on Alcohol and Related Conditions. J Clin Psychiatry. 2005;66:1351–61. https://doi.org/10.4088/JCP.v66n1102.

    Article  Google Scholar 

  4. Terlecki MA, Ecker AH, Buckner JD. The role of underutilization of protective behavioral strategies in the relation of social anxiety with risky drinking. Addict Behav. 2020;100:106122. https://doi.org/10.1016/j.addbeh.2019.106122.

    Article  Google Scholar 

  5. D’Avanzato C, Dalrymple KL. Recent Insight Into the Subtypes of Social Anxiety Disorder. Curr Psychiatry Rep. 2016;18:50. https://doi.org/10.1007/s11920-016-0688-z.

    Article  Google Scholar 

  6. Hyett MP, McEvoy PM. Social anxiety disorder: looking back and moving forward. Psychol Med. 2018;48:1937–44. https://doi.org/10.1017/S0033291717003816.

    Article  Google Scholar 

  7. Binelli C, Muñiz A, Sanches S, Ortiz A, Navines R, Egmond E, et al. New evidence of heterogeneity in social anxiety disorder: Defining two qualitatively different personality profiles taking into account clinical, environmental and genetic factors. Eur Psychiatry. 2015;30:160–5. https://doi.org/10.1016/j.eurpsy.2014.09.418.

    Article  CAS  Google Scholar 

  8. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR). vol. 1. 4th ed. Arlington, VA: American Psychiatric Association; 2000. https://doi.org/10.1176/appi.books.9780890423349.

  9. Blöte AW, Kint MJW, Miers AC, Westenberg PM. The relation between public speaking anxiety and social anxiety: A review. J Anxiety Disord. 2009;23:305–13. https://doi.org/10.1016/j.janxdis.2008.11.007.

    Article  Google Scholar 

  10. Rambau S, Forstner AJ, Wegener I, Mücke M, Wissussek CTS, Staufenbiel SM, et al. Childhood adversities, bonding, and personality in social anxiety disorder with alcohol use disorder. Psychiatry Res. 2018;262:295–302. https://doi.org/10.1016/j.psychres.2018.02.006.

    Article  Google Scholar 

  11. Schneier FR, Foose TE, Hasin DS, Heimberg RG, Liu S-M, Grant BF, et al. Social anxiety disorder and alcohol use disorder co-morbidity in the National Epidemiologic Survey on Alcohol and Related Conditions. Psychol Med. 2010;40:977–88. https://doi.org/10.1017/S0033291709991231.

    Article  CAS  Google Scholar 

  12. Lampe L. Avoidant personality disorder as a social anxiety phenotype: risk factors, associations and treatment. Curr Opin Psychiatry. 2016;29:64–9. https://doi.org/10.1097/YCO.0000000000000211.

    Article  Google Scholar 

  13. Costache ME, Frick A, Månsson K, Engman J, Faria V, Hjorth O, et al. Higher- and lower-order personality traits and cluster subtypes in social anxiety disorder. PLoS ONE. 2020;15:e0232187. https://doi.org/10.1371/journal.pone.0232187.

    Article  CAS  Google Scholar 

  14. Cloninger CR. A Systematic Method for Clinical Description and Classification of Personality Variants: A Proposal. Arch Gen Psychiatry. 1987;44:573. https://doi.org/10.1001/archpsyc.1987.01800180093014.

    Article  CAS  Google Scholar 

  15. Cloninger CR. A Psychobiological Model of Temperament and Character. Arch Gen Psychiatry. 1993;50:975. https://doi.org/10.1001/archpsyc.1993.01820240059008.

    Article  CAS  Google Scholar 

  16. De Fruyt F, Van De Wiele L, Van Heeringen C. Cloninger’s Psychobiological Model of Temperament and Character and the Five-Factor Model of Personality. Personal Individ Differ. 2000;29:441–52. https://doi.org/10.1016/S0191-8869(99)00204-4.

    Article  Google Scholar 

  17. Aluja A, Blanch A. The Five and Seven Factors Personality Models: Differences and Similitude between the TCI-R, NEO-FFI-R and ZKPQ-50-CC. Span J Psychol. 2011;14:659–66. https://doi.org/10.5209/rev_SJOP.2011.v14.n2.14.

    Article  Google Scholar 

  18. Kampman O, Viikki M, Leinonen E. Anxiety Disorders and Temperament—an Update Review. Curr Psychiatry Rep. 2017;19:27. https://doi.org/10.1007/s11920-017-0779-5.

    Article  Google Scholar 

  19. Kampman O, Viikki M, Järventausta K, Leinonen E. Meta-Analysis of Anxiety Disorders and Temperament. Neuropsychobiology. 2014;69:175–86. https://doi.org/10.1159/000360738.

    Article  Google Scholar 

  20. Richter P, Werner J, Heerlein A, Kraus A, Sauer H. On the Validity of the Beck Depression Inventory. Psychopathology. 1998;31:160–8. https://doi.org/10.1159/000066239.

    Article  CAS  Google Scholar 

  21. Yang S, Sung J, Kim J-H, Song Y-M, Lee K, Kim H-N, et al. Some personality traits converge gradually by long-term partnership through the lifecourse – Genetic and environmental structure of Cloninger’s temperament and character dimensions. J Psychiatr Res. 2015;63:43–9. https://doi.org/10.1016/j.jpsychires.2015.01.020.

    Article  CAS  Google Scholar 

  22. Zwir I, Arnedo J, Del-Val C, Pulkki-Råback L, Konte B, Yang SS, et al. Uncovering the complex genetics of human temperament. Mol Psychiatry. 2020;25:2275–94. https://doi.org/10.1038/s41380-018-0264-5.

    Article  CAS  Google Scholar 

  23. GlavinaJelaš I, Dević I, Karlović D. Cloninger’s temperament and character dimensions and dopaminergic genes: DAT1 VNTR and COMT Val158Met polymorphisms. Psychiatr Danub. 2018;30:47–56. https://doi.org/10.24869/psyd.2018.47.

    Article  CAS  Google Scholar 

  24. He Y, Martin N, Zhu G, Liu Y. Candidate genes for novelty-seeking: a meta-analysis of association studies of DRD4 exon III and COMT Val158Met. Psychiatr Genet. 2018;28:97–109. https://doi.org/10.1097/YPG.0000000000000209.

    Article  CAS  Google Scholar 

  25. Oreland L, Lagravinese G, Toffoletto S, Nilsson KW, Harro J, Robert Cloninger C, et al. Personality as an intermediate phenotype for genetic dissection of alcohol use disorder. J Neural Transm. 2018;125:107–30. https://doi.org/10.1007/s00702-016-1672-9.

    Article  Google Scholar 

  26. Pawlak J, Dmitrzak-Węglarz M, Maciukiewicz M, Kapelski P, Czerski P, Leszczyńska-Rodziewicz A, et al. Personality traits as an endophenotype in genetic studies on suicidality in bipolar disorder. Acta Neuropsychiatr. 2017;29:115–21. https://doi.org/10.1017/neu.2016.43.

    Article  CAS  Google Scholar 

  27. Service SK, Verweij KJH, Lahti J, Congdon E, Ekelund J, Hintsanen M, et al. A genome-wide meta-analysis of association studies of Cloninger’s Temperament Scales. Transl Psychiatry. 2012;2:e116–e116. https://doi.org/10.1038/tp.2012.37.

    Article  Google Scholar 

  28. Kampman O, Poutanen O. Can onset and recovery in depression be predicted by temperament? A systematic review and meta-analysis. J Affect Disord. 2011;135:20–7. https://doi.org/10.1016/j.jad.2010.12.021.

    Article  Google Scholar 

  29. Rodebaugh TL, Levinson CA, Langer JK, Weeks JW, Heimberg RG, Brown PJ, et al. The structure of vulnerabilities for social anxiety disorder. Psychiatry Res. 2017;250:297–301. https://doi.org/10.1016/j.psychres.2017.01.073.

    Article  Google Scholar 

  30. Miettunen J, Raevuori A. A meta-analysis of temperament in axis I psychiatric disorders. Compr Psychiatry. 2012;53:152–66. https://doi.org/10.1016/j.comppsych.2011.03.008.

    Article  Google Scholar 

  31. Sher KJ, Bartholow BD, Wood MD. Personality and substance use disorders: A prospective study. J Consult Clin Psychol. 2000;68:818–29. https://doi.org/10.1037/0022-006X.68.5.818.

    Article  CAS  Google Scholar 

  32. Chatterjee S, Sunitha TA, Velayudhan A, Khanna S. An investigation into the psychobiology of social phobia: personality domains and serotonergic function. Acta Psychiatr Scand. 1997;95:544–50. https://doi.org/10.1111/j.1600-0447.1997.tb10144.x.

    Article  CAS  Google Scholar 

  33. Faytout M, Tignol J, Swendsen J, Grabot D, Aouizerate B, Lépine JP. Social phobia, fear of negative evaluation and harm avoidance. Eur Psychiatry. 2007;22:75–9. https://doi.org/10.1016/j.eurpsy.2005.07.009.

    Article  CAS  Google Scholar 

  34. Hofmann SG, Loh R. The Tridimensional Personality Questionnaire: Changes during psychological treatment of social phobia. J Psychiatr Res. 2006;40:214–20. https://doi.org/10.1016/j.jpsychires.2005.03.010.

    Article  Google Scholar 

  35. Lochner C, Hemmings S, Seedat S, Kinnear C, Schoeman R, Annerbrink K, et al. Genetics and personality traits in patients with social anxiety disorder: A case-control study in South Africa. Eur Neuropsychopharmacol. 2007;17:321–7. https://doi.org/10.1016/j.euroneuro.2006.06.010.

    Article  CAS  Google Scholar 

  36. Mörtberg E, Bejerot S, Åberg WA. Temperament and character dimensions in patients with social phobia: Patterns of change following treatments? Psychiatry Res. 2007;152:81–90. https://doi.org/10.1016/j.psychres.2006.10.003.

    Article  Google Scholar 

  37. Pélissolo A, André C, Pujol H, Yao SN, Servant D, Braconnier A, et al. Personality dimensions in social phobics with or without depression: Personality dimensions in social phobia. Acta Psychiatr Scand. 2002;105:94–103. https://doi.org/10.1034/j.1600-0447.2002.01115.x.

    Article  Google Scholar 

  38. Marteinsdottir I, Tillfors M, Furmark T, Anderberg UM, Ekselius L. Personality dimensions measured by the Temperament and Character Inventory (TCI) in subjects with social phobia. Nord J Psychiatry. 2003;57:29–35. https://doi.org/10.1080/08039480310000239.

    Article  Google Scholar 

  39. Paavonen V, Luoto K, Lassila A, Leinonen E, Kampman O. Temperament clusters associate with anxiety disorder comorbidity in depression. J Affect Disord. 2018;236:252–8. https://doi.org/10.1016/j.jad.2018.04.084.

    Article  Google Scholar 

  40. Kashdan TB, Elhai JD, Breen WE. Social anxiety and disinhibition: An analysis of curiosity and social rank appraisals, approach–avoidance conflicts, and disruptive risk-taking behavior. J Anxiety Disord. 2008;22:925–39. https://doi.org/10.1016/j.janxdis.2007.09.009.

    Article  Google Scholar 

  41. Kashdan TB, McKnight PE, Richey JA, Hofmann SG. When social anxiety disorder co-exists with risk-prone, approach behavior: Investigating a neglected, meaningful subset of people in the National Comorbidity Survey-Replication. Behav Res Ther. 2009;47:559–68. https://doi.org/10.1016/j.brat.2009.03.010.

    Article  Google Scholar 

  42. Kashdan TB, Hofmann SG. The high-novelty–seeking, impulsive subtype of generalized social anxiety disorder. Depress Anxiety. 2008;25:535–41. https://doi.org/10.1002/da.20382.

    Article  Google Scholar 

  43. Kashdan TB, McKnight PE. The Darker Side of Social Anxiety: When Aggressive Impulsivity Prevails Over Shy Inhibition. Curr Dir Psychol Sci. 2010;19:47–50. https://doi.org/10.1177/0963721409359280.

    Article  Google Scholar 

  44. Nicholls J, Staiger PK, Williams JS, Richardson B, Kambouropoulos N. When social anxiety co-occurs with substance use: Does an impulsive social anxiety subtype explain this unexpected relationship? Psychiatry Res. 2014;220:909–14. https://doi.org/10.1016/j.psychres.2014.08.040.

    Article  Google Scholar 

  45. Adams T, Rapinda KK, Frohlich JR, O’Connor RM, Keough MT. Impulsivity moderates the effect of social anxiety on in-lab alcohol craving. Addict Behav. 2019;97:70–6. https://doi.org/10.1016/j.addbeh.2019.05.025.

    Article  Google Scholar 

  46. Mörtberg E, Tillfors M, van Zalk N, Kerr M. An atypical anxious-impulsive pattern of social anxiety disorder in an adult clinical population. Scand J Psychol. 2014;55:350–6. https://doi.org/10.1111/sjop.12117.

    Article  Google Scholar 

  47. Lipton MF, Weeks JW, Daruwala SE, De Los RA. Profiles of Social Anxiety and Impulsivity Among College Students: A Close Examination of Profile Differences in Externalizing Behavior. J Psychopathol Behav Assess. 2016;38:465–75. https://doi.org/10.1007/s10862-015-9531-9.

    Article  Google Scholar 

  48. First MB, Spitzer RL, Gibbon M, Williams JBW. Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I). Clinician Version: Administration Booklet. American Psychiatric Association Publishing; 1996.

    Google Scholar 

  49. Wittchen H-U, Zaudig M, Fydrich T. SKID: Strukturiertes Klinisches Interview für DSM-IV. Achse I und II. Handanweisung. Göttingen: Hogrefe; 1997.

    Google Scholar 

  50. Connor KM, Davidson JR, Churchill LE, Sherwood A, Foa E, Weisler RH. Psychometric properties of the Social Phobia Inventory (SPIN). New self-rating scale. Br J Psychiatry J Ment Sci. 2000;176:379–86.

    Article  CAS  Google Scholar 

  51. Sosic Z, Gieler U, Stangier U. Screening for social phobia in medical in- and outpatients with the German version of the Social Phobia Inventory (SPIN). J Anxiety Disord. 2008;22:849–59. https://doi.org/10.1016/j.janxdis.2007.08.011.

    Article  CAS  Google Scholar 

  52. Lobbestael J, Leurgans M, Arntz A. Inter-rater reliability of the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID I) and Axis II Disorders (SCID II). Clin Psychol Psychother. 2011;18:75–9. https://doi.org/10.1002/cpp.693.

    Article  Google Scholar 

  53. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch Gen Psychiatry. 1961;4:561–71.

    Article  CAS  Google Scholar 

  54. Hautzinger M, Bailer M, Worrall H, Keller F. Beck-Depressions-Inventar (BDI). Testhandbuch. 1. Auflage. Bern: Hans Huber; 1994.

  55. Richter J, Eisemann M, Richter G. Zur deutschsprachigen Version des Temperament- und Charakterinventars. Z Für Klin Psychol Psychother. 2000;29:117–26. https://doi.org/10.1026//0084-5345.29.2.117.

    Article  Google Scholar 

  56. Gray JA. A Critique of Eysenck’s Theory of Personality. In: Eysenck HJ, editor. Model Personal., Berlin, Heidelberg: Springer Berlin Heidelberg; 1981, p. 246–76. https://doi.org/10.1007/978-3-642-67783-0_8.

  57. Eysenck HJ. Dimensions of personality. New York: Praeger; 1947.

    Google Scholar 

  58. Eysenck HJ. Biological Basis of Personality. Nature. 1963;199:1031–4. https://doi.org/10.1038/1991031a0.

    Article  CAS  Google Scholar 

  59. Eysenck SBG, Eysenck HJ, Barrett P. A revised version of the psychoticism scale. Personal Individ Differ. 1985;6:21–9. https://doi.org/10.1016/0191-8869(85)90026-1.

    Article  Google Scholar 

  60. Carver CS, White TL. Behavioral inhibition, behavioral activation, and affective responses to impending reward and punishment: The BIS/BAS Scales. J Pers Soc Psychol. 1994;67:319–33. https://doi.org/10.1037/0022-3514.67.2.319.

    Article  Google Scholar 

  61. Mardaga S, Hansenne M. Relationships between Cloninger’s biosocial model of personality and the behavioral inhibition/approach systems (BIS/BAS). Personal Individ Differ. 2007;42:715–22. https://doi.org/10.1016/j.paid.2006.08.013.

    Article  Google Scholar 

  62. Hartig J, Moosbrugger H. Die, “ARES-Skalen” zur Erfassung der individuellen BIS- und BAS-Sensitivität. Z Für Differ Diagn Psychol. 2003;24:293–310. https://doi.org/10.1024/0170-1789.24.4.293.

    Article  Google Scholar 

  63. Bacher J. Teststatistiken zur Bestimmung der Clusterzahl für Quick Cluster. ZA-Inf Zentralarchiv Für Empirische Sozialforschung. 2001;48:71–97.

  64. Fleiss JL, Levin B, Paik MC. Statistical methods for rates and proportions. 3rd ed. Hoboken, N.J: J. Wiley; 2003.

  65. Benjamini Y, Hochberg Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. J R Stat Soc Ser B Methodol. 1995;57:289–300. https://doi.org/10.1111/j.2517-6161.1995.tb02031.x.

    Article  Google Scholar 

  66. IBM SPSS Statistics for Macintosh 2020.

  67. Lieb R, Müller N. Epidemiologie und Komorbidität der Sozialen Phobie. Soz. Phobie Soz. Angststörung, Göttingen: Hogrefe; 2002, p. 34–65.

  68. Wittchen H-U, Fehm L. Epidemiology and natural course of social fears and social phobia: Epidemiology and course of SP. Acta Psychiatr Scand. 2003;108:4–18. https://doi.org/10.1034/j.1600-0447.108.s417.1.x.

    Article  Google Scholar 

  69. Wiltink J, Hoyer J, Beutel ME, Ruckes C, Herpertz S, Joraschky P, et al. Do Patient Characteristics Predict Outcome of Psychodynamic Psychotherapy for Social Anxiety Disorder? PLoS ONE. 2016;11:e0147165. https://doi.org/10.1371/journal.pone.0147165.

    Article  CAS  Google Scholar 

  70. Wells A, Clark DM, Salkovskis P, Ludgate J, Hackmann A, Gelder M. Social phobia: The role of in-situation safety behaviors in maintaining anxiety and negative beliefs. Behav Ther. 1995;26:153–61. https://doi.org/10.1016/S0005-7894(05)80088-7.

    Article  Google Scholar 

  71. Jakšić N, Aukst-Margetić B, Marčinko D, Brajković L, Lončar M, Jakovljević M. Temperament, character, and suicidality among Croatian war veterans with posttraumatic stress disorder. Psychiatr Danub. 2015;27:60–3.

    Google Scholar 

  72. Krysinska K, Heller TS, De Leo D. Suicide and deliberate self-harm in personality disorders. Curr Opin Psychiatry. 2006;19:95–101. https://doi.org/10.1097/01.yco.0000191498.69281.5e.

    Article  Google Scholar 

  73. Svrakic DM. Differential Diagnosis of Personality Disorders by the Seven-Factor Model of Temperament and Character. Arch Gen Psychiatry. 1993;50:991. https://doi.org/10.1001/archpsyc.1993.01820240075009.

    Article  CAS  Google Scholar 

  74. Crescentini C, Garzitto M, Paschetto A, Brambilla P, Fabbro F. Temperament and character effects on late adolescents’ well-being and emotional-behavioural difficulties. PeerJ. 2018;6:e4484. https://doi.org/10.7717/peerj.4484.

    Article  Google Scholar 

  75. Spittlehouse JK, Vierck E, Pearson JF, Joyce PR. Temperament and character as determinants of well-being. Compr Psychiatry. 2014;55:1679–87. https://doi.org/10.1016/j.comppsych.2014.06.011.

    Article  Google Scholar 

  76. Balestri M, Porcelli S, Souery D, Kasper S, Dikeos D, Ferentinos P, et al. Temperament and character influence on depression treatment outcome. J Affect Disord. 2019;252:464–74. https://doi.org/10.1016/j.jad.2019.04.031.

    Article  Google Scholar 

  77. Zaninotto L, Solmi M, Toffanin T, Veronese N, Cloninger CR, Correll CU. A meta-analysis of temperament and character dimensions in patients with mood disorders: Comparison to healthy controls and unaffected siblings. J Affect Disord. 2016;194:84–97. https://doi.org/10.1016/j.jad.2015.12.077.

    Article  Google Scholar 

  78. Terock J, Janowitz D, Spitzer C, Miertsch M, Freyberger HJ, Grabe HJ. Alexithymia and self-directedness as predictors of psychopathology and psychotherapeutic treatment outcome. Compr Psychiatry. 2015;62:34–41. https://doi.org/10.1016/j.comppsych.2015.06.007.

    Article  Google Scholar 

  79. Mullola S, Hintsanen M, Elovainio M, Pulkki-Råback L, Lipsanen J, Josefsson K, et al. Adulthood temperament and educational attainment: A population-based cohort study. Learn Instr. 2015;40:39–53. https://doi.org/10.1016/j.learninstruc.2015.07.002.

    Article  Google Scholar 

  80. Vilaplana-Pérez A, Pérez-Vigil A, Sidorchuk A, Brander G, Isomura K, Hesselmark E, et al. Much more than just shyness: the impact of social anxiety disorder on educational performance across the lifespan. Psychol Med. 2021;51:861–9. https://doi.org/10.1017/S0033291719003908.

    Article  Google Scholar 

  81. Brook CA, Willoughby T. The Social Ties That Bind: Social Anxiety and Academic Achievement Across the University Years. J Youth Adolesc. 2015;44:1139–52. https://doi.org/10.1007/s10964-015-0262-8.

    Article  Google Scholar 

  82. Ponniah K, Hollon SD. Empirically supported psychological interventions for social phobia in adults: a qualitative review of randomized controlled trials. Psychol Med. 2008;38:3–14. https://doi.org/10.1017/S0033291707000918.

    Article  CAS  Google Scholar 

  83. Domschke K, Schiele MA, Romanos M. Prävention von Angsterkrankungen. Nervenarzt. 2021;92:450–6. https://doi.org/10.1007/s00115-020-01045-1.

    Article  CAS  Google Scholar 

Download references

Acknowledgements

Not applicable.

Funding

Open Access funding enabled and organized by Projekt DEAL. The study was funded by the Else Kröner-Fresenius-Stiftung (2019_A127). The funding source was not involved in study design, in the collection, analysis and interpretation of data in the writing of the manuscript and in the decision to submit the article for publication.

Author information

Author notes

  1. Man-Long Chung and Laura-Effi Seib-Pfeifer shared first authorship.

Authors and Affiliations

  1. Department of Psychosomatic Medicine and Psychotherapy, University Hospital Bonn, Venusberg Campus 1, 53127, Bonn, Germany

    Man-Long Chung, Christina Elling, Franziska Geiser & Rupert Conrad

  2. Department of Psychology, University of Bonn, Kaiser-Karl-Ring 9, 53111, Bonn, Germany

    Laura-Effi Seib-Pfeifer

  3. Institute of Human Genetics, University of Bonn, School of Medicine & University Hospital Bonn, Venusberg Campus 1, 53127, Bonn, Germany

    Andreas J. Forstner

  4. Institute of Neuroscience and Medicine (INM-1), Research Centre Jülich, Wilhelm-Johnen-Straße, 52428, Jülich, Germany

    Andreas J. Forstner

  5. Centre for Human Genetics, University of Marburg, Baldingerstraße, 35033, Marburg, Germany

    Andreas J. Forstner & Johannes Schumacher

  6. Department of Psychosomatic Medicine and Psychotherapy, University Hospital Münster, Albert-Schweitzer-Campus 1, 48149, Münster, Germany

    Rupert Conrad

Authors
  1. Man-Long Chung
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Laura-Effi Seib-Pfeifer
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Christina Elling
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Franziska Geiser
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Andreas J. Forstner
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Johannes Schumacher
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Rupert Conrad
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

Man-Long Chung: Conceptualization, Data curation, Methodology, Validation, Formal analysis, Software, Investigation, Visualization, Writing – original draft, Writing – review & editing. Laura-Effi Seib-Pfeifer: Conceptualization, Data curation, Methodology, Validation, Formal analysis, Software, Investigation, Visualization, Writing – original draft, Writing – review & editing. Christina Elling: Data curation, Software, Formal analysis, Writing – review & editing. Franziska Geiser: Validation, Formal analysis, Supervision, Writing – review & editing. Andreas J. Forstner: Conceptualization, Funding acquisition, Validation, Formal analysis, Supervision, Writing – review & editing. Johannes Schumacher: Validation, Formal analysis, Supervision, Writing – review & editing. Rupert Conrad: Conceptualization, Funding acquisition, Methodology, Validation, Formal analysis, Software, Supervision, Writing – review & editing. The author(s) read and approved the final manuscript.

Corresponding author

Correspondence to Man-Long Chung.

Ethics declarations

Ethics approval and consent to participate

The study was approved by the local ethics committee of the University of Bonn and was in accordance with the declaration of Helsinki. All participants gave written informed consent.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1:

Table A. Comparison of the compared cluster solutions.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Chung, ML., Seib-Pfeifer, LE., Elling, C. et al. Personality subtypes in adults with social anxiety disorder - novelty seeking makes the difference. BMC Psychiatry 22, 832 (2022). https://doi.org/10.1186/s12888-022-04484-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12888-022-04484-z

Keywords

BMC Psychiatry

ISSN: 1471-244X

Contact us