Original Article
Temperature Perception in the Blushing Region
A Pilot Study Comparing Blushing-Fearful Individuals With and Without Social Phobia and Controls
Abstract
Abstract.Background: Individuals with social phobia are characterized by a heightened sensitivity for physical symptom such as blushing. Objectives: We exploratorily examined whether blushing-fearful individuals with / without social phobia were more sensitive to perceiving facial temperature increases compared to healthy controls. Methods: 50 participants were tested using an adaptive two-alternative forced-choice task. A Peltier element was attached to the cheek and repeatedly warmed up in one of two phases. The participants’ task was to indicate in which of the two phases the temperature had increased. Results: Contrary to our hypothesis, blushing-fearful individuals with social phobia (n = 11) were less sensitive to facial temperature increases, i. e., exhibited higher temperature perception threshold, than the subclinical group (n = 14) and healthy controls (n = 23). Limitations: The preliminary nature of our study should be noted, and replication in a larger sample is warranted. Conclusions: The findings may be best understood within a predictive coding framework: Individuals with social phobia may rely on priors when assessing their physical symptoms to form an impression of their public self in social situations.
Zusammenfassung.Hintergrund: Personen mit sozialer Phobie sollen sich durch eine erhöhte Sensibilität für körperliche Symptome wie etwa Erröten auszeichnen. Zielsetzung: Wir untersuchten, ob errötungsängstliche Personen mit und ohne Sozialphobie im Vergleich zu gesunden Kontrollpersonen Temperaturerhöhungen in der Errötungsregion sensitiver wahrnehmen. Methode: 50 Teilnehmer wurden mit einer adaptiven Zwei-Alternativen-Forced-Choice-Aufgabe getestet. Ein an der Wange befestigtes Peltier-Element wurde wiederholt zufällig in einer von zwei Phasen erwärmt. Die Teilnehmer sollten angeben, in welcher Phase der Temperaturanstieg stattgefunden hatte. Ergebnisse: Entgegen unserer Hypothese nahmen Personen mit Errötungsangst und Sozialphobie Temperaturerhöhungen im Gesicht weniger empfindlich wahr, d. h. sie wiesen eine höhere Temperaturwahrnehmungsschwelle auf als Personen mit subklinischer Errötungsangst und gesunde Kontrollpersonen. Limitationen: Der vorläufige Charakter unserer Studie sollte beachtet werden und die Replikation in einer größeren Stichprobe ist gerechtfertigt. Schlussfolgerungen: Unsere Ergebnisse lassen sich am besten im Rahmen eines predictive coding Modells verstehen: Personen mit sozialer Phobie verlassen sich bei der Bewertung ihrer körperlichen Symptome möglicherweise hauptsächlich auf Vorannahmen (priors), um sich in sozialen Situationen ein Bild von ihrem öffentlichen Selbst zu machen.
Blushing has repeatedly been acknowledged as a prominent physiological feature of social phobia (SP, also termed social anxiety; Nikolić et al., 2015). According to Leary and Toner (2012), blushing is defined as an uncontrollable experience of heat, usually accompanied by redness of the skin on the face, neck, ears, and upper chest. People sometimes experience this physiological response in connection with real or perceived evaluations, social attention, or self-insecurity. The blush is often accompanied by feelings of embarrassment, shame, guilt (Drummond, 2012), social anxiety, or fear when social norms are violated (Leary & Toner, 2012). Moreover, studies on childhood SP show that blushing in social situations is not only associated with but also contributes to the development of SP symptoms in childhood (Nikolić et al., 2016, 2020), indicating the etiological role of blushing in SP.
Indeed, self-perceived blushing is strongly related to SP (e. g., Leary & Meadows, 1991; Neto, 1996; for a meta-analysis see Nikolić et al., 2015). In contrast, physiological blushing has been reported to be only weakly related to SP. Even if individuals are formally diagnosed with SP, their physiological blushing may not be stronger than that of healthy controls. For example, in a study by Gerlach et al. (2001), individuals suffering from SP exhibited a stronger increase in facial blood flow in only one of three social situations, although they self-reported more blushing in all three situations. Similarly, only a group of individuals suffering from both SP and fear of blushing blushed physiologically more than healthy controls, whereas individuals with SP without fear of blushing did not (Voncken & Bögels, 2009). Again, both groups nonetheless self-reported substantially more blushing. These findings were previously interpreted in two ways: 1) Individuals with SP and fear of blushing may overestimate their blushing (i. e., in terms of an interpretation bias); 2) blushing-fearful individuals with SP may be able to perceive physiological blushing even with only slight increases in facial blood flow (i. e., exhibit increased interoceptive accuracy). In other words, individuals with SP might assume that they are blushing more often because of an elevated ability to detect even small changes in skin temperature as an indication of blushing.
Several theoretical models suggest a heightened sensitivity for bodily symptoms. For example, Gerlach (2005) postulated a vicious circle model of social anxiety, explicitly including potentially visible somatic symptoms of anxiety. This model suggests that the experience of anxiety leads to increased sympathetic activation, which in turn increases the occurrence of visible physical symptoms of anxiety (e. g., sweating, shaking, and blushing). With heightened internal attention, individuals with social anxiety are more prone to perceive these physical symptoms and may then interpret them catastrophically. This again increases anxiety and physiological arousal. Similarly, Drummond (2012) also argued that hypervigilance toward blushing results in a heightened perception of blushing. Both models are thematically linked to other vicious cycle models of anxiety, such as the psychophysiological model of panic attacks (Ehlers & Margraf, 1989). For example, panic patients often exhibit an enhanced ability to perceive their heart rate (Ehlers, 1993). Additionally, a more accurate heartbeat perception was found repeatedly not only among individuals with panic disorder (Domschke et al., 2010) but also in individuals suffering from generalized anxiety or social anxiety (Van der Does et al., 2000).
Previous research on SP indicates that individuals with SP usually focus their attention inwardly (e. g., Deiters et al., 2013; Lin et al., 2021). Arguably, such heightened self-focused attention is one of the psychological risk factors for fear of blushing (Drummond et al., 2020). Moreover, individuals with SP tend to interpret visible bodily symptoms such as blushing, sweating, or trembling as a threat (Gerlach et al., 2004). However, it has not yet been thoroughly explored whether individuals with SP better perceive their bodily processes (i. e., blushing). Some evidence for better interoceptive sensitivity (i. e., heartbeat perception) in individuals with SP was provided by Stevens et al. (2011). Similarly, Judah et al. (2018) found that individuals with SP exhibited increased processing of cardiac activity as measured by heartbeat-evoked potentials. In contrast, in a recent review of thermosensation and thermoregulation in anxiety disorders, the authors conclude that individuals with anxiety disorders do not manifest altered thermal sensitivity compared to healthy controls (Fischer et al., 2021). However, this conclusion is derived from four studies mainly investigating pain (e. g., pain perception, Defrin et al., 2008; cold pain thresholds, Mostoufi et al., 2014). In sum, it remains unclear whether individuals with SP perceive facial blushing more sensitively.
Schandry and Poth (1983) first measured blushing using physiological equipment. Cheek temperature was used as an index for blushing based on the notion that increased blood flow to the facial skin from blushing should increase skin temperature. Not surprisingly, a feeling of warmth is the physical sensation most often reported when blushing, and such changes in facial temperature are reported irrespective of skin complexion (Simon & Shields, 1996). Thus, should the vicious cycle of blushing and anxiety in SP include heightened interoceptive sensitivity for blushing, the physiological system that warrants examination is the ability to perceive temperature increases in the facial skin.
Therefore, the present study explores the sensitivity to perceive facial temperature increases among blushing-fearful individuals with SP and healthy controls. A novel paradigm using an adaptive two-alternative forced-choice task to measure the perception of facial temperature increases was tested, in which a Peltier element was attached to the cheek of participants and repeatedly warmed up in one of two phases. The participants’ task was to detect in which of the two phases the temperature increase had occurred. We hypothesized that blushing-fearful individuals with SP would be more sensitive to perceiving temperature increases in the face than individuals fearful of blushing without SP or healthy controls. We expected to validate the paradigm and to exploratorily examine the difference between clinical and subclinical individuals considering that sample type (i. e., blushing fearfulness and / or diagnosis of SP) was found to influence the relationship between self-perceived blushing, physiological blushing, and social anxiety according to a previous meta-analysis (Nikolić et al., 2015).
Method
Participants
Data from 50 participants were collected through convenience sampling. Information about the study was posted on various social media (e. g., Facebook) and local newspapers, specifically targeting individuals with a fear of blushing. 25 individuals who reported a fear of blushing were recruited, 11 of whom fulfilled the criteria for SP. In compensation for participating in the experiment, participants were offered a workshop on dealing with blushing anxiety. The control group was collected with a small delay in time (starting after the first five blushing-fearful participants had been recruited), allowing to match variables of gender, age, and education of the control group. Among the total 50 participants, there were 9 males and 41 females. The age ranged from 18 to 48 years, with an average of 29.96 (SD = 9.17).
Measurements
ICD-10 Checklist Social Phobia
This semistructured interview was developed by Hiller et al. (1995) to assess clinically significant SP and associated impairment according to the ICD-10 criteria. The severity of the symptoms is assessed in terms of a three-level grading: (1) no SP, (2) subclinical SP, or (3) diagnosis of SP (F40.1). Test-retest reliability for the diagnosis checklist for anxiety disorders is good (kappa = 0.76; Hiller et al., 1993).
Patient Health Questionnaire (PHQ-D)
The PHQ-D measures symptoms of various mental disorders and the degree of associated impairment (Löwe et al., 2002). The present study used an extended version of PHQ-D to assess the presence (yes / no), the degree of impairment (not / little / severely impaired), or the frequency of impairment (not at all / on individual days / on more than half of the days / almost every day) of depression, somatic symptoms, and social anxiety. In the current sample, the internal consistency was satisfactory to very high, with Cronbach’s α = .71 for the depression subscale, α = .73 for the somatic symptoms’ subscale, and α = .92 for the social anxiety subscale.
Brief Symptom Inventory 18 (BSI-18)
The BSI-18 (Derogatis, 2000) is an 18-item self-reported screening inventory designed to assess participants’ level of psychological distress on three subscales: somatization, depression, and anxiety, with 6 items each. Participants rated how much they had been “distressed or bothered” in the past 7 days (including the current day) by any given symptom, on a 5-point Likert-type scale ranging from 0 (not at all) to 4 (extremely). The global severity index of distress represents the sum across the three subscales. In the current sample, the internal consistency of the global severity index is satisfactory, with Cronbach’s α = .78.
Blushing Propensity Scale (BPS)
The BPS (Leary & Meadows, 1991) is a self-report scale to measure the tendency for blushing in everyday situations. Participants were asked to indicate on a scale of 1 (never) to 5 (always) how often they had the impression of blushing in 14 social situations (“… when I am introduced to someone”; “… if I act stupidly or incompetently in front of others”). In the current sample, the internal consistency was very high, with Cronbach’s α = .96.
Fear of Blushing Questionnaire (FEA)
The FEA assesses self-reported perceived blushing and the fear of it (Härtling et al., 2013). It includes seven subscales: (1) “fear of blushing,” which measures the extent of blushing and how far it is experienced as problematic (e. g., “How anxious are you to turn red”) with 6 items using Likert-type scale from 0 (not at all / never) to 10 (very much / always); (2) “behavioral problems,” which is rated from 0 (never) to 4 (always) and captures the impairment in the last seven days by six possible blushing-related reactions (e. g., “black-outs”); (3) “positive cognitions” (10 items); (4) “negative cognitions” (52 items), which relate to thoughts and ideas a person may have regarding blushing (“Everyone gets it when I blush,” “Blushing is adorable”), with each item to be estimated from 0 (I don’t believe) to 10 (I believe) to indicate the extent the thought is believed; (5) “safety behavior,” which queries the frequency of 10 strategies used in the last 7 days to hide blushing (e. g., “wear makeup”) on a scale from 0 (never) to 4 (always); (6) “body reactions,” which includes six items for experiencing body reactions in connection with blushes (”face felt hot”) on a scale of 0 (never) to 4 (always); (7) “frequency,” which consists of an open item and requests the estimated number of blush reactions in the last week including the current day. In the current sample, the internal consistency for the first six scales was Cronbach’s α = .95, .88, .89, .98, .80, and .94, respectively.
Paradigm for Thermal Sensitivity
A 4 cm x 4 cm Peltier element, which could be heated and cooled, was fixed with an adhesive strip to the right cheek of the participants. A thermal probe was attached on the left cheek which measured the temperature of the facial skin approximately at the same location opposite to where the Peltier element was placed. The skin temperature at the beginning of the experiment served as reference temperature, on which the initial temperature of the Peltier element was based.
The paradigm included 27 trials (one test trial and 26 regular trials), each with two 7-second-long phases. The beginning and end of the two phases were indicated both on the screen with the words “Phase 1” and “Phase 2” and three flanking tones. The Peltier element warmed up randomly in one of these two phases. The task of the participant was to indicate in which of the two phases the temperature increase had occurred by pressing a button labeled “Phase 1” or “Phase 2” after each trial. The next trial started after the response with a cooling phase, in which the temperature of the Peltier element was cooled down to the reference temperature. (Note that it is important to not cool down the Peltier element between the two test phases because the skin is much better at detecting decreases in temperature than increases given that cold receptors are much more numerous than warm receptors by a ratio of up to 30:1 [Jones & Berris, 2002].)
The target temperature to which the Peltier element warmed up in the respective trial was determined by calculating the sum of the reference temperature plus a variable value: In the first trial, the Peltier element warmed up to the reference temperature plus 2° centigrade (C). This large difference was selected to allow the subject to get a feeling for warming and to easily discover it. From the second trial on, the difference of reference and target temperature was decreased in case of a correct answer and increased in case of an incorrect response.
In the second trial, the Peltier element warmed up to the reference temperature plus 1 °C (target temperature). If the participant chose the correct phase, the target temperature was reduced by 0.05℃ from the next trial on (reference plus 0.95℃, reference plus 0.90℃, reference plus 0.85℃). As from a difference of 0.4℃ between the reference and target temperature, the target temperature was reduced by 0.025℃.
If the participant answered incorrectly, the target temperature rose by three levels for the next trial, and then dropped again by one level if correctly responded; if incorrectly responded, it rose again by three levels. Thus, as the experiment progressed, it became increasingly difficult to detect in which of the phases heating had occurred. This simple up-down adaptive procedure converges at a stimulus intensity that indicates the individual perception threshold (Levitt, 1971). To exclude external temperature as a possible confounding variable, the room temperature was recorded at the beginning of the experiment.
Procedure
The experiment took place between October and December 2014. Participants were tested individually. Upon arrival, they gave informed consent and were interviewed using the ICD-10 Social Phobia Checklist conducted by one of the experimenters. Experimenters were trained to use the Checklist and supervised by ALG, who has extensive experience in diagnosing and treating individuals with SP. Subsequently, a set of questionnaires including demographical data, PHQ-D, BSI-18, BPS, and FEA were filled in.
Afterward, the Peltier element was attached to the right cheek and a thermal probe to the left cheek of the participant. Participants then read the instruction on the computer and confirmed it by pressing a button to start the test trial. Questions were answered if there were any. The participants were then asked to stay relaxed for 1 minute before the experiment to examine thermal sensibility started with 26 trials, each with two phases. The participants’ task was to decide in each trial in which phase the temperature increased. The duration of the entire experiment was about 20 to 25 minutes.
The experiment ended with debriefing the participants and offering compensation for participation in the form of either joining a 3-hour workshop on how to deal with blushing, credits for courses, or €8, depending on the individual choice of the participant.
Data Analysis
We performed data analysis using SPSS version 24. MANOVA and utilized chi-square tests to determine group differences regarding demographic and clinical characteristics. Note that we split the two fearful groups posthoc after realizing that only 11 blushing-fearful participants fulfilled the criteria for social phobia. The results of the temperature sensitivity test were analyzed using MANOVA. All statistical tests were conducted with an alpha level of p = .05. Effect sizes are reported using partial eta squares (ηp2) and Cohen’s d.
Results
Description of the Sample
The participants were classified into three groups according to the results of the ICD-10 Checklist Social Phobia as described above: (1) control group (no self-reported fear of blushing and no diagnosis of SP); (2) subclinical group (self-reported fear of blushing but no diagnosis of SP; (3) clinical group (self-reported fear of blushing and diagnosis of SP). Two participants from the control group did not fill out some of the questionnaires and were therefore excluded. Table 1 presents the demographic and clinical characteristics of the sample. A one-way MANOVA revealed that the three groups differed significantly in terms of all subscales of fear of blushing, blushing propensity, level of depression, and level of social anxiety.
Experimental Results
One participant in the subclinical group had to be excluded from the analysis for the temperature perception task because of a technical malfunction. The average reference temperature of the skin as measured on the left cheek was 32.95℃ (SD = 1.44). The three groups did not differ significantly concerning their initial skin temperature, F(2, 46) = 0.38, p = .69, ηp2 = .016. The room temperature was 21.52℃ (SD = 1.36) on average. No significant group differences were found for the room temperature, F(2, 46) = 1.27, p = .29, ηp2 = .052.
Visual inspection of the graph depicted in Figure 1 showed that a plateau was reached in approximately the last quintile of the trials. Therefore, a mean of the last eight temperature differences was calculated for each subject through trials 20 to 27, which was used as an index for the individual threshold to reliably detect temperature increases. The internal consistency of this measure was very high with Cronbach’s α = .97. Figure 1 depicts the temperature differences per trial distinguishing between the three groups.
Within the 27 trials, participants responded incorrectly to 2.20 (SD = 1.35) trials on average. The average numbers of incorrectly responded trials were 1.84 (SD = 1.28) for the control group, 2.00 (SD = 1.29) for the subclinical group, and 3.27 (SD = 1.10) for the SP group. On average, the threshold to detect temperature increases across all participants was 0.27℃ (SD = 0.086). The average thresholds for the control group, subclinical group, and SP group were 0.25℃ (SD = 0.074), 0.25℃ (SD = 0.079), and 0.34℃ (SD = 0.087), respectively. A MANOVA revealed a significant difference among the three groups, F(4,92) = 2.96, p = .024, ηp2 = .11. Both the number of incorrectly responded trials and the threshold differed significantly among the three groups, F(2,46) = 5.28, p = .009, ηp2 = .19, and F(2,46) = 6.14, p = .004, ηp2 = .211, respectively. According to posthoc tests, the number of incorrectly responded trials in the SP group was significantly higher than in the subclinical group (MD = 1.27, p = .049, d = 1.06) and control group (MD = 1.43, p = .008, d = 1.20), whereas there was no difference between the subclinical group and the control group (MD = .16, p = 1.00, d = 0.12). For the threshold to detect facial temperature increases, the SP group was significantly higher than in the subclinical group (MD = .095, p = .015, d = 1.08) and control group (MD = .094, p = .006, d = 1.14). The subclinical and control groups did not differ (MD = .001, p = 1.00, d = 0.013). Therefore, the SP group was less sensitive to temperature changes than the subclinical and control groups1.
Correlations
Bivariate correlation analysis showed a significant correlation between the FEA’s threshold and score of “behavioral problems” subscale (r = .30, p = .037), indicating that a higher tendency of blushing-related impairment is associated with less sensitivity to perceiving facial temperature changes. We found no significant correlations between the threshold and scores of other subscales of FEA, PHQ, BPS, or BSI (all r < .20, p > .17).
Discussion
The main aim of the present study was to exploratorily examine whether blushing-fearful individuals with and without SP were more sensitive to perceiving temperature changes on the face compared to healthy controls. To this end, we tested a novel paradigm for thermal sensitivity. Models of SP predict that individuals suffering from social anxiety or SP pay more attention to ongoing internal processes and thus are better able to detect physical changes (e. g., Drummond, 2012; Gerlach, 2005). In contrast to this hypothesis, the threshold of perceiving temperature differences was higher among the blushing-fearful group with SP compared to the subclinical and control groups. In other words, individuals with SP and additional blushing concerns were the least able to perceive facial temperature changes. Interestingly, our preliminary finding suggests that the clinical severity of SP might play a crucial role here: The effect was found only in the SP group but not the subclinical group, despite both groups being concerned about blushing. This latter result follows findings regarding the intensity of blushing as highlighted in the Introduction (Nikolić et al., 2015).
The results of our study indicate that the diagnosis of SP might be accompanied by a qualitative change regarding temperature perception (i. e., a decrease in the ability to accurately perceive physiological blushing), although contradicting our original hypothesis. In addition, results of the correlational analyses only partly support this finding: The score of the “behavioral problems” subscale of the FEA was significantly associated with the threshold, while no significant correlations were found between the threshold and scores of other subscales of FEA, PHQ, BPS, or BSI. We also acknowledge that our results are based on a relatively small sample, and that the robustness of the findings is thus not well validated. Therefore, interpretation should only be made with caution. However, if replicated in larger samples, our findings can be well understood in the context of predictive coding theory. According to this framework, the perception of body sensations is a constructive and hypothesis-driven process (Barrett & Simmons, 2015). The prediction of the presence or intensity of a bodily sensation is compared to the somatosensory input using the resulting prediction error to create the perception. The perception corresponds more to the somatosensory input (i. e., the perceived temperature) if this input is distinct. However, it corresponds more to the prediction if (a) the signal is vague, (b) the probability for a certain sensation is especially likely, or (c) if the prediction itself is especially important and thus processed with priority (e. g., Van den Bergh et al., 2017). Because of the relative vagueness of interoceptive stimuli, people are often not sensitive in describing bodily states and symptoms (Petersen et al., 2015). This is also true for the perception of blushing (e. g., Voncken & Bögels, 2009). Barrett and Simmons (2015) argued that interoceptive perception generally depends more on prior expectations than on sensory input. Stimuli predicting threatening outcomes are given even more weight (Paulus & Stein, 2006) because the system can react more adaptively when detecting threat (for an overview of negative affectivity on interoceptive processing, see also Van den Bergh et al., 2017).
In the eyes of an individual suffering from SP, blushing signals a threat as it increases the likelihood of negative evaluations. Not surprisingly, the mere belief of having increased arousal leads to poorer performance and overestimation of the visibility of anxiety in SP (Wild et al., 2008). Studies with false feedback of blushing confirmed that the belief that one is blushing leads to even more feelings of blushing and higher anxiety (Dijk et al., 2009; Drummond et al., 2003). In sum, the sensory experience of blushing is rather vague. Individuals diagnosed with SP are more likely to expect blushing and are thus threatened by this prospect. Taking this theoretical conceptualization and our results together, it seems entirely possible that, in the course of SP, the perception of blushing becomes gradually more dependent on predictions (i. e., priors) and is thus increasingly decoupled from sensory input, especially if individuals are less able to perceive their blushing adequately. In line with this, one previous study provided preliminary evidence of decreased accuracy of blush estimation among individuals with SP: Self-perceived blushing was found to be positively related to physiological blushing (cheek blood flow) in healthy controls but not in individuals suffering from SP (Voncken & Bögels, 2009).
It is worth noting that we used a novel paradigm to measure thermal sensitivity. To the best of our knowledge, this is the first study to directly measure the threshold of perceiving facial temperature increases of individuals with SP and comparison groups. Previous studies investigating thermal sensitivity usually measured pain perception or cold pain thresholds (see Fischer et al., 2021). Thus our paradigm expands the research domain of thermal sensitivity. However, some modifications may be beneficial for future replications. For example, termination of the task after a fixed number of reversals instead of a fixed number of trials might further increase the reliability of the task (Levitt, 1971). Further, adding more time points for facial temperature measurement during the task might be necessary. In our study, the facial temperature was measured only at the beginning to determine the necessary initial temperature for the Peltier element. It is unclear whether the actual facial temperature of the three groups remained the same during the task. Note, however, that both the room temperature and the initial facial temperature were the same in all three groups. Therefore, it can reasonably be assumed that the actual facial temperatures of the three groups remained the same, given that physiological blushing has been reported to be only weakly associated with the severity of SP (Fischer et al., 2021; Nikolić et al., 2015) and only in situations designed to induce blushing (Gerlach et al. 2001).
If SP is indeed associated with a decreased sensitivity to perceive blushing, our results may have important clinical implications. Therapists could explicitly address this in psychoeducation and cognitive restructuring of dysfunctional beliefs about blushing. Our findings also support treatments that aim at improving the accuracy of perception of symptoms that are not entirely physiologically explained (Schaefer et al., 2014; Meyerholz et al., 2019). This approach might be extended by improving therapeutically how individuals with SP form a perception of themselves. For example, video feedback helps to develop a more realistic impression of how they appear to others and to correct distorted negative self-images (Warnock-Parkes et al., 2017). Alternatively, task concentration training, in which participants are trained to shift attention away from physical symptoms to the actual social task, has been proven effective (e. g., Drummond et al., 2020; Härtling et al., 2016).
Several limitations of this study should be noted. As mentioned above, one major limitation lies in the small sample size and uneven distribution of the sample, although we were able to detect a significant difference. Note that our study is exploratory in nature, so it is necessary to replicate these findings in a larger and independent sample. Another limitation is that the perception of facial temperature change was not examined in an anxious state. Some studies have compared the perception of physiological processes both in anxious and nonanxious states (e. g., before and during speech anticipation, see Stevens et al., 2011). It is worthwhile to examine whether the lesser sensitivity to perceiving blushing is a trait-like disposition and how it might be influenced by situational anxiety. In addition, the interview to assess the severity of SP symptoms, ICD-10 Checklist Social Phobia, was not double-coded. However, the measurement by itself has good reliability according to the original literature (Hiller et al., 1993).
In conclusion, our study found that individuals with SP were less sensitive to perceiving facial temperature increases compared to subclinical individuals with a fear of blushing and healthy controls, indicating that they may rely on other sources of information (e. g., inner beliefs and predictions) to form an impression of themselves in social situations. If our findings can be replicated in larger samples, it may have important implications for clinical practice.
Electronic Supplementary Material
The electronic supplementary material is available with the online version of the article at https://doi.org/10.1026/1616-3443/a000682
Literatur
(2015). Interoceptive predictions in the brain. Nature Reviews Neuroscience, 16 (7), 419 – 429. https://doi.org/10.1038/nrn3950
(2008). Quantitative testing of pain perception in subjects with PTSD: Implications for the mechanism of the coexistence between PTSD and chronic pain. Pain, 138 (2), 450 – 459. https://doi.org/10.1016/j.pain.2008.05.006
(2013). Internal and external attention in speech anxiety. Journal of Behavior Therapy and Experimental Psychiatry, 44 (2), 143 – 149. https://doi.org/10.1016/j.jbtep.2012.09.001
(2000). BSI-18: Brief Symptom Inventory 18: Administration, scoring, and procedures manual. Pearson.
(2009). I blush, therefore I will be judged negatively: Influence of false blush feedback on anticipated others’ judgments and facial coloration in high and low blushing-fearfuls. Behaviour Research and Therapy, 47 (7), 541 – 547. https://doi.org/10.1016/j.brat.2009.03.005
(2010). Interoceptive sensitivity in anxiety and anxiety disorders: An overview and integration of neurobiological findings. Clinical Psychology Review, 30 (1), 1 – 11. https://doi.org/10.1016/j.cpr.2009.08.008
(2012).
Psychophysiology of the blush . In W. R. CrozierP. J. de Jong (Eds.), The psychological significance of the blush (pp. 15 – 38). Cambridge University Press. https://doi.org/10.1017/CBO9781139012850.004(2003). The impact of verbal feedback about blushing on social discomfort and facial blood flow during embarrassing tasks. Behaviour Research and Therapy, 41 (4), 413 – 425. https://doi.org/10.1016/S0005-7967(02)00021-9
(2020). Treatment options for fear of blushing. Current Psychiatry Reports, 22 (6), 28 https://doi.org/10.1007/s11920-020-01152-5
(1993). Interoception and panic disorder. Advances in Behaviour Research and Therapy, 15 (1), 3 – 21. https://doi.org/10.1016/0146-6402(93)90001-I
(1989).
The psychophysiological model of panic attacks . In P. M. G. EmmelkampW. EveraerdW. KraaimaatM. J. M. van Son (Eds.), Fresh perspectives on anxiety disorders (pp. 1 – 29). Swets & Zeitlinger.(2021). A systematic review of thermosensation and thermoregulation in anxiety disorders. Frontiers in Physiology, 12, 784943. https://doi.org/10.3389/fphys.2021.784943
(2005).
Angst vor körperlichen Symptomen im Rahmen der Sozialen Angststörung: Ursachen und Behandlung [Fear of physical symptoms in the context of social anxiety disorder: causes and treatment ]. In N. VriendsJ. Margraf (Eds.), Soziale Kompetenz, soziale Unsicherheit, soziale Phobie. Verstehen und Verändern [Social competence, social insecurity, social phobia. Understanding and changing ] (pp. 131 – 142). Schneider Hohengehren.(2004). Public and private heart rate feedback in social phobia: A manipulation of anxiety visibility. Cognitive Behaviour Therapy, 33 (1), 36 – 45. https://doi.org/10.1080/16506070310014682
(2001). Blushing and physiological arousability in social phobia. Journal of Abnormal Psychology, 110 (2), 247 – 258.
(2013). Psychometrische Eigenschaften des Fragebogens zur Errötungsangst (FEA) [Psychometric properties of the Fear of Blushing Questionnaire (FEA)]. Diagnostica, 59 (3), 142 – 153. https://doi.org/10.1026/0012-1924/a000090
(2016). Cognitive therapy and task concentration training applied as intensified group therapies for social anxiety disorder with fear of blushing: A randomized controlled trial. Clinical Psychology & Psychotherapy, 23 (6), 509 – 522. https://doi.org/10.1002/cpp.1975
(1995). IDCL Internationale Diagnosen Checkliste für ICD-10 und DSM-IV (Testbox). [IDCL International Diagnostic Checklists for ICD-10 and DSM-IV]. Huber.
(1993). Routine psychiatric examinations guided by ICD-10 Diagnostic Checklists (International Diagnostic Checklists). European Archives of Psychiatry and Clinical Neuroscience, 242 (4)., 218 – 223. https://doi.org/10.1007/bf02189966
(2002). The psychophysics of temperature perception and thermal-interface design. Proceedings 10th Symposium on Haptic Interfaces for Virtual Environment and Teleoperator Systems, 137 – 142. https://doi.org/10.1109/Haptic.2002.998951
(2018). The relationship between social anxiety and heartbeat evoked potential amplitude. Biological Psychology, 139, 1 – 7. https://doi.org/10.1016/j.biopsycho.2018.09.013
(1991). Predictors, elicitors, and concomitants of social blushing. Journal of Personality and Social Psychology, 60 (2), 254 – 262. https://doi.org/10.1037/0022-3514.60.2.254
(2012).
Psychological theories of blushing . In W. R. CrozierP. J. de Jong (Eds.), The psychological significance of the blush (pp. 63 – 76). Cambridge University Press. https://doi.org/10.1017/CBO9781139012850.007(1971). Transformed up-down methods in psychoacoustics. The Journal of the Acoustical Society of America, 49(2B), 467 – 477. https://doi.org/10.1121/1.1912375
(2021). Self-focused attention vs. negative attentional bias during public speech task in socially anxious individuals. Behaviour Research and Therapy, 136, 103766. https://doi.org/10.1016/j.brat.2020.103766
(2002). PHQ-D. Gesundheitsfragebogen für Patienten (2nd ed.) [PHQ-D Patient Health Questionnaire]. Pfizer.
(2019). Contingent biofeedback outperforms other methods to enhance the accuracy of cardiac interoception: A comparison of short interventions. Journal of Behavior Therapy and Experimental Psychiatry, 63, 12 – 20. https://doi.org/10.1016/j.jbtep.2018.12.002
(2014). Pain sensitivity in posttraumatic stress disorder and other anxiety disorders: A preliminary case control study. Annals of General Psychiatry, 13 (1), 31 https://doi.org/10.1186/s12991-014-0031-1
(1996). Correlates of social blushing. Personality and Individual Differences, 20 (3), 365 – 373. https://doi.org/10.1016/0191-8869(95)00180-8
(2015). Blushing and social anxiety: A meta-analysis. Clinical Psychology: Science and Practice, 22 (2), 177 – 193. https://doi.org/10.1111/cpsp.12102
(2016). Autonomic arousal in children of parents with and without social anxiety disorder: A high-risk study. Journal of Child Psychology and Psychiatry, 57 (9), 1047 – 1055. https://doi.org/10.1111/jcpp.12563
(2020). The unique contribution of blushing to the development of social anxiety disorder symptoms: Results from a longitudinal study. Journal of Child Psychology and Psychiatry, 61 (12), 1339 – 1348. https://doi.org/10.1111/jcpp.13221
(2006). An insular view of anxiety. Biological Psychiatry, 60 (4), 383 – 387. https://doi.org/10.1016/j.biopsych.2006.03.042
(2015). Interoception and the uneasiness of the mind: Affect as perceptual style. Frontiers in Psychology, 6, 1408. https://doi.org/10.3389/fpsyg.2015.01408
(2014). Improving heartbeat perception in patients with medically unexplained symptoms reduces symptom distress. Biological Psychology, 101, 69 – 76. https://doi.org/10.1016/j.biopsycho.2014.05.012
(1983). Eine experimentelle Untersuchung zur Psychologie und Physiology des Errötens [An experimental study on the psychology and physiology of blushing]. Psychologische Beiträge, 25, 503 – 514.
(1996). Does complexion color affect the experience of blushing? Journal of Social Behavior and Personality, 11 (1), 177 – 188.
(2011). Heartbeat perception in social anxiety before and during speech anticipation. Behaviour Research and Therapy, 49 (2), 138 – 143. https://doi.org/10.1016/j.brat.2010.11.009
(2017). Symptoms and the body: Taking the inferential leap. Neuroscience & Biobehavioral Reviews, 74(Pt A), 185 – 203. https://doi.org/10.1016/j.neubiorev.2017.01.015
(2000). Heartbeat perception in panic disorder: A reanalysis. Behaviour Research and Therapy, 38 (1), 47 – 62. https://doi.org/10.1016/s0005-7967(98)00184-3
(2009). Physiological blushing in social anxiety disorder patients with and without blushing complaints: Two subtypes? Biological Psychology, 81 (2), 86 – 94. https://doi.org/10.1016/j.biopsycho.2009.02.004
(2017). Seeing is believing: Using video feedback in cognitive therapy for social anxiety disorder. Cognitive and Behavioral Practice, 24 (2), 245 – 255. https://doi.org/10.1016/j.cbpra.2016.03.007
(2008). Perception of arousal in social anxiety: Effects of false feedback during a social interaction. Journal of Behavior Therapy and Experimental Psychiatry, 39 (2), 102 – 116. https://doi.org/10.1016/j.jbtep.2006.11.003
1Similar results were found using only the last trial, instead of the mean of the last eight trials, as dependent variable. The results of this additional analysis are provided in the Electronic Supplementary Material (ESM 1).
