Social Anxiety in Conversations: Self-Focus, Silence, and Fear of Judgment
Author: Ukrainian Psychological Hub · Published: October 1, 2026 · Editorial Policy
A conversation can become strangely difficult when attention turns inward. Instead of simply listening and responding, a person may start monitoring every word, facial expression, pause, gesture, and sign of anxiety while simultaneously trying to predict what the other person thinks. The result can be a familiar social-anxiety pattern: the mind seems to go blank, speech becomes shorter or quieter, silence feels dangerous, and ordinary uncertainty is interpreted as evidence of judgment.
These experiences are psychologically real, but they do not by themselves establish a diagnosis. Social Anxiety Disorder is a clinical condition defined by a broader pattern of persistent fear or anxiety about social scrutiny, avoidance or endurance with marked distress, and meaningful impairment. The National Institute of Mental Health describes fear of being judged, difficulty talking to people, speaking very softly, and a sense that the mind has gone blank among common manifestations of the disorder. Similar experiences can also occur in people who do not meet diagnostic requirements.
This article focuses narrowly on what happens inside conversation: self-focused attention, fear of negative evaluation, silence or reduced talking, conversational safety behaviors, anticipatory worry, and the replay that can follow afterward. It does not treat every awkward pause as pathology, and it does not assume that conversational difficulty is caused by anxiety whenever anxiety is present.
Social anxiety in conversations: the short answer
Social anxiety can make conversation feel less like an exchange and more like a live evaluation. Cognitive-behavioral models propose that when a person expects negative evaluation, attention can shift toward internal threat cues: “How do I sound?”, “Am I blushing?”, “Was that stupid?”, “What should I say next?”, “Why are they quiet?”, “Do I look nervous?” The person then has less access to the actual conversation and more exposure to an internally generated picture of how badly the interaction may be going. This mechanism is central to influential models of social anxiety, including the model proposed by Rapee and Heimberg and the Clark and Wells model incorporated into NICE clinical guidance.
Conversation can then narrow behavior. A person may speak less, rehearse sentences before saying them, ask questions mainly to keep attention off themselves, avoid disagreement, speak quietly, hide visible anxiety, leave early, overuse humor, check the other person's face for reassurance, or mentally review each sentence while the interaction is still happening. These strategies can reduce anxiety for a moment while also making it harder to learn that an imperfect, spontaneous conversation can be tolerated.
The important distinction is between the feeling of conversational failure and what actually happened. Social anxiety can bias self-evaluation, yet research also shows that some safety behaviors can genuinely alter social performance. The evidence therefore supports a more useful question than “Was I objectively good or bad at talking?”: which processes were operating, what did they make harder, and what happened when those processes changed?
Why self-focused attention changes a conversation
Self-focused attention means attention is disproportionately directed toward one's own body, thoughts, feelings, appearance, voice, or imagined social image. In ordinary life, some self-monitoring is useful. It helps people notice when they are interrupting, choose words carefully, read the room, and repair misunderstandings. Social anxiety becomes more difficult when monitoring becomes threat-driven, continuous, and expensive.
A classic review by Spurr and Stopa concluded that self-focused attention is associated with social anxiety and may increase access to negative self-information while reducing processing of external information that could contradict feared predictions. A later systematic review by Gregory and Peters found that self-related constructs, including self-focused attention, are central in cognitive models of Social Anxiety Disorder and are modified during cognitive behavioral therapy.
The conversation becomes two tasks at once
Imagine answering a simple question while also checking your voice, face, hands, posture, eye contact, wording, timing, and the other person's reaction. That is not one task. It is a conversation plus an internal surveillance task. The second task can become so dominant that listening, memory for what was just said, spontaneity, and access to one's own interests all become more difficult.
This does not mean that social anxiety produces a universal cognitive deficit. Laboratory findings are heterogeneous, and many studies are small or task-specific. The better-supported claim is narrower: in socially anxious states, self-monitoring and threat processing can compete with the external information needed for fluid interaction. That interpretation is consistent with experimental conversation studies and with the treatment mechanisms emphasized by NICE.
The internal camera can feel more convincing than the external interaction
People with social anxiety often form an observer-like image of themselves: a picture of looking rigid, boring, awkward, red-faced, shaky, or visibly anxious. Because that image is generated from internal sensations and expectations, it can feel like direct evidence of how others see them. It is not direct evidence.
The distinction matters clinically. Socially anxious people may sometimes be judged as more nervous, and in some tasks there can be observable performance differences. Yet the other person does not have access to the full intensity of the internal experience. Research on interpersonal perception has repeatedly found a gap between how negatively socially anxious people believe they are seen and how others actually evaluate them. The gap is not identical in every study or situation, which is one reason careful behavioral experiments are more informative than reassurance.
Fear of judgment is the organizing threat
Conversation anxiety is often less about words themselves than about what words might reveal. A person may fear appearing unintelligent, boring, needy, strange, inexperienced, too quiet, too talkative, unattractive, incompetent, anxious, rude, or socially out of place. The feared outcome can be rejection, ridicule, visible embarrassment, loss of status, or simply the sense that another person has formed a negative opinion.
NICE recommends that assessment of possible Social Anxiety Disorder examine the feared social situations, what the person thinks might happen, self-image, safety-seeking behavior, focus of attention, and anticipatory and post-event processing. That list is useful because “fear of judgment” is not one thought. It is a network of predictions about what will be noticed, what it will mean, and what another person will do next.
The social context also matters. Human beings do evaluate one another, and some environments contain real hostility, bullying, prejudice, exploitation, or large power differences. Evidence-based work with social anxiety is not a demand to assume that every social setting is safe or that every judgment fear is irrational. The useful distinction is between actual contextual risk and a threat estimate that expands beyond the available evidence.
Why your mind can go blank
“My mind goes blank” is one of the most searched and most recognizable conversation complaints in social anxiety. NIMH explicitly includes this experience among common symptoms that can occur when a person with Social Anxiety Disorder is around other people or performing in front of them. The phrase describes a subjective experience, not a specific neurological mechanism.
Several processes can converge. Attention may be occupied by monitoring bodily sensations and appearance. The person may be pre-editing possible replies and rejecting them before speaking. Threat predictions may trigger a search for the “perfect” response. At the same time, less attention is available for the other person's last sentence, the surrounding context, and associative ideas that normally make conversation feel spontaneous.
Conversation research supports parts of this account. Experimental work by McManus, Sacadura, and Clark found that experimentally increasing self-focus and safety behaviors was unhelpful in socially anxious participants. In adolescents, Leigh, Chiu, and Clark experimentally compared conversations conducted with self-focus and safety behaviors versus external focus without those behaviors; the self-focused condition increased anxiety and undermined performance ratings. These studies support a causal role for the process, while their specific samples and laboratory tasks limit how literally the results should be applied to every real conversation.
Blanking can also have many other explanations. Fatigue, sleep deprivation, depression, trauma responses, substance effects, medication effects, language demands, hearing difficulty, attention problems, autism-related overload, speech or language conditions, and ordinary uncertainty can all affect verbal access. If blanking is new, pervasive, accompanied by neurological symptoms, or occurs well beyond social-evaluative situations, it should not be automatically explained by social anxiety.
Silence, short answers, and reduced talking
Silence is not a symptom with one meaning. A pause can mean thoughtfulness, comfort, fatigue, uncertainty, cultural style, loss of interest, processing time, emotional regulation, or nothing in particular. In social anxiety, however, silence can become both an outcome of anxiety and a feared event in its own right.
One conversation study of people with Social Anxiety Disorder by Stevens and colleagues found that talking time was a strong predictor of observer-rated social performance, and group differences in performance disappeared when talking time was taken into account. The authors proposed that reduced talking time may function as a safety behavior or as a downstream behavioral result of several anxiety-related processes. This does not mean that quiet people are socially deficient. It means that in some anxious interactions, speaking less can materially change how the conversation unfolds.
A common loop is easy to recognize: fear of saying something wrong leads to waiting for certainty; waiting creates a longer pause; the pause is interpreted as proof of awkwardness; anxiety rises; the threshold for speaking becomes even higher. Another loop goes in the opposite direction: silence feels intolerable, so the person rushes to fill every gap, asks question after question, or says things they do not mean simply to prevent a pause. Both patterns treat silence as danger.
The therapeutic target is therefore not “never be quiet.” It is flexibility. A person who can tolerate a pause has more options: think, listen, change subject, ask a genuine question, say they lost their train of thought, or let the interaction end naturally. A person who experiences every pause as a social emergency has fewer options.
Conversational safety behaviors
Safety behaviors are actions used to prevent a feared social catastrophe or conceal anxiety. A major review by Piccirillo, Dryman, and Heimberg concluded that these behaviors are important in the maintenance of social anxiety because they can block opportunities to learn that feared outcomes are less likely, less visible, or more manageable than expected.
In conversation, safety behaviors can be obvious: avoiding eye contact, staying near an exit, letting someone else speak for you, leaving early, avoiding names or personal details, or not initiating. They can also be nearly invisible: rehearsing every sentence internally, monitoring grammar, keeping answers deliberately short, smiling continuously to signal friendliness, speaking only when certain, mentally scripting the next three turns, agreeing with everything, asking questions solely to redirect attention, checking a phone to escape a pause, or scanning the other person's face for signs of disapproval.
A behavior is not automatically a safety behavior simply because it appears on a list. Asking questions, preparing for an interview, using notes, choosing where to sit, or taking a break can be sensible. The functional question is why the behavior is being used and what the person believes would happen without it. If the rule is “I must do this or they will discover something unacceptable about me,” the behavior is more likely to be part of the anxiety-maintaining system.
This nuance matters because dropping every coping behavior indiscriminately can be counterproductive. Treatment experiments usually target specific behaviors that are hypothesized to maintain fear and then test what happens without them. NICE's recommended Clark and Wells CBT explicitly includes experiential exercises on self-focus and safety-seeking behaviors, externally focused attention, video feedback, and behavioral experiments.
Anticipatory anxiety can start the conversation before it happens
Many socially anxious conversations begin hours or days before the first word. The person may simulate possible questions, rehearse answers, imagine awkward pauses, predict visible anxiety, inspect previous failures, or plan escape routes. Some preparation is useful. Anticipatory processing becomes costly when it functions as repeated threat simulation rather than practical planning.
The difference can be framed operationally. Practical preparation produces a finite plan: “I will arrive at seven, I know two people there, and I can ask Sam about the new job.” Threat rehearsal keeps generating new conditions for safety: “What if Sam leaves? What if I cannot think of anything? What if I blush? What if they notice I rehearsed?” The second process rarely reaches completion because its goal is certainty about an inherently uncertain social event.
After the conversation: post-event processing
Social anxiety often continues after the interaction through post-event processing: detailed, repetitive review of what happened, usually with selective attention to possible mistakes and signs of negative evaluation. A 2024 systematic review and meta-analysis by Edgar and colleagues found a moderate association between post-event rumination and social-anxiety symptoms across studies. A 2025 scoping review by Flynn and Yoon described links between post-event processing, negative memory biases, worsening performance appraisals, and self-focused attention, while also emphasizing unanswered questions about causality.
Conversation studies show why the process can feel so convincing. In one study, highly socially anxious participants evaluated their performance more negatively and engaged in more negative post-event processing after a conversation than low-anxiety participants; negative review remained linked to later performance appraisals. The problem is not remembering a conversation. The problem is treating an internally reconstructed version as if it were a neutral recording.
A useful post-conversation question is therefore not “How do I stop thinking about it?” but “What kind of review am I doing?” Evidence-based reflection asks what happened, what was observed, what prediction was tested, and what could be learned. Rumination repeatedly asks why the person was not flawless.
Does social anxiety mean poor social skills?
The evidence does not support a single answer. Some studies find that socially anxious participants perform similarly to controls while judging themselves more negatively. Other studies find observable differences in fluency, talking time, or interaction quality, especially in spontaneous conversations. A study by Strahan and Conger found no general performance deficit in a simulated interview, while work comparing speech and conversation tasks found that actual performance differences may be more visible in conversational interactions than in structured speeches.
That mixed evidence is clinically important. Social anxiety should not be reduced to a social-skills deficit, because a person can possess strong social skills and still experience intense fear, self-monitoring, avoidance, and post-event rumination. At the same time, years of avoidance can reduce practice, and anxiety-driven behaviors such as whispering, giving minimal answers, not asking follow-up questions, or leaving rapidly can affect the interaction.
The practical implication is assessment before remediation. If a person knows how to converse when relaxed but loses access to those abilities under scrutiny, the treatment target is different from a genuine skills gap. Skills practice can still be useful, especially after long avoidance or when another communication condition is present, but it should not automatically replace treatment of the fear system.
Face-to-face conversation and video calls
Video calls add a distinctive form of self-monitoring: a live image of the self can remain on screen throughout the interaction. Research by Vriends and colleagues found heightened self-focused attention during video chat in women with Social Anxiety Disorder, and greater self-focus was associated with greater self-rated anxiety. A 2026 randomized online-interaction study by Özhan, Graf, and Garner found that some self-view and audience-view configurations affected performance evaluation and negative post-event processing in participants with elevated social anxiety.
These findings are preliminary relative to the much larger face-to-face literature. They support a plausible practical principle: if the self-view becomes a mirror used to inspect appearance, expression, or visible anxiety, it can intensify the same self-monitoring process seen offline. Hiding self-view may help some people focus on the interaction, although it should not become a rigid rule that is believed necessary for safety.
What helps during a socially anxious conversation?
For a person with clinically significant Social Anxiety Disorder, a single conversational trick is not a substitute for treatment. Still, several strategies follow directly from the mechanisms targeted in evidence-based CBT. The goal is not to manufacture confidence before speaking. It is to make the conversation a place where attention, behavior, and predictions can be tested more accurately.
Shift attention from self-surveillance to the interaction
External attention does not mean forcing yourself to stare at the other person or suppressing internal sensations. It means repeatedly returning attention to information outside the internal performance monitor: the actual words being said, tone, context, shared topic, objects in the room, and what you genuinely want to understand.
NICE specifically includes systematic training in externally focused attention in Clark and Wells CBT for Social Anxiety Disorder. The rationale is straightforward: a prediction about another person's judgment cannot be tested if attention is almost entirely occupied by imagined evidence from inside the self.
Let a pause be a pause
A pause becomes more threatening when it is interpreted as a verdict. Instead of immediately filling silence, try allowing one ordinary pause to exist without rescue. Notice the prediction: “If I do not speak instantly, this will become unbearable” or “They will think I am boring.” Then observe what actually happens. The other person may speak. You may think of something naturally. The conversation may shift. Or the interaction may end without catastrophe.
This is not a rule to create artificial silence. It is an experiment in removing the assumption that conversational continuity must be controlled every second.
Say an adequate sentence before you find the perfect sentence
Perfectionistic pre-editing is a common form of covert safety behavior. If five possible replies are generated and rejected before one is spoken, the external conversation can move on while the internal conversation becomes increasingly elaborate. A useful experiment is to lower the standard from “the best response” to “a relevant, honest response.”
The aim is not impulsivity. It is to notice when editing is serving communication and when it is serving threat avoidance.
Reduce one safety behavior rather than all of them at once
Choose a specific behavior that appears tied to a feared prediction. Examples might include speaking at a normal volume instead of intentionally softening the voice, allowing a small disagreement instead of automatic agreement, keeping the phone away during a short interaction, or answering one personal question without over-rehearsing. The experiment should be challenging enough to generate learning and safe enough to be reasonable in context.
Behavioral experiments differ from tests of courage. Their purpose is to collect information. What did you predict? What did you change? What happened? What did the other person actually do? What remains uncertain? This method is central to CBT for Social Anxiety Disorder and closely related to exposure and behavioral experiments.
Afterward, compare prediction with evidence
If the conversation is replaying, make the review structured and finite. Write down the feared prediction in one sentence, the observable outcome in one sentence, and one alternative explanation. Avoid trying to reconstruct every facial expression. Human social information is ambiguous, and certainty cannot be recovered by replaying ambiguity for another hour.
For persistent post-event rumination, treatment can target the process directly. A systematic review and meta-analysis of psychological interventions found that pre-event and post-event rumination can decrease during psychological treatment for social anxiety, with CBT showing substantial within-group reductions, although study quality and intervention heterogeneity limit precise conclusions.
A simple conversation experiment
A useful self-help experiment can be built around one feared prediction rather than around the vague goal of “being more confident.” Consider this sequence.
Before the conversation: write one prediction that could, at least partly, be observed. “If there is a three-second pause, the other person will visibly disengage.” “If I say something without rehearsing it, I will say something unacceptable.” “If I stop smiling continuously, they will think I am unfriendly.” Avoid predictions that cannot be tested, such as “They will secretly think I am weird forever.”
Choose one process to change: direct more attention outward, reduce one safety behavior, or remain in the interaction slightly longer than usual. Keep the change small enough that you can actually observe the result.
During the conversation: let anxiety be present without using anxiety level as the outcome measure. The experiment can succeed while you feel anxious. Your job is to gather information about the prediction.
Afterward: record observable evidence for and against the prediction. Separate facts from interpretations. “They checked their phone once” is an observation; “they were desperate to escape me” is an interpretation. Then stop the formal review.
This exercise is educational and is not a substitute for individualized treatment. If social anxiety is severe, if exposure attempts repeatedly become overwhelming, or if there is significant depression, substance use, trauma, or another condition, working with a qualified clinician can make the experiments safer and more precise.
When conversation anxiety is not Social Anxiety Disorder
Conversation anxiety exists on a continuum. A person can feel shy, self-conscious, awkward, or afraid of judgment without having a mental disorder. The WHO ICD-11 Clinical Descriptions and Diagnostic Requirements and current clinical guidance distinguish a diagnosable disorder from ordinary variation by the overall clinical pattern, including persistence, disproportionate fear or anxiety, avoidance or marked distress, and interference with functioning.
The companion article Social Anxiety vs Social Anxiety Disorder explains this boundary in detail, while Social Anxiety vs Shyness addresses the separate distinction between temperament or ordinary shyness and clinical disorder.
One uncomfortable conversation, a quiet personality, disliking small talk, preferring written communication, or needing time to warm up does not establish Social Anxiety Disorder. Neither does a screening score. Diagnosis requires clinical assessment of the full pattern and possible alternative explanations.
Important differential explanations for conversational silence
Silence and verbal difficulty are diagnostically nonspecific. Social anxiety is one possible explanation among many, and more than one process can be present at the same time.
Selective mutism has its own diagnostic territory and should not be used as a synonym for “I freeze” or “I am quiet when anxious.” Autism can involve differences in social communication, sensory processing, reciprocity, masking, or processing load, and autistic people can also have co-occurring social anxiety. Depression can reduce energy, speech, interest, and social engagement. Trauma-related states can produce freezing, detachment, or threat sensitivity. Stuttering and other speech-language differences can create secondary fear of evaluation. Hearing difficulties, language barriers, sleep deprivation, substance effects, medication effects, and neurological or medical conditions can also change communication.
NICE recommends that assessment of possible Social Anxiety Disorder consider coexisting conditions, including mood disorders, other anxiety disorders, substance misuse, psychosis, and autism, and that assessment in younger people also consider speech, language, communication, and physical-health needs. The point is not to create a long list of self-diagnoses. It is to prevent a single symptom such as silence from being treated as diagnostic proof.
For broader diagnostic reasoning, see How Is Anxiety Diagnosed? Assessment, History, and Differential Diagnosis.
When social anxiety in conversations becomes a clinical problem
A clinically important pattern is more likely when fear and avoidance are persistent, the situations repeatedly involve scrutiny or possible negative evaluation, the anxiety is out of proportion to the actual context, and the pattern limits relationships, education, work, health care, independence, or other valued activities. The exact diagnostic formulation depends on a clinical assessment rather than on one conversational symptom.
Signs that assessment may be useful include repeatedly avoiding calls, meetings, dates, classes, shops, or social events because conversation is expected; relying on alcohol or other substances to talk; losing opportunities because speaking feels impossible; having intense anticipatory anxiety for days before ordinary interactions; spending long periods replaying conversations; or organizing life around not being noticed.
The broader overview Social Anxiety Disorder: Symptoms, Causes, Diagnosis, and Treatment covers the diagnostic territory, while Social Anxiety Symptoms maps emotional, cognitive, physical, and behavioral signs without turning any single sign into a diagnosis.
What treatment evidence says
For adults diagnosed with Social Anxiety Disorder, NICE recommends individual CBT specifically developed for the disorder, based on the Clark and Wells or Heimberg models. Clark and Wells CBT explicitly targets self-focused attention, safety-seeking behavior, negative self-imagery, behavioral experiments, and pre- and post-event processing. These are the same mechanisms that often become visible in conversation.
The broader treatment evidence is substantial. A 2024 systematic review and meta-analysis by de Ponti and colleagues included 66 randomized controlled trials with 5,560 participants and found psychotherapy effective for reducing adult Social Anxiety Disorder symptoms, while noting significant risk of bias in the literature. More recent network meta-analyses continue to support CBT as a major evidence-based option, but comparative rankings should be interpreted with attention to study quality and differences among protocols.
Treatment is not simply “practice talking more.” Exposure can be useful, but evidence-based social-anxiety treatment also addresses the way the exposure is conducted. Repeated conversations in which a person remains intensely self-focused, performs many safety behaviors, and then spends hours concluding the interaction went badly may produce less corrective learning than conversations used as structured behavioral experiments.
The full treatment owner is Social Anxiety Disorder Treatment: CBT, Exposure, Medication, and Self-Help. This page stays focused on the conversational mechanism rather than duplicating that treatment guide.
Frequently asked questions
Why do I go quiet around people even when I want to talk?
Social anxiety can shift attention toward self-monitoring and threat prediction, which can make spontaneous responding harder. Some people also reduce talking as a safety behavior to lower the chance of saying something they fear will be judged. Going quiet has many other possible explanations, so the pattern and context matter.
Can social anxiety make your mind go blank in conversation?
Yes. NIMH lists a mind-going-blank experience among common manifestations of Social Anxiety Disorder. Experimental research also supports a role for self-focused attention and safety behavior in disrupting social interaction. A blank mind is not specific to social anxiety and should not be used as a diagnostic test.
Is awkward silence a sign of social anxiety?
It can become a trigger or focus of social anxiety, but silence itself is not diagnostic. Pauses are normal in conversation. The clinically relevant pattern is the meaning attached to the pause, the intensity of fear, the behaviors used to prevent it, and the degree to which the pattern causes avoidance or impairment.
Does social anxiety mean I have poor social skills?
Not necessarily. Research is mixed: some socially anxious people perform as well as controls while evaluating themselves more negatively, whereas some studies find observable differences in spontaneous conversation, often involving reduced talking or anxiety-driven safety behavior. Assessment should distinguish skill, access to skill under threat, and avoidance history.
Why do I replay every conversation afterward?
Post-event processing is a well-studied feature of social anxiety. It involves repetitive review of a social event, often with emphasis on perceived mistakes or signs of judgment. Meta-analytic evidence shows a moderate association with social-anxiety symptoms, but rumination also occurs in other conditions and in people without a diagnosis.
Should I memorize conversation scripts?
A small amount of preparation can help with a genuinely unfamiliar situation. Rigid scripts can become safety behaviors if the person concludes they cannot cope without them. A more flexible approach is to prepare a few topics or questions while allowing unscripted responses, pauses, and imperfect wording.
Should I force eye contact to beat social anxiety?
No universal amount of eye contact is required. The useful question is whether gaze avoidance is being used rigidly to prevent feared judgment and whether changing it is an appropriate behavioral experiment. Culture, neurodivergence, trauma history, sensory needs, and individual communication style all matter.
Can video calls make social anxiety worse?
They can for some people, particularly when a live self-view increases self-monitoring. Experimental work has linked self-view conditions with self-focused attention, negative performance appraisal, or post-event processing. The evidence is still developing, so hiding self-view is best treated as a practical experiment rather than a universal rule.
When should I seek professional help?
Consider an assessment when fear of conversation is persistent, distressing, increasingly avoided, or meaningfully interferes with relationships, education, work, health care, or daily functioning. Help is also important when alcohol or other substances are becoming necessary to socialize, depression is significant, or another condition may be contributing. A clinician can distinguish symptoms from a formal diagnosis and discuss evidence-based treatment options.
Related Articles
References
de Ponti N, Matbouriahi M, Franco P, et al. The efficacy of psychotherapy for social anxiety disorder, a systematic review and meta-analysis. Journal of Anxiety Disorders. 2024;104:102881. doi:10.1016/j.janxdis.2024.102881.
Donohue HE, Modini M, Abbott MJ. Psychological interventions for pre-event and post-event rumination in social anxiety: A systematic review and meta-analysis. Journal of Anxiety Disorders. 2024;102:102823. doi:10.1016/j.janxdis.2023.102823.
Edgar EV, Richards A, Castagna PJ, Bloch MH, Crowley MJ. Post-event rumination and social anxiety: a systematic review and meta-analysis. Journal of Psychiatric Research. 2024;173:87–97. doi:10.1016/j.jpsychires.2024.03.013.
Flynn AJ, Yoon KL. Post-event processing in social anxiety: A scoping review. Journal of Anxiety Disorders. 2025;109:102947. doi:10.1016/j.janxdis.2024.102947.
Gregory B, Peters L. Changes in the self during cognitive behavioural therapy for social anxiety disorder: A systematic review. Clinical Psychology Review. 2017;52:1–18. doi:10.1016/j.cpr.2016.11.008.
Leigh E, Chiu K, Clark DM. Self-focused attention and safety behaviours maintain social anxiety in adolescents: An experimental study. PLOS ONE. 2021;16:e0247703. doi:10.1371/journal.pone.0247703.
McManus F, Sacadura C, Clark DM. Why social anxiety persists: an experimental investigation of the role of safety behaviours as a maintaining factor. Journal of Behavior Therapy and Experimental Psychiatry. 2008;39(2):147–161. doi:10.1016/j.jbtep.2006.12.002.
National Institute for Health and Care Excellence. Social anxiety disorder: recognition, assessment and treatment — Recommendations. Clinical guideline CG159.
National Institute of Mental Health. Social Anxiety Disorder: What You Need to Know. National Institutes of Health.
Özhan N, Graf EW, Garner M. Social anxiety in online social interactions: Examining the effects of self and audience images on anxiety, self-awareness and performance evaluations. Journal of Anxiety Disorders. 2026;118:103111. doi:10.1016/j.janxdis.2026.103111.
Piccirillo ML, Dryman MT, Heimberg RG. Safety Behaviors in Adults With Social Anxiety: Review and Future Directions. Behavior Therapy. 2016;47(5):675–687. doi:10.1016/j.beth.2015.11.005.
Rapee RM, Heimberg RG. A cognitive-behavioral model of anxiety in social phobia. Behaviour Research and Therapy. 1997;35(8):741–756. doi:10.1016/S0005-7967(97)00022-3.
Spurr JM, Stopa L. Self-focused attention in social phobia and social anxiety. Clinical Psychology Review. 2002;22(7):947–975. doi:10.1016/S0272-7358(02)00107-1.
Stevens S, Hofmann M, Kiko S, et al. What determines observer-rated social performance in individuals with social anxiety disorder? Journal of Anxiety Disorders. 2010;24(8):830–836. doi:10.1016/j.janxdis.2010.06.005.
Strahan E, Conger AJ. Social anxiety and its effects on performance and perception. Journal of Anxiety Disorders. 1998;12(4):293–305. doi:10.1016/S0887-6185(98)00016-4.
Voncken MJ, Bögels SM. Social performance deficits in social anxiety disorder: reality during conversation and biased perception during speech. Journal of Anxiety Disorders. 2008;22(8):1384–1392. doi:10.1016/j.janxdis.2008.02.001.
Vriends N, Meral Y, Bargas-Avila JA, Stadler C, Bögels SM. How do I look? Self-focused attention during a video chat of women with social anxiety (disorder). Behaviour Research and Therapy. 2017;92:77–86. doi:10.1016/j.brat.2017.02.008.
World Health Organization. Clinical descriptions and diagnostic requirements for ICD-11 mental, behavioural and neurodevelopmental disorders. Geneva: World Health Organization; 2024. ISBN 978-92-4-007726-3.