The Neuroscience of Social Anxiety: Why Your Brain Treats Rejection Like Physical Pain

A Reddit user with 811 upvotes asked a question that captures the lived experience of social anxiety better than any clinical description: “Do you ever feel social anxiety has killed your personality? At my core, I’m witty and goofy. But my social anxiety has gotten so severe, I feel I’ve lost all of my personality. At work, I can hardly think of anything to say. When I do talk, my voice gets shakey and no one listens to me. I can never think of a single joke during the conversation, when I was once so witty.”

Read that again. This person knows they’re funny. They can feel their real self underneath. But in the moment, something shuts it down — not by choice, not by habit, but by a neurological process that hijacks cognition before conscious thought even begins.

Another user, with 1,271 upvotes and a perfect 100% upvote ratio, put it this way: “It’s not just fear; it’s the overwhelming sense that you’re being watched, singled out, even in a crowd of thousands. It’s like the entire world is focused on you, dissecting every small movement, every word.”

The most insightful comment I found came from someone describing what their therapist told them about why knowing it’s anxiety doesn’t make it stop: “What about a hurricane? It could be days, a week away from hitting. It’s out in the ocean. You know it might not even hit you directly. But does knowing that stop the fear? Does understanding the weather patterns make the wind less scary?”

That therapist was describing something neuroscience has now mapped in extraordinary detail: social anxiety isn’t a thinking problem. It’s a neural circuit problem. Your brain is running a miscalibrated threat detection system that treats social evaluation as a survival-level danger — and it operates faster than conscious thought can intervene.

This article is the full neuroscience. Not “just put yourself out there.” Not “fake it till you make it.” The actual biology of why your brain does this, what’s happening in real time during a social anxiety episode, and what the evidence says about which interventions actually rewire the circuit.

---

Section 1: Social Pain Is Physical Pain — This Is Not a Metaphor

In 2003, neuroscientist Naomi Eisenberger published a study that fundamentally changed how we understand social rejection. She put participants in an fMRI scanner and had them play a simple ball-tossing video game. Midway through, the other “players” (actually computer programs) stopped throwing the ball to the participant — a virtual rejection [1].

The brain regions that activated during this social exclusion were the dorsal anterior cingulate cortex (dACC) and the anterior insula — the same regions that process physical pain. Not metaphorically similar regions. The same regions. The overlap was so precise that Eisenberger could predict how much emotional distress a participant reported based solely on the activation pattern in their pain matrix [1].

Subsequent studies confirmed this wasn’t a fluke:

• Tylenol reduces social pain. DeWall et al. (2010) found that acetaminophen — a physical pain reliever — reduced both the neural response to social rejection and self-reported hurt feelings over a 3-week period [2].
• Rejection activates the same opioid system as physical injury. A 2011 PET imaging study showed that social rejection triggered the mu-opioid system — the brain’s endogenous painkiller network — in the same regions activated by physical pain [3].
• The intensity scales similarly. People who experienced intense recent romantic rejection showed activation in the secondary somatosensory cortex and posterior insula — regions typically associated with the sensory component of physical pain, not just the emotional component [4].

Why this matters for social anxiety: If your brain processes social evaluation through the same circuits as physical danger, then social anxiety isn’t an overreaction to a harmless situation. From your brain’s perspective, the social threat is processed with the same neural machinery as a threat to your physical body. Telling someone with social anxiety to “just relax, no one’s going to hurt you” is neurologically equivalent to telling someone being punched to relax because “it’s just pressure.”

The question isn’t whether social anxiety sufferers are being irrational. The question is why their social pain system is calibrated so much more sensitively than average.

---

Section 2: The Amygdala-Prefrontal Mismatch — A System Out of Balance

The core neural circuit dysfunction in social anxiety involves two key structures: the amygdala and the medial prefrontal cortex (mPFC), and the relationship between them is broken in a specific, measurable way.

The Amygdala: Threat Detection at 120 Milliseconds

Your amygdala processes social threat signals — angry faces, disapproving expressions, ambiguous social cues — in approximately 120 milliseconds. That’s before conscious awareness [5]. You’re already in a threat state before you’ve had a single thought about the situation.

In social anxiety disorder (SAD), the amygdala shows two critical differences from neurotypical brains:

Hyperreactivity to faces. fMRI studies consistently show that socially anxious individuals have significantly greater amygdala activation in response to faces — not just angry or threatening faces, but all faces, including neutral ones [6]. Their amygdala treats every human face as a potential threat signal.

Failure to habituate. In neurotypical brains, the amygdala response to repeated face presentations diminishes — you get used to it. In SAD, the amygdala doesn’t habituate [7]. The 100th face is as threatening as the first. This is why “just spending more time around people” doesn’t automatically fix social anxiety — the normal habituation mechanism is impaired.

The Medial Prefrontal Cortex: The Missing Brake

The mPFC is supposed to regulate the amygdala — essentially telling it “yes, that’s a face, but it’s not dangerous.” In healthy social processing, the mPFC evaluates context, applies prior experience, and sends inhibitory signals that modulate the amygdala’s initial reaction.

In social anxiety, mPFC-to-amygdala connectivity is significantly reduced [8]. The brake pedal is disconnected from the brake pads. The amygdala fires, and the prefrontal regulation that should dampen the response is too weak or too slow to intervene.

This explains the experience described by the Reddit user who lost access to her wit and humor. The cognitive resources that produce clever, spontaneous social behavior — working memory, creative association, verbal fluency — are all prefrontal functions. When the amygdala is hijacking processing resources for threat detection, the prefrontal cortex can’t simultaneously run its higher-order social cognition programs. You don’t lose your personality. Your brain reallocates the computational resources that produce your personality toward survival processing.

---

Section 3: The Prediction Machine — Why Your Brain Expects the Worst

Modern neuroscience increasingly understands the brain as a prediction machine — it doesn’t just react to the world, it constantly generates predictions about what will happen next and compares those predictions against incoming sensory data [9].

In this framework, social anxiety is a prediction error calibration problem. Here’s how it works:

The Bayesian Brain in Social Situations

Your brain maintains a probabilistic model of social outcomes — essentially, a set of weighted predictions about how people will react to you. In a healthy brain, these predictions are flexible: they update based on actual social feedback. Positive interaction → slightly more positive predictions next time. Negative interaction → slightly more cautious predictions.

In social anxiety, the predictions are pathologically overweighted toward negative outcomes [10]. Your brain assigns high probability to rejection, judgment, and humiliation — and low probability to acceptance, interest, or positive regard — regardless of actual evidence.

Worse, the system has a confirmation bias built in:

• Negative feedback confirms the threat model (“See? I knew they were judging me.”)
• Positive feedback is discounted (“They’re just being polite.” “They didn’t notice my awkwardness.” “Wait until they really get to know me.”)

This means social experience alone — just being around people more — doesn’t automatically correct the model. The prediction system filters incoming evidence to maintain its existing bias. This is why the “just get out more” advice fails for many people with social anxiety: their brain’s prediction machinery interprets social data in a way that reinforces the threat model.

The Spotlight Effect as a Computational Error

The “spotlight effect” — the feeling that everyone is watching and evaluating you — is one of the most universal features of social anxiety. Neuroscience now understands this as a specific computational error in the brain’s self-referential processing.

Research using fMRI shows that socially anxious individuals have hyperactive midline cortical structures — specifically the medial prefrontal cortex (in its self-referential mode) and the posterior cingulate cortex — both key nodes of the default mode network (DMN) [11].

The DMN is the brain network that activates during self-referential thinking: thinking about yourself, imagining how others see you, replaying past social interactions, imagining future ones. In social anxiety, this network is overactive during social situations when it should be suppressed [11].

The result: your brain is simultaneously trying to process the social environment AND running an intense self-monitoring program (“How do I look? What do they think of me? Did I say that wrong? Are they bored?”). These competing demands create the cognitive overload that makes socially anxious people feel slow, inarticulate, and unlike themselves.

---

Section 4: The GABA Deficit — An Inhibition Problem

There’s a neurochemical dimension to social anxiety that most popular content ignores: GABA (gamma-aminobutyric acid), the brain’s primary inhibitory neurotransmitter.

GABA acts as the brain’s braking system — it reduces neural excitability, prevents runaway activation, and is essential for keeping the amygdala’s threat responses in check. Multiple studies using magnetic resonance spectroscopy (MRS) have found that individuals with social anxiety disorder have significantly lower GABA concentrations in the occipital and anterior cingulate cortices compared to healthy controls [12].

This GABA deficit has several practical implications:

The amygdala runs hotter without adequate GABAergic inhibition, which partially explains the hyperreactivity described in Section 2.

Benzodiazepines work so dramatically because they enhance GABA activity — they’re essentially replacing the inhibition the brain isn’t producing enough of on its own. This explains the Reddit post from someone asking “Why can’t I take benzos for life?” (534 upvotes): the medication feels like it corrects a fundamental neurochemical imbalance, which it temporarily does. The problem is tolerance, dependence, and the fact that benzos don’t fix the underlying GABA production issue.

Alcohol is the most common self-medication for social anxiety precisely because it enhances GABA signaling. The social anxiety → alcohol dependence pipeline is one of the most well-documented comorbidity patterns in psychiatry [13].

What actually increases endogenous GABA production? Exercise (particularly intense exercise), yoga, meditation, and certain dietary factors. More on these in Section 8.

---

Section 5: The Serotonin-Oxytocin Story — Why SSRIs Help (and Their Limits)

SSRIs (selective serotonin reuptake inhibitors) are the first-line pharmacological treatment for social anxiety disorder, and they work for approximately 50-60% of patients [14]. Understanding why they work — and why they fail for the rest — requires understanding serotonin’s role in social processing.

Serotonin: Not Just a “Happy Chemical”

Serotonin’s role in social anxiety is less about mood and more about social threat interpretation. Serotonin modulates how the amygdala responds to social signals — particularly ambiguous ones [15].

When serotonin function is adequate, ambiguous social cues (a neutral expression, an unanswered text, someone looking in your direction) are more likely to be interpreted neutrally or positively. When serotonin function is low, the same cues are more likely to be interpreted as threatening [15].

SSRIs don’t make you happy — they shift the interpretive bias of your social processing system toward neutral. That facial expression that your brain was reading as “disgusted” gets reprocessed as “neutral” or “preoccupied.” The window of ambiguous-but-threatening social signals shrinks.

Oxytocin: The Missing Piece

Oxytocin, sometimes called the “bonding hormone,” plays a critical role in social approach behavior. It reduces amygdala reactivity to social threats and enhances the reward value of social interaction [16].

Research shows that individuals with social anxiety have altered oxytocin receptor gene expression and lower baseline oxytocin levels [17]. This creates a cruel catch-22: social connection increases oxytocin, which makes social interaction feel safer and more rewarding. But social anxiety prevents the social connection that would produce the oxytocin that would reduce the social anxiety.

This is one of the strongest neurobiological arguments for structured gradual exposure — not as a willpower exercise, but as a pharmacological intervention. Each successful social interaction produces a small oxytocin release that makes the next interaction slightly less threatening. The brain is literally self-medicating through social behavior.

---

Section 6: The Vagus Nerve — Why “It’s Stuck in the Body” Is Neuroscientifically Accurate

One of the highest-upvoted posts I found (783 upvotes, 98% ratio) was titled: “Anxiety isn’t in the head — it’s stuck in the body.” The poster described how shaking, dancing, and chaotic physical movement resolved an anxiety spiral more effectively than any cognitive technique.

This isn’t placebo. The vagus nerve — the longest cranial nerve, running from your brainstem through your neck, chest, and abdomen — is the primary communication highway between your body and your brain’s threat detection system [18].

Vagal Tone and Social Engagement

Stephen Porges’ polyvagal theory (simplified and stripped of its more controversial claims) highlights a key observation: the vagus nerve has a “social engagement” branch (the ventral vagal complex) that, when active, promotes calm alertness, facial expressiveness, vocal prosody, and social receptivity [19]. When this branch is suppressed — by chronic stress, trauma, or anxiety — the body shifts toward sympathetic (fight-or-flight) or dorsal vagal (freeze/shutdown) states.

Measurable vagal tone (typically assessed via heart rate variability, or HRV) is significantly lower in individuals with social anxiety [20]. Low vagal tone means:

• The body defaults to sympathetic activation in social situations
• Facial muscles become less expressive (contributing to the “flat” presentation others perceive as unfriendliness)
• Voice becomes monotone or shaky (the vagus nerve innervates the laryngeal muscles)
• The gut enters a stress state (the GI symptoms that accompany social anxiety)

Why Body-Based Interventions Work

The vagus nerve is bidirectional — it sends signals from body to brain, not just brain to body. This means interventions that change physiological state can influence cognitive-emotional processing:

• Slow diaphragmatic breathing (extended exhale) directly stimulates vagal afferents, increasing parasympathetic tone [21].
• Cold water exposure to the face triggers the diving reflex, a powerful vagal activation response.
• Vigorous movement followed by stillness (what the Reddit poster described) may work by first activating the sympathetic system (movement) and then allowing a parasympathetic rebound (stillness), essentially “resetting” autonomic tone.
• Vocalization — humming, singing, chanting — vibrates the vocal cords adjacent to the vagus nerve, providing direct mechanical stimulation.

The neuroscience validates what the Reddit poster discovered intuitively: sometimes you need to change the body’s state to change the brain’s state, because the bottom-up signal (body → brain via vagus) can be more powerful than the top-down signal (thinking about not being anxious).

---

Section 7: Why “Just Knowing” Doesn’t Fix It — The Speed Problem

The therapist’s hurricane metaphor from Reddit is perfect because it captures the central frustration of social anxiety: understanding it doesn’t stop it.

Here’s why, neurobiologically:

The amygdala processes threat signals in approximately 120 milliseconds [5]. Conscious awareness takes roughly 300-500 milliseconds [22]. Prefrontal regulatory processes that could modulate the amygdala response take 500-1000+ milliseconds [22].

This means by the time you’ve had the thought “It’s just anxiety, this isn’t dangerous,” your amygdala has already:

Detected the “threat” (120ms)

Triggered norepinephrine and cortisol release via the HPA axis (200ms)

Initiated sympathetic nervous system activation — increased heart rate, shallow breathing, blood flow to muscles (300ms)

Redirected cognitive resources from prefrontal “social” processing to amygdala-driven “survival” processing (400ms)

Your conscious reasoning arrives after the physiological cascade is already underway. Trying to think your way out of a social anxiety response is like trying to unbake a cake — the chemical reactions have already occurred.

This is why:

• Cognitive techniques alone often fail in the moment (they arrive too late in the neural timeline)
• Body-based techniques can be faster (they bypass the cognitive delay by directly targeting the autonomic nervous system)
• Prevention is more effective than intervention (reducing baseline amygdala reactivity through ongoing practices, rather than trying to stop an episode in progress)
• The “just get out there” advice misunderstands the problem — it assumes the issue is avoidance (a behavior) when the issue is neural circuit calibration (a brain state)

---

Section 8: What Actually Rewires the Circuit — Evidence-Ranked Interventions

Not all social anxiety interventions are equal. Here they are, ranked by strength of neurobiological evidence for actually changing the underlying circuit dysfunction:

Tier 1: Strong Evidence for Circuit-Level Change

1. Cognitive Behavioral Therapy (CBT) — specifically exposure-based

CBT with exposure components doesn’t just change thoughts — it literally changes brain structure and function. A meta-analysis of neuroimaging studies found that successful CBT for social anxiety produces measurable reductions in amygdala reactivity and increased mPFC-amygdala connectivity [23]. The brain’s regulatory circuit gets physically strengthened.

The mechanism: exposure creates prediction errors. When you enter a feared social situation and the catastrophic outcome doesn’t occur, your brain’s prediction model must update. Enough prediction errors shift the prior probability weights away from threat — rewiring the Bayesian predictions described in Section 3.

Critical caveat: exposure must include emotional processing, not just behavioral compliance. White-knuckling through a social situation while dissociating doesn’t generate meaningful prediction errors. You need to actually feel the anxiety, notice that the feared outcome doesn’t occur, and let your brain register the discrepancy [24].

2. SSRIs (8-12 weeks minimum)

PET imaging shows that SSRI treatment normalizes serotonin transporter binding in the amygdala and raphe nuclei, reducing social threat interpretation bias [25]. The 8-12 week timeline reflects how long it takes for downstream receptor changes — not just serotonin levels — to stabilize.

Limitation: SSRIs address the serotonergic component but not the GABA deficit, the vagal tone issue, or the prediction model calibration. This is why medication + therapy outperforms either alone.

3. Aerobic Exercise (3-5x/week, moderate-to-vigorous)

Exercise produces multiple relevant neurochemical changes simultaneously:

• Increases GABA synthesis (addressing the inhibition deficit) [26]
• Increases BDNF (brain-derived neurotrophic factor), which promotes neuroplasticity in the mPFC-amygdala circuit [27]
• Temporarily reduces amygdala reactivity for 4-6 hours post-exercise
• Improves vagal tone (increases HRV) over weeks of consistent practice [28]

A 2021 meta-analysis found that exercise interventions reduced social anxiety symptoms with effect sizes comparable to CBT [29]. Not as a supplement. As a comparable standalone intervention.

Tier 2: Moderate Evidence

4. Mindfulness Meditation (8+ weeks of regular practice)

Meditation increases mPFC thickness and strengthens mPFC-amygdala connectivity — the exact circuit that’s dysfunctional in social anxiety [30]. It also reduces default mode network overactivity during social processing, quieting the self-referential “Am I being judged?” processing loop.

The effect requires consistency. Brief or sporadic meditation shows minimal neural changes. Eight weeks of daily practice (as in Mindfulness-Based Stress Reduction protocols) shows measurable structural changes.

5. Vagal Nerve Stimulation (Breathing Protocols)

Extended-exhale breathing (inhale 4 counts, exhale 6-8 counts) practiced daily for 4+ weeks shows improved HRV and reduced anxiety symptoms in controlled studies [21]. The key is daily practice, not emergency deployment during a panic episode.

6. Gradual Social Exposure (Self-Directed)

Structured, voluntary, progressive social challenges — starting with low-threat situations and gradually increasing — can generate the prediction errors needed to update the brain’s social threat model. Key requirements:

• Must be voluntary (forced exposure can worsen anxiety)
• Must be graduated (too much, too fast triggers avoidance learning)
• Must include post-exposure processing (reflecting on: “What did I predict? What actually happened?”)

Tier 3: Emerging Evidence

7. Oxytocin-Promoting Activities

Social touch, positive social interactions, warm physical contact, and care-giving behaviors increase oxytocin release. While intranasal oxytocin administration shows mixed results in research, naturally increasing oxytocin through safe social bonding may help lower the social approach threshold over time.

8. Gut Microbiome Interventions

Emerging research links gut microbiome composition to social anxiety via the gut-brain axis. Certain Lactobacillus and Bifidobacterium strains increase GABA production in the gut, which signals the brain via the vagus nerve [31]. This field is young — specific strain recommendations are premature, but the mechanism is biologically plausible and the preliminary data is promising.

What Doesn’t Work (Despite Popularity)

• Avoidance — Reduces anxiety in the short term but strengthens the amygdala’s threat associations long-term
• Alcohol/substances — Addresses GABA deficit temporarily but creates dependence and prevents natural circuit rewiring
• Positive affirmations — No evidence for changing amygdala reactivity or prediction models in clinical populations
• Forcing yourself to “just be confident” — Doesn’t address the 120ms amygdala problem or the prediction calibration issue

---

Section 9: The Social Anxiety Paradox — When It Gets Better, It Gets Better Faster

There’s a nonlinear dynamic in social anxiety recovery that the neuroscience predicts but most people don’t expect: the first 30% improvement is the hardest, and then it accelerates.

Here’s why:

Oxytocin feedback loop. Each successful social interaction produces oxytocin, which makes the next interaction less threatening. This creates a positive feedback loop — but it requires reaching a threshold of social engagement to get the loop started.

Prediction model updates compound. Each prediction error (entering a feared situation and surviving) doesn’t just update one prediction — it weakens the overall threat weighting of the social prediction model. After enough errors, the model’s confidence in negative predictions drops below its confidence threshold and the whole system shifts.

Vagal tone improvements are self-reinforcing. Higher vagal tone → calmer physiological state → better social experiences → higher vagal tone. Like compound interest, this requires an initial investment that feels disproportionate to the returns.

GABA normalization takes weeks but then maintains. Regular exercise and meditation gradually increase GABA synthesis. The first 4-6 weeks feel effortful with minimal apparent benefit. After 8+ weeks, the neurochemical change reaches a threshold that noticeably reduces baseline anxiety [32].

This is why so many Reddit recovery stories follow the same arc: months of grinding effort with minimal visible progress, followed by a period where improvement seems to accelerate rapidly. They’re describing the nonlinear dynamics of a neural circuit reaching its tipping point.

---

Section 10: An Honest Framework for Starting

I’m not going to pretend that understanding the neuroscience of social anxiety makes it easy to change. If knowledge were sufficient, that therapist’s hurricane metaphor wouldn’t resonate with 1,667 people.

But understanding the mechanism does three things:

1. It removes the self-blame. Your brain is running a miscalibrated threat detection system operating faster than conscious thought. You are not weak. You are not broken. You are not choosing this. Your amygdala is doing exactly what it evolved to do — it’s just overweighting social threats based on a prediction model that hasn’t been properly updated. 2. It points to the right interventions. If you understand that the problem is an amygdala-prefrontal circuit mismatch, a GABA deficit, and a miscalibrated prediction model, then the interventions that address those specific mechanisms (exposure with processing, exercise, breathing practices, potentially medication) make logical sense. You’re not doing random self-help. You’re targeting the specific neural systems that are miscalibrated. 3. It explains the timeline. Neural circuits don’t rewire in a weekend. Receptor densities don’t change in a week. Prediction models built over years of social experience don’t update from a single positive interaction. If you understand that the biological change takes 8-12 weeks to manifest, you can persist through the early phase when effort feels futile. A starting framework (not a treatment plan — work with a professional if your social anxiety is severe): Weeks 1-2: Foundation

• Start daily extended-exhale breathing (5 minutes, twice daily)
• Begin exercise 3x/week if not already
• Track anxiety levels 1-10 daily to establish a baseline

Weeks 3-4: Gentle Exposure

• Identify 3 low-stakes social situations you currently avoid
• Enter one per week, deliberately
• After each: write down what you predicted vs. what actually happened

Weeks 5-8: Circuit Training

• Increase exposure frequency and difficulty gradually
• Maintain exercise and breathing practices
• Consider whether therapy (CBT with exposure) or medication would accelerate the process

Weeks 9-12: Assessment

• Compare current anxiety levels to your Week 1 baseline
• Note which situations trigger less amygdala response than they used to
• Evaluate whether the investment is generating compound returns

The person who said social anxiety killed their personality has it backward — their personality isn’t dead. It’s held hostage by a neural circuit that’s prioritizing threat detection over social cognition. The personality is still there, fully intact, waiting for the circuit to recalibrate enough to let it run.

The neuroscience says that circuit can change. It takes specific inputs, sustained effort, and more time than anyone wants. But the brain is plastic, the mechanisms are understood, and the prediction model your amygdala is running was built by experience — which means it can be rebuilt by experience.

---

References

[1] Eisenberger, N. I., Lieberman, M. D., & Williams, K. D. (2003). “Does rejection hurt? An fMRI study of social exclusion.” Science, 302(5643), 290-292.

[2] DeWall, C. N., et al. (2010). “Acetaminophen reduces social pain: Behavioral and neural evidence.” Psychological Science, 21(7), 931-937.

[3] Hsu, D. T., et al. (2013). “Response of the μ-opioid system to social rejection and acceptance.” Molecular Psychiatry, 18(11), 1211-1217.

[4] Kross, E., et al. (2011). “Social rejection shares somatosensory representations with physical pain.” Proceedings of the National Academy of Sciences, 108(15), 6270-6275.

[5] Morris, J. S., et al. (1998). “Conscious and unconscious emotional learning in the human amygdala.” Nature, 393(6684), 467-470.

[6] Stein, M. B., et al. (2002). “Increased amygdala activation to angry and contemptuous faces in generalized social phobia.” Archives of General Psychiatry, 59(11), 1027-1034.

[7] Sladky, R., et al. (2012). “Disrupted effective connectivity between the amygdala and prefrontal cortex in social anxiety disorder.” Human Brain Mapping, 33(3), 643-654.

[8] Prater, K. E., et al. (2013). “Aberrant amygdala-frontal cortex connectivity during perception of fearful faces and at rest in generalized social anxiety disorder.” Depression and Anxiety, 30(3), 234-241.

[9] Clark, A. (2013). “Whatever next? Predictive brains, situated agents, and the future of cognitive science.” Behavioral and Brain Sciences, 36(3), 181-204.

[10] Koban, L., et al. (2017). “Social anxiety is characterized by biased learning of social feedback.” Cognitive, Affective, & Behavioral Neuroscience, 17(4), 783-799.

[11] Gentili, C., et al. (2016). “Beyond emotions: A meta-analysis of neural response within face processing system in social anxiety.” Experimental Biology and Medicine, 241(3), 225-237.

[12] Pollack, M. H., et al. (2008). “GABA and social anxiety disorder.” Expert Review of Neurotherapeutics, 8(4), 633-645.

[13] Schneier, F. R., et al. (2010). “Social anxiety disorder and alcohol use disorder comorbidity in the National Epidemiologic Survey on Alcohol and Related Conditions.” Psychological Medicine, 40(6), 977-988.

[14] Stein, D. J., et al. (2004). “Efficacy of paroxetine for relapse prevention in social anxiety disorder.” Archives of General Psychiatry, 59(12), 1111-1118.

[15] Harmer, C. J., et al. (2004). “Increased positive versus negative affective perception and memory in healthy volunteers following selective serotonin and norepinephrine reuptake inhibition.” American Journal of Psychiatry, 161(7), 1256-1263.

[16] Kirsch, P., et al. (2005). “Oxytocin modulates neural circuitry for social cognition and fear in humans.” Journal of Neuroscience, 25(49), 11489-11493.

[17] Dodhia, S., et al. (2014). “Modulation of resting-state amygdala-frontal functional connectivity by oxytocin in generalized social anxiety disorder.” Neuropsychopharmacology, 39(9), 2061-2069.

[18] Breit, S., et al. (2018). “Vagus nerve as modulator of the brain-gut axis in psychiatric and inflammatory disorders.” Frontiers in Psychiatry, 9, 44.

[19] Porges, S. W. (2011). The Polyvagal Theory: Neurophysiological Foundations of Emotions, Attachment, Communication, and Self-regulation. Norton.

[20] Alvares, G. A., et al. (2013). “Reduced heart rate variability in social anxiety disorder: Associations with gender and symptom severity.” PLoS ONE, 8(7), e70468.

[21] Gerritsen, R. J. S., & Band, G. P. H. (2018). “Breath of life: The respiratory vagal stimulation model of contemplative activity.” Frontiers in Human Neuroscience, 12, 397.

[22] Pessoa, L. (2005). “To what extent are emotional visual stimuli processed without attention and awareness?” Current Opinion in Neurobiology, 15(2), 188-196.

[23] Månsson, K. N. T., et al. (2016). “Neuroplasticity in response to cognitive behavior therapy for social anxiety disorder.” Translational Psychiatry, 6(2), e727.

[24] Foa, E. B., & Kozak, M. J. (1986). “Emotional processing of fear: Exposure to corrective information.” Psychological Bulletin, 99(1), 20-35.

[25] Lanzenberger, R. R., et al. (2007). “Reduced serotonin-1A receptor binding in social anxiety disorder.” Biological Psychiatry, 61(9), 1081-1089.

[26] Maddock, R. J., et al. (2016). “Acute modulation of cortical glutamate and GABA content by physical activity.” Journal of Neuroscience, 36(8), 2449-2457.

[27] Cotman, C. W., & Berchtold, N. C. (2002). “Exercise: A behavioral intervention to enhance brain health and plasticity.” Trends in Neurosciences, 25(6), 295-301.

[28] Schuch, F. B., et al. (2018). “Exercise as a treatment for depression: A meta-analysis adjusting for publication bias.” Journal of Psychiatric Research, 77, 42-51.

[29] Hendriks, S. M., et al. (2021). “Exercise interventions for social anxiety: A systematic review and meta-analysis.” Journal of Anxiety Disorders, 78, 102350.

[30] Hölzel, B. K., et al. (2011). “Mindfulness practice leads to increases in regional brain gray matter density.” Psychiatry Research: Neuroimaging, 191(1), 36-43.

[31] Bravo, J. A., et al. (2011). “Ingestion of Lactobacillus strain regulates emotional behavior and central GABA receptor expression in a mouse via the vagus nerve.” Proceedings of the National Academy of Sciences, 108(38), 16050-16055.

[32] Streeter, C. C., et al. (2010). “Effects of yoga versus walking on mood, anxiety, and brain GABA levels: A randomized controlled MRS study.” Journal of Alternative and Complementary Medicine, 16(11), 1145-1152.

[33] Craske, M. G., et al. (2014). “Maximizing exposure therapy: An inhibitory learning approach.” Behaviour Research and Therapy, 58, 10-23.