Why You Perform Worse When Someone Better Is Watching You

You've done this a hundred times. You know the material cold. Your track record is solid. And then someone walks into the room — a senior partner, a founder you admire, a mentor you've wanted to impress for years — and something short-circuits. Your thoughts scatter. Your words come out wrong. The competence you've built over years seems to evaporate in real time. And the worst part? You're aware of every second of it happening, which only makes it worse.

This isn't imposter syndrome in the way most people describe it — it's something more specific, more humiliating, and far more common among genuinely capable people than anyone admits. It's not that you feel like a fraud in general. It's that the presence of someone who clearly knows more than you do triggers a cascade that actively dismantles your performance in the moment. Understanding exactly why that happens is the first step to stopping it.

In 1965, psychologist Nicholas Cottrell built on Robert Zajonc's social facilitation research to propose what he called Evaluation Apprehension Theory. The core finding was this: the mere presence of others doesn't uniformly help or hurt performance — what matters is whether those others are perceived as evaluators. When you believe you're being judged by someone competent enough to detect your mistakes, arousal levels spike in a very specific way.

For well-learned, automatic tasks, that arousal can actually enhance output — which is why athletes sometimes perform better in front of a crowd. But for complex, cognitively demanding tasks that require working memory, creative thinking, or nuanced judgment, that same arousal becomes interference. The more sophisticated the task, and the more expert the observer, the harder the performance collapse hits. You're not imagining it. Your nervous system is doing exactly what it was designed to do — it's just doing it at the worst possible moment.

Here's what makes this phenomenon distinct from general social anxiety: it scales with the observer's perceived competence. You can present to a room of fifty strangers and feel fine. But put one person in that room who you believe could genuinely assess whether you're good — and the whole system tips.

Researchers have replicated this across domains. A 2011 study by Gimmig, Huguet, Caverni, and Counti found that chess players made significantly more errors when observed by players ranked higher than them, compared to being watched by lower-ranked players or being alone. The degradation wasn't about nervousness in a general sense — it was specifically triggered by the upward comparison. Your brain isn't asking 'am I being watched?' It's asking 'is the person watching me capable of seeing through me?' That's a completely different threat calculation.

This is why imposter syndrome spikes most violently not in incompetent people — but in people surrounded by genuine expertise. The more credentialed your environment, the more your threat-detection system treats observation as exposure.

When evaluation apprehension fires, your prefrontal cortex — the part responsible for flexible thinking, working memory, and novel problem-solving — gets partially hijacked by the limbic system's threat response. Cortisol and adrenaline flood in. Your attentional spotlight narrows. This is the physiological definition of choking: it's not a failure of skill, it's the skill becoming inaccessible because your cognitive resources have been partly commandeered by threat monitoring.

Sian Beilock's research at the University of Chicago — documented in her book Choke — showed that when experts are asked to think carefully about a well-practiced skill, performance degrades. The expert observer triggers exactly this: you start consciously monitoring what should be automatic, which breaks the fluency. A pianist who starts watching their own fingers mid-performance. A surgeon who becomes aware of their hands. A founder who starts editing every sentence while they're still speaking it.

The cruel irony is that self-monitoring — the very thing your brain deploys to protect you from judgment — is what generates the errors that make the judgment worse. The mechanism designed to help you perform is what causes the collapse.

• ◆Attentional narrowingyour focus collapses inward onto self-monitoring rather than outward onto the task, stealing the cognitive bandwidth needed for fluent performance
• ◆Explicit processingautomatic, practiced skills get 'de-automatized' as you start consciously supervising them, breaking the very fluency that makes them good
• ◆Cortisol interferenceelevated stress hormones compress working memory capacity, making it harder to hold complex ideas in mind or generate novel responses under pressure
• ◆Anticipatory shameyour brain runs worst-case simulations before the task is complete, consuming processing power and creating a self-fulfilling prediction
• ◆Comparison anchoringyou stop performing against your own standard and start performing against your mental model of the observer's standard, which you consistently overestimate

What makes expert-observer collapse uniquely brutal is that it's not just anxiety — it's real-time upward social comparison triggering what Leon Festinger called 'opinion evaluation.' When we can't objectively assess our own ability, we look to social comparison for calibration. The problem is that in the presence of someone visibly more accomplished, we don't compare neutrally — we compare catastrophically.

You don't think 'I'm somewhat less experienced than them.' You think 'they can see exactly how much I don't know.' The comparison triggers what researchers call contrast effects: the other person's competence makes your own feel dimmer by proximity, even when nothing about your actual capability has changed. This is why being in a room with brilliant people can make you feel briefly stupid, even when you're not — and why the most intellectually insecure people at elite institutions are often the smartest people who came from less elite environments. They're not less capable. They're doing more violent upward comparison.

There's a painful paradox here that doesn't get named enough: the more invested you are in being competent, the more devastating expert observation becomes. People with high performance identity — founders, senior professionals, high achievers — have more to lose from being seen as inadequate. Their self-concept is heavily tied to the perception of competence. So the threat isn't just social discomfort. It's identity threat.

This is why you might have noticed that your sharpest, most capable colleagues sometimes go inexplicably quiet or flat when a genuinely impressive external figure enters a conversation. They're not suddenly less intelligent. They've entered a threat state that's incompatible with the cognitive flexibility that makes them good at what they do. The people who seem unfazed in those moments are often either less invested in the outcome — or they've done specific psychological work to decouple their identity from moment-to-moment performance evaluation.

The hard truth is that expertise itself can deepen the vulnerability. The more you know, the more acutely you understand what you don't know. Dunning and Kruger documented the other end of this — the incompetent don't know enough to recognize their gaps. Genuine experts are perpetually aware of their gaps, which means they're perpetually aware of what an observer who knows more could detect.

The standard advice — 'just be yourself,' 'fake it till you make it,' 'don't overthink it' — is worse than useless here because it addresses the surface behavior without touching the mechanism. Telling someone who's already in explicit self-monitoring mode to 'stop overthinking' is like telling someone hyperventilating to 'just breathe normally.' The instruction competes with itself.

What research actually supports is more counterintuitive. Beilock's work found that expressive writing about performance anxiety before a high-stakes event — just ten minutes of putting fears on paper — measurably improved scores under pressure. The hypothesis is that writing offloads the ruminative content from working memory, freeing up cognitive resources for the task. Similarly, pre-performance self-affirmation of core values (not ability affirmations — values) has been shown to reduce the identity-threat component of evaluation apprehension, making the stakes feel lower without requiring you to pretend you're not nervous.

The more durable fix is upstream: systematically decoupling your performance identity from moment-to-moment observer evaluation at the level of belief, not behavior. That's not a reframe you apply in the moment — it's a structural change in how your nervous system codes the meaning of being observed.

• ◆Expressive pre-performance writingten minutes journaling your specific fears before a high-stakes situation offloads anxious content from working memory, documented to improve objective performance scores
• ◆Values affirmation, not ability affirmationaffirming what you care about (not how good you are) reduces identity threat and lowers the psychological stakes of the interaction
• ◆Process focus over outcome focusredirecting attention to specific procedural steps rather than the overall impression you're making reduces explicit monitoring of automatic skills
• ◆Deliberate low-stakes exposurerepeatedly performing in front of progressively more expert observers in contexts where the outcome doesn't matter, building a new association between expert presence and safety
• ◆Reframing the observer's roleresearch on 'challenge vs. threat appraisal' shows that mentally recasting the expert as a resource rather than an evaluator (someone who can help you see what you're missing) shifts the physiological response from threat to challenge arousal, which actually enhances performance

At the bottom of all of this is a belief that most high performers carry without naming it: that their competence is a performance they're giving rather than a reality they inhabit. When no one credible is watching, you can relax into what you actually know. The moment an expert enters the frame, the performance metaphor activates — and then you're not working, you're acting, and you know it.

Resolving this at the root isn't about techniques or cognitive tricks. It's about shifting the foundational identity structure that makes external evaluation feel like a verdict on your worth rather than just information about your current output. That shift doesn't come from insight alone — reading this article might name the problem clearly, but naming it and reprogramming it are very different things. The pattern lives in the subconscious architecture of how you relate to judgment, and that's where the real work has to happen.