Why Thinking About How You're Doing Makes You Do It Worse

You've done this a hundred times. In practice, alone, or in low-stakes moments, it flows. You speak clearly, perform sharply, move with ease. Then the moment actually matters — a board presentation, a difficult conversation, a competition, a first date with someone you genuinely like — and something short-circuits. Not nerves exactly. Something stranger. You become aware of yourself in a way that feels almost clinical, like watching a recording of your own face while trying to hold a conversation. And the awareness itself is what destroys you.

This isn't a confidence problem. It isn't imposter syndrome in the traditional sense. It's something more specific, more neurological, and significantly more interesting — and once you understand the mechanism, you'll stop trying to fix the wrong thing.

Psychologists Richard Masters and Jon Maxwell developed what's known as Reinvestment Theory in the 1990s — and it remains one of the most underappreciated explanations for why skilled people choke. The core idea is this: when you first learn a complex skill, it requires conscious, explicit attention. You think about each component. A tennis player thinks about grip, stance, swing, follow-through. A speaker thinks about pacing, eye contact, breath. This is procedural knowledge being built from scratch.

But as competence develops, that knowledge gets automated. It moves from the prefrontal cortex — the slow, deliberate, verbal part of your brain — into procedural memory networks, where it runs fast, fluid, and largely unconscious. This is why experts make things look effortless. They're not thinking less because they care less. They're thinking less because the skill has been encoded at a level that doesn't require conscious supervision.

The catastrophe happens when a high-stakes moment triggers what Masters called 'reinvestment' — the conscious mind reaching back in and trying to manually supervise a process that runs better on autopilot. You start narrating your own performance while performing it. The prefrontal cortex, helpful for planning and terrible for real-time execution, suddenly becomes a backseat driver. And the car swerves.

The trigger isn't the high stakes themselves. Research by Sian Beilock at the University of Chicago found something telling: expert performers choke specifically because they have more automated skill to disrupt. Novices don't choke in the same way — they're still performing consciously anyway, so additional monitoring doesn't make things significantly worse. Experts, by contrast, have built intricate, efficient subconscious programs. When conscious attention re-enters the loop, it jams the signal.

Beilock's studies on golfers found that when skilled players were asked to monitor a specific aspect of their swing — to think explicitly about what their knee was doing, or where their elbow was — their performance deteriorated sharply. The same players performed fine under time pressure or distraction, because those conditions actually prevented over-monitoring. The enemy wasn't pressure. The enemy was self-directed attention.

This maps onto something you've probably experienced. Your performance doesn't degrade when you're slightly distracted or in a light flow state. It degrades when you're watching yourself. When you're evaluating your delivery mid-sentence. When you're asking, 'Is this landing? Am I coming across well? Did that sound strange?' — while simultaneously trying to perform the thing you're asking about.

There's a cruel irony here. The more competent you are, the more deeply your skills are automated — and the more damage self-monitoring inflicts when it kicks in. High achievers also tend to have higher dispositional reinvestment: a measurable personality trait describing the extent to which someone is inclined to consciously control their own movements, words, and behavior. Masters' Reinvestment Scale identifies this as a consistent individual difference, and high performers — analytical, conscientious, self-aware — score higher.

The same traits that make you rigorous in preparation make you dangerous to yourself under execution. You've trained yourself to be thoughtful, evaluative, precise. Those are extraordinary capacities in a planning room. In the moment of performance, they become a wrecking ball.

There's also a social component. Research on evaluation apprehension — the anxiety produced by knowing you're being watched and judged — shows it activates the same neural threat circuitry as physical danger. Your brain doesn't distinguish cleanly between 'this presentation could go badly' and 'I am under physical threat.' The stress response that floods your system is real, physiological, and specifically designed to narrow your focus onto survival. Which is the opposite of what nuanced performance requires.

Reinvestment-driven performance breakdown shows up differently depending on the domain, but the signatures are recognizable once you know what to look for:

Several patterns recur across domains and people:

• ◆The mid-sentence dissolutionYou begin a point confidently, then become aware you're making it, and the second half of the sentence loses coherence. The monitoring interrupts the generation.
• ◆The fluency dropSkills that feel smooth in private become effortful in public. A musician's fingers feel strange. A speaker's voice sounds foreign to themselves.
• ◆The recursive loopYou become aware of your awareness. You notice you're self-monitoring, which makes you self-monitor the self-monitoring. The loops compound.
• ◆The retroactive replayAfter the performance, you dissect every moment with surgical precision, treating it as evidence of fundamental inadequacy rather than as a temporary disruption of an automated process.
• ◆The phantom audienceYou continue performing for imaginary evaluators even when alone, rehearsing, or in genuinely low-stakes contexts. The internal observer is always present, always watching.

One of the more disorienting realizations for high performers is that the internal critic they've always treated as a quality-control mechanism is often producing the very errors it's designed to prevent. The voice that says 'make sure this sounds intelligent' is what makes you stumble over your words. The observer that demands 'perform well' is the mechanism of poor performance.

Psychologist Timothy Gallwey articulated this in The Inner Game of Tennis with elegant simplicity: there is a Self 1 — the conscious, analytical narrator — and a Self 2 — the body's accumulated intelligence, the automated skill system. Self 1 doesn't trust Self 2. It intervenes. It corrects in real time. And every intervention costs you. Gallwey's insight, decades before the neuroscience caught up, was that performance excellence isn't about trying harder — it's about learning to silence Self 1 long enough for Self 2 to do what it already knows how to do.

The problem is that for most high achievers, Self 1 has been running so long and so loudly that it feels synonymous with identity. Telling someone to 'get out of their own head' isn't enough. The head-occupying isn't random. It's a learned response with deep roots — often tracing back to environments where conscious self-monitoring was necessary for safety, approval, or survival.

The conventional advice — 'just relax,' 'stop overthinking,' 'be present' — fails because it gives Self 1 another job to do. 'Be present' becomes another instruction to monitor your own presence. 'Stop overthinking' becomes a thought about thinking. The directive lands in the exact cognitive system you're trying to quiet, and that system dutifully tries to comply by generating more internal commentary.

What the research actually supports is more counterintuitive:

• ◆Attentional redirectionRather than suppressing self-monitoring, redirect conscious attention outward. Focus on a specific external cue: a listener's left eye, the texture of an object in your hand, the rhythm of your own breath as sensation rather than evaluation. This occupies Self 1 without asking it to shut up.
• ◆Pre-performance chunkingBeilock's research found that skilled performers who wrote about their worries immediately before a high-stakes event performed significantly better. The act of externalizing the anxious internal monologue appears to decompress its hold on working memory, freeing cognitive resources for the actual task.
• ◆Process-focus over outcome-focusExplicitly orienting toward task-relevant processes ('what am I communicating?' rather than 'how am I coming across?') redirects monitoring toward productive targets. The monitoring doesn't disappear; it gets aimed somewhere useful.
• ◆Practiced exposure under simulated pressureBeilock also found that repeatedly practicing under evaluative conditions builds what she calls 'practiced attention' — the skill of performing while being observed, which itself becomes automated over time. Avoiding high-stakes situations to spare yourself discomfort does the opposite of helping.
• ◆Acceptance rather than suppressionTrying to eliminate the internal observer increases its power. Research on ironic processing (Wegner's white bear studies) confirms this: the harder you try not to think about something, the more it activates. Acknowledging the observer — 'yes, I'm aware I'm being watched, and I'm continuing anyway' — consistently outperforms suppression attempts.

Reinvestment-driven choking is a surface symptom. Underneath it is almost always a belief — implicit, subconscious, rarely examined — that the automated self cannot be trusted. That without conscious supervision, something will go wrong. That the skill, the intelligence, the capability that functions so well in private will be exposed as insufficient the moment the stakes are real.

This belief didn't appear from nowhere. For most high achievers, there was a period — often in childhood or early professional life — when their environment genuinely required hyper-vigilant self-monitoring. Maybe approval was conditional. Maybe mistakes were punished visibly. Maybe being watched felt dangerous, and being watched performing felt existential. The brain learned: when the stakes are high, supervise everything. That was an adaptive response then. It's a performance-destroying one now.

The work, ultimately, isn't technique. It's rebuilding trust in your own competence at the level where that competence actually lives — below conscious awareness, in the body, in the accumulated thousands of hours that don't disappear just because someone important is watching. The skill is there. The observer just doesn't believe it yet.