Why You Can’t Focus Under Pressure: The Neuroscience of Attentional Choking in High Performers

In twenty-six years of practice at MindLAB Neuroscience, I have not met a single client whose attention was genuinely broken when it failed under pressure. What broke, reliably, was the handoff. The moment the salience network tagged a situation as high-stakes, control was supposed to transfer cleanly to the central executive network — and it didn’t. The wiring was intact. The calibration was off. That distinction is the entire game, and it is the difference between a capacity problem (which would require something most people don’t actually need) and a calibration problem (which responds to mechanism-targeted intervention).

This article describes the neuroscience underneath that pattern. It explains why high performers — the very people whose brains are optimized for stakes-detection — are also the people who find themselves staring at a familiar task on a high-stakes day and feeling their working memory dissolve. And it lays out what is actually known about retraining the handoff, separating the established mechanisms from the wishful claims that crowd this space.

Why Do You Choke When the Stakes Are High?

You choke when the stakes are high because your salience network — anterior insula and dorsal anterior cingulate — detects the high-stakes signal but stalls before transferring control to the central executive network. The salience signal stays elevated, the executive handoff never completes, and focus fragments because two networks fire in conflict.

The architecture is itself well-mapped. The salience network’s job is to monitor internal and external signals for importance and then switch the brain into the appropriate mode — task-focused executive control, internally-oriented default mode, or threat-vigilance. In ordinary cognition, the switch operates so smoothly that you never notice it. Under high stakes, the switching circuit is exactly where the failure occurs.

Hermans and colleagues showed that acute stress triggers a large-scale reallocation of brain resources toward the salience network — anterior insula, dorsal ACC, amygdala — at the expense of the executive control network. The reallocation is mediated by catecholamines and corticosteroids, and it is not subtle: under stress, the brain systematically downshifts prefrontal engagement and upshifts salience and reactive subcortical processing. This is the systems-level substrate beneath the lived experience of pressure choke.

Consider a young professional facing the first major pitch of their career. They have rehearsed the deck for weeks. They know the material cold in a quiet room. The moment they walk into the conference room and register the senior partners watching, the salience network does its job — it tags this as a high-stakes situation. What is supposed to happen next is a clean handoff: the central executive takes over, working memory loads the rehearsed plan, the dlPFC sustains the goal, and the talk runs. What happens instead is that the anterior insula keeps firing, the executive network never fully comes online, and the rehearsed plan is right there — visible at the edge of cognition — but unreachable.

The mechanism is not unique to public performance. It shows up the same way in a difficult medical conversation, a critical email that has to be drafted in twenty minutes, an interview that decides a board vote, a chess match where one move determines the season. What varies is the trigger; what stays constant is the circuit.

What Happens in the Brain When You Can’t Focus Under Pressure?

Under pressure, the locus coeruleus floods the prefrontal cortex with norepinephrine, pushing arousal past the Yerkes-Dodson optimal. Alpha-1 adrenergic receptors saturate in the dlPFC. Working-memory representations collapse. The anterior cingulate registers the resulting conflict but cannot resolve it, because the executive circuitry needed to resolve it has already gone offline.

The Yerkes-Dodson curve has been a textbook fixture for over a century, but the circuit-level mechanism beneath it is recent. Beerendonk and colleagues’ 2024 paper provides converging human evidence — using pupil-indexed arousal as the readout — for the inverted-U shape and proposes a disinhibitory interneuron circuit that explains it. At low arousal, cortical signal is weak; at moderate arousal, the disinhibitory circuit selectively boosts task-relevant processing; at high arousal, the same circuit becomes saturated and indiscriminate, so signal and noise amplify together. Performance peaks in the narrow band where the circuit is engaged but not saturated.

Norepinephrine is the proximate driver. The locus coeruleus — a small brainstem nucleus — projects noradrenergic fibers across the entire cortex. At moderate firing rates, NE binds primarily to high-affinity alpha-2 receptors that enhance prefrontal function. As firing rises, NE recruits the lower-affinity alpha-1 receptors. Alpha-1 binding suppresses dlPFC activity. At very high arousal, alpha-1 saturation collapses the working-memory representations that the executive network was holding — which is the moment the rehearsed plan goes blank.

The anterior cingulate cortex sits in the middle of this and registers what is happening. The ACC’s job is conflict monitoring — detecting when goals and outputs are out of alignment and signaling that more cognitive control is needed. Under choke, the ACC dutifully fires the conflict signal. But the dlPFC, which would normally translate that signal into resolution, is offline. The result is a felt experience of knowing something is wrong while being unable to fix it. That gap is not a failure of will; it is a circuit that has lost the ability to act on its own conflict alarms.

A telling human-clinical demonstration of this pattern comes from surgical training. Modi and colleagues showed that under matched time-pressure tasks, surgeons whose performance held steady displayed sustained ventrolateral and dorsolateral prefrontal activation, while surgeons whose performance declined showed prefrontal deactivation. Same task. Same demand. Two opposite neural signatures — and the difference predicted who choked. This is the architecture I am describing, captured in a real high-stakes setting.

"The gap between knowing something is wrong and being able to fix it is not a failure of will. It is a circuit that has lost the ability to act on its own conflict alarms."

A burnt-out executive in their early fifties presents this picture in a particular way. The ACC’s conflict signal has been firing for years, often unrecognized as such, while the dlPFC has been incrementally less able to resolve it. By the time they arrive at MindLAB Neuroscience, the choke moments have stopped being occasional — they have become a default. The fix is not stress-management language. It is to recalibrate the noradrenergic operating point and rebuild the executive engagement that has been chronically dampened.

How Do Elite Performers Maintain Focus Under Stress?

Elite performers don’t suppress arousal — they maintain the salience-to-executive handoff under arousal. Their prefrontal cortex stays engaged when other people’s prefrontal cortex shuts down. The neural signature is sustained dlPFC activation under matched task demand, not lower arousal. The mechanism is calibration: the operating point on the inverted-U is shifted, not flattened.

The Modi surgical study is one of the cleanest demonstrations of this in the literature, and it is worth sitting with. The two cohorts of surgeons faced the same task under the same time pressure. The “stable” cohort sustained prefrontal engagement; the “decline” cohort showed prefrontal deactivation. Neither cohort was operating at low arousal. The stable surgeons were not calmer — they were better calibrated. The disinhibitory circuit Beerendonk’s group described was holding the signal-amplification window open longer for them.

This reframes a great deal of common advice about pressure performance. Telling someone to “stay calm” misidentifies the lever. The lever is the operating point on the arousal curve, not the arousal level itself. Cools and Arnsten’s 2021 review of monoaminergic neuromodulation in primate prefrontal cortex makes this explicit: “Most neuromodulators have a narrow inverted-U dose response, which coordinates arousal state with cognitive state.” Recalibration moves the whole curve, not the moment-to-moment level.

Consider a different scenario. A partner in their mid-forties is chairing a charity board meeting where a difficult endowment decision must be made in the next ninety minutes. There is no boardroom. There is a kitchen table, two adult children moving in and out of the room, a phone vibrating with another family commitment, and a dossier to be processed before the call. The stakes are high — relationally, financially, reputationally — and the salience load is dispersed across multiple domains rather than concentrated in a single arena. This is a non-corporate variant of the same neural problem: the salience network is firing on too many inputs, the executive handoff is fragmenting, and “trying harder” makes it worse, because trying harder pushes NE further past optimal.

The recalibration that elite performers use, whether they articulate it or not, is to shape the salience input and load the noradrenergic system progressively. They reduce the salience input distribution that would push them past optimal. They build, over years, a higher tolerance to arousal before alpha-1 saturation kicks in. And — critically — they treat the live high-stakes moment as the rep, not the test. The training happens in the moment that fires the circuit, not in a quiet room afterward. This is the neural mechanism behind every credible pressure-performance method I have seen work, across athletes, surgeons, traders, and leaders.

Is Choking Under Pressure a Sign of ADHD or Anxiety?

Choking under pressure is not the same as ADHD or anxiety, though they share surface features. ADHD involves a lifelong catecholaminergic signature in cognitive-control circuitry that develops differently from baseline. Anxiety involves chronic amygdala-driven threat detection. Situational choking involves an acute, calibration-specific salience-network hyperactivation in an otherwise-intact attention architecture.

The distinction matters because the interventions diverge. ADHD is a neurodevelopmental pattern with a stable circuit-level signature, often present from childhood, and best evaluated by a qualified clinician. The cognitive-control machinery is structurally different, not miscalibrated by recent experience. Friedman and Robbins’ 2021 review of prefrontal cortex in cognitive control documents how working memory updating, set shifting, and inhibition are dissociable executive-function components — each with partly distinct circuit substrates that can be selectively disrupted. ADHD’s signature spans multiple components; situational choking does not.

Anxiety is a different circuit story. Eysenck’s attentional control theory frames anxious attention as a chronic pull-toward-threat — the goal-directed attentional system loses ground to the stimulus-driven system because the threat-detection circuit holds priority too aggressively. The mechanism overlaps with the salience-network hyperactivation seen in choking, but with a critical difference: anxiety’s salience hyperactivation is trait-like and present at baseline, whereas choking’s is state-like and emerges only under specific high-stakes triggers. A person can have an entirely normal baseline attention profile and still choke in the live moment. A person with anxiety carries the hyperactive salience tag everywhere.

The phenotypic overlap is real, and it is the source of a common confusion. Many high-performing professionals present at MindLAB Neuroscience convinced they have either undiagnosed ADHD or generalized anxiety. Sometimes that is correct. More often, the actual mechanism is acute attentional choking in a brain that is otherwise functioning at the upper end of normal — and the intervention path is different. ADHD and clinical anxiety warrant evaluation by a qualified clinician; attentional choking warrants recalibration, not assessment for a condition.

The differential question is not “do these conditions feel similar” — they do — but “what is the temporal pattern of the signal?” Trait conditions persist across contexts. State choking emerges in specific arousal regimes. The two have different mechanisms, different timelines, and different paths forward.

Can You Train Your Brain to Perform Better Under Pressure?

Yes — the salience-to-executive handoff is plastic and recalibrates with progressive exposure. Stress-inoculation methodologies graduate the salience load until the executive network completes the transfer reliably under arousal. The training works because the circuit is most plastic in the live high-stakes moment, where the misfire originates, rather than in a neutral room afterward.

The empirical literature on choking interventions is more developed than the popular discourse suggests. Gröpel and Mesagno’s systematic review of 47 empirical studies in sport identified three intervention classes that consistently work: pre-performance routines that anchor attention before the salience load arrives, attention-redirection techniques such as quiet-eye training that recruit specific perceptual circuits during the moment, and acclimatisation training — graduated exposure to pressure in conditions that mimic the eventual stakes. The same review found null or negative results for goal-setting, neurofeedback, and reappraisal cues used in isolation. Training is not a generic category. The mechanism has to match the failure mode.

Stress-inoculation as a methodology — the Meichenbaum-derived approach that systematically grades stressors from low to high — works because it loads the noradrenergic and salience systems in calibrated steps. The trainee’s executive network is asked to complete the handoff under progressively heavier salience input. Each successful completion strengthens the handoff. Eventually the operating point on the inverted-U shifts, and the executive network holds under conditions that previously collapsed it. This is plain neuroplasticity: a circuit becomes more reliable with repetition under appropriate load.

MindLAB Neuroscience’s Real-Time Neuroplasticity™ methodology applies this same logic to attentional choking specifically. The intervention occurs in the live high-stakes moment, when the salience network is actively firing past optimal — which is the window in which the handoff circuit is most malleable. Working in the live moment rather than retrospectively is what produces the durable shift. This is the one mechanism RTN brings to this topic; it is not a generic pressure protocol, and there is no branded “stress-inoculation protocol” that would apply across every domain. The methodology is the methodology; the topic dictates the specific application.

What this looks like in practice is a paced reload of the executive system under increasing salience input. A young professional might begin with low-stakes presentations to a single observer, advancing through progressively more consequential audiences as the handoff stabilizes. An executive might begin with simulated high-stakes decisions in a controlled environment, advancing into the live boardroom as the noradrenergic operating point recalibrates. A partner managing complex family systems might begin by training the salience filter on a single domain at a time, progressively reintegrating the multi-domain load. The protocol design varies; the underlying mechanism — graduated load, executive reload, live-moment plasticity — does not.

What the First Conversation Looks Like

When someone reaches out to MindLAB Neuroscience about pressure choke, the first conversation is not a symptom inventory. It is a calibration mapping — I want to understand where on the arousal curve your executive system is going offline, what is loading the salience network beyond its current tolerance, and whether what you’re describing is a state problem (choking) or a trait condition that warrants different evaluation. We identify which arena is the highest-leverage training ground, what the noradrenergic substrate is doing, and which of the three established intervention classes fits your specific failure pattern. Most clients leave the strategy call with a clearer picture of their own pressure response than they have had in years.

FAQ