Cardiac arrest is technically simple to manage. I know that sounds strange, maybe even cynical. But it is true: when a patient has no pulse, is not breathing, and loses consciousness in front of you, the range of decisions narrows almost completely. You start compressions. You attach the defibrillator. You follow the algorithm. You are not choosing, you are executing something you have practiced hundreds of times.
The paradox of critical emergency is this: the more severe the situation, the less you have to decide.
System 1 and System 2: two brains, one patient.
Daniel Kahneman describes human reasoning as the product of two systems operating in parallel [4]. System 1 is fast, automatic, intuitive: it recognizes patterns without conscious effort, draws on accumulated experience, acts before you have finished thinking. System 2 is slow, deliberate, analytical: it compares hypotheses, weighs probabilities, requires sustained attention and costs cognitive effort.
Under normal conditions the two systems alternate and correct each other. In emergency, time pressure, emotional load and the stakes shift the weight toward System 1, for better and for worse. For better, because the expert recognizes and acts in seconds. For worse, because cognitive biases are the offspring of System 1 working without supervision.
The brain under pressure does not work worse. It works differently.
When we face a critical patient, our cognitive system does not collapse, it reorganizes. In extreme emergency, System 1 takes command. Not out of weakness, out of efficiency. The experienced clinician does not evaluate options in sequence: they recognize the situation and act. This is the core of Gary Klein's Recognition-Primed Decision Making (RPDM) model: in high-stakes, time-constrained contexts, experts do not choose between alternatives, they match the situation to a known pattern and activate the associated response [5][8].
Cardiac arrest is the perfect example. AHA/ERC guidelines provide an explicit, sequential, validated algorithm [6]. The protocol is not a limitation, it is a framework that offloads working memory and frees mental resources to manage operational variables: compression quality, vascular access, reversible causes [3][7].
The result? Less cognitive stress, not more. Paradoxically, extreme criticality protects the clinician from uncertainty.
But mountain polytrauma is not cardiac arrest.
This is where the paradox risks breaking down, and it is worth being honest about it.
A polytrauma patient in an alpine environment, with hypothermia, limited resources and a helicopter that cannot land, is not a simple scenario in any sense of the word. Operational complexity is at its maximum. And yet, even in this case, the decision space tends to narrow, not because an exhaustive protocol exists, but because the available resources decide for you. You cannot do everything. So you decide with what you have.
The distinction is crucial: operational complexity increases, decisional complexity decreases.
You are not choosing between ten therapeutic options, you are optimizing the three you have available. This does not make the work easier. It makes it more focused.
Where the brain really breaks down.
The scenario that puts the most pressure on clinical reasoning is not cardiac arrest. It is not even mountain polytrauma. It is the patient with no name.
The patient with worsening dyspnea, a confused history, borderline parameters. The one who could have heart failure, pulmonary embolism, atypical pneumonia, or all three at once. There is no protocol deciding for you. You have to generate hypotheses, weigh them, gather data, reassess. System 2 has no choice, it has to step in.
And that is exactly where acute stress does the most damage. Research shows that under pressure the most compromised cognitive functions are divided attention, working memory and complex reasoning [1]. Cognitive biases, anchoring, premature closure, availability bias, thrive in ambiguous scenarios, not in extreme ones [2][12].
The COVID-19 pandemic offered an involuntary case study: analysis of clinical cases managed during that period documented a surge in diagnostic errors in atypical scenarios, while protocols for severe cases held [12].
What to do with this, concretely.
Awareness of this mechanism is not purely theoretical. It has direct implications for how we prepare and how we act in the field.
First: recognize which scenario you are in. If the patient is critical and the protocol exists, trust the protocol. Do not add complexity where the cognitive system is already optimized for simplicity. Slowing down to "reason better" during cardiac arrest is almost always counterproductive.
If instead the patient is ambiguous, parameters that do not convince, fragmented history, a picture that does not add up, switch modes deliberately. Slow down. Verbalize your hypotheses. Ask yourself what you are taking for granted. Clinical uncertainty must be recognized before it can be managed: the signals are cognitive ("I do not know what is happening"), somatic ("something does not feel right"), emotional ("I feel uncomfortable") [9].
Finally: train both systems. Protocols become automatic through repetition, stress training, high-fidelity simulation [11]. Analytical reasoning is trained through deliberate complexity: ambiguous cases, structured debriefings, controlled exposure to uncertainty [10].
In closing.
The next time you are faced with a patient in cardiac arrest, do not feel guilty if you do not feel panic. It is not pathological emotional detachment. It is your brain working exactly as it should: recognizing, activating, executing.
Worry instead when the patient is the one you cannot figure out. That is where the real cognitive work begins. And there, calm is not automatic, it has to be built.
References
- LeBlanc VR. The Effects of Acute Stress on Performance: Implications for Health Professions Education. Academic Medicine. 2009;84(10 Suppl):S25-33.
- Hartigan S, Brooks M, Hartley S, et al. Review of the Basics of Cognitive Error in Emergency Medicine: Still No Easy Answers. Western Journal of Emergency Medicine. 2020;21(6):125-131.
- Szulewski A, Howes D, van Merriënboer JJG, Sweller J. From Theory to Practice: The Application of Cognitive Load Theory to the Practice of Medicine. Academic Medicine. 2021;96(1):24-30.
- Harris E, Santhosh L. Dual Process Theory and Cognitive Load: How Intensivists Make Diagnoses. Critical Care Clinics. 2022;38(1):27-36.
- Bond S, Cooper S. Modelling Emergency Decisions: Recognition-Primed Decision Making. Journal of Clinical Nursing. 2006;15(8):1023-32.
- Panchal AR, Bartos JA, Cabañas JG, et al. Part 3: Adult Basic and Advanced Life Support: 2020 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation. 2020;142(16 Suppl 2):S366-S468.
- Ford JS, Malhotra A, Pearce AK, Wardi G. CPADS-30: Mastering the First 30 Seconds of Adult Cardiac Arrest Resuscitation. American Journal of Emergency Medicine. 2025;96:166-169.
- Hudson D. Neuroanatomical Basis for Recognition Primed Decision Making. Studies in Health Technology and Informatics. 2013;183:107-10.
- Jain A, Bhoi S, V T A. Emergency Medicine as Risk Management: A Cognitive Scaffold for Safe Decisions Under Uncertainty. Emergency Medicine Journal. 2026.
- Ilgen JS, Dhaliwal G. Educational Strategies to Prepare Trainees for Clinical Uncertainty. New England Journal of Medicine. 2025;393(16):1624-1632.
- Aronson M, Henderson T, Dodd KW, et al. Effects of Brief Mental Skills Training on Emergency Medicine Residents' Stress Response During a Simulated Resuscitation. Western Journal of Emergency Medicine. 2022;23(1):79-85.
- Coen M, Sader J, Junod-Perron N, Audétat MC, Nendaz M. Clinical Reasoning in Dire Times: Analysis of Cognitive Biases in Clinical Cases During the COVID-19 Pandemic. Internal and Emergency Medicine. 2022;17(4):979-988.
