Inert Gas Narcosis

What Really Happens to Your Brain at Depth

Inert gas narcosis is a reversible alteration of central nervous system function caused by breathing inert gases at elevated partial pressures under hyperbaric conditions. It causes confusion, poor judgment, and sometimes a false sense of confidence. The martini analogy — one cocktail per 10 metres — is wrong in a specific and important way: alcohol alters brain chemistry through metabolism. Narcosis is physical. Nitrogen dissolves into nerve membranes under pressure and interferes with how the brain processes information. And you can feel fine — or even good — while being meaningfully impaired.

What Is Actually Happening in the Brain

The primary site is the myelin sheath of neurons — approximately 70–80% lipid by dry weight, one of the most lipid-rich tissues in the body. The Meyer-Overton hypothesis links narcotic potency to lipid solubility: nitrogen is moderately lipid-soluble, dissolves into the myelin under pressure, and physically alters the geometry of voltage-gated Na⁺/K⁺ ion channels. This slows axonal conduction and progressively impairs higher neural functions.

Later research shifted toward protein interactions: gases appear to bind with hydrophobic pockets within receptor proteins, producing allosteric modulation. The onset follows a sigmoidal pressure-effect curve rather than a linear one, which is consistent with receptor saturation rather than simple membrane flooding.

Neurotransmitter effects add another layer. Hyperbaric exposures alter dopamine, norepinephrine, and nitric oxide pathways in animals, with region-specific effects in the hypothalamus and striatum. Inhibiting nitric-oxide synthase in rats reduced behavioural narcosis signs, suggesting NO-mediated signalling plays a role.

The CFFF Test and the Biphasic Response

The Critical Flicker Fusion Frequency (CFFF) test is the most reliable tool for measuring narcosis objectively. It measures the frequency at which a flickering light appears to stop flickering — a direct test of cortical integration speed, which is precisely what narcosis degrades. CFFF was developed for diving research by Peter Bennett (founder of DAN USA) in 1960 and correlates with EEG activity and subjective mental state.

What CFFF data reveal is a biphasic response the martini analogy completely misses. In the first few minutes at depth, scores rise above baseline — oxygen activates dopamine and glutamate synthesis, briefly improving apparent performance. After 15–20 minutes, scores fall below baseline as nitrogen's inhibitory load builds and overtakes the initial stimulation. And surface impairment persists at least 30 minutes post-dive — ascending does not immediately restore normal cognition.

Hemelryck et al. (2013) demonstrated this pattern across chamber, pool, and open-water exposures. Divers who completed a 30-metre dive showed lower-than-baseline CFFF scores for at least 30 minutes after surfacing. The implication for post-dive activity — driving, supervising new divers, making logistical decisions — should be obvious.

Why Simple Tests at Depth Miss the Problem

Motor skills and simple verbal tasks (finger counting, arithmetic on slates) are the last functions lost to narcosis. By the time a diver fails a simple task, executive function, novel problem-solving, and information processing speed are already substantially degraded.

A moderately narcotised diver can correctly count fingers, read a gauge, and perform over-learned procedures while being completely unable to improvise a solution to an unfamiliar problem. This is the same mechanism that lets a moderately drunk person walk to the kitchen. Over-learned tasks survive because they run from procedural memory in the basal ganglia, not from the prefrontal cortex — the first area narcosis compromises.

CFFF has a key advantage over these tests: it requires no motor skill and cannot be passed by rehearsed responses. It also requires a pre-dive individual baseline, without which no meaningful decline can be detected. This is why "I feel fine at 40 m" is not useful evidence of tolerance.

Decision-Making at Recreational Depths

Ahti and Wikgren (2023) used the Iowa Gambling Task (IGT) in an open-water study with 22 divers. The IGT involves 100 card selections across four decks; participants gradually learn which decks are advantageous. It tests prefrontal cortex function — the region affected by alcohol and hypothesised to be impaired by narcosis.

Mean IGT score at 30 m was 1,584.5 (SD 436.7), compared to 2,062.5 (SD 584.1) at 5 m. The difference was statistically significant. Age, BMI, gender, number of previous dives, and self-reported depth comfort did not affect performance. This was the first study to measure narcosis-related decision impairment using the IGT in open water.

The finding matters because 30 m is conventionally cited as the depth where serious narcosis begins. The study shows that safety-critical executive function — the kind needed to assess an emergency, decide to abort, or evaluate a buddy's state — is already compromised at a depth most recreational divers reach regularly.

The Equivalent Narcotic Depth

You can quantify the narcotic load of any gas mix at any depth using the Equivalent Narcotic Depth (END) formula:

END (m) = (PN₂_current / 0.79 − 1) × 10

Where PN₂_current is the partial pressure of nitrogen in bar at the diving depth.

Example: EAN32 at 40 m. Absolute pressure is 5 bar. FN₂ in EAN32 is 0.68. PN₂ = 0.68 × 5 = 3.40 bar. END = (3.40 / 0.79 − 1) × 10 ≈ 33 m.

Breathing EAN32 at 40 m is narcotically equivalent to breathing air at 33 m. A 2019 double-blinded hyperbaric chamber study by Lafère et al., using CFFF and cognitive testing, confirmed measurably less narcosis on EAN32 at 40 m versus air at 40 m. This validates the END formula as a practical planning tool, not just a theoretical one.

Helium is non-narcotic. Adding helium (trimix) reduces narcosis in direct proportion to the helium fraction, which is the primary reason trimix is used beyond approximately 40 m.

Gas Density and Hypercapnia: A Confounding Problem

Air at 40 m (5 bar absolute) has a density of 1.29 × 5 = 6.45 g/L. The recommended maximum gas density is 6.2 g/L (Mitchell and Doolette). Above this threshold, respiratory muscles approach their ventilation limit under moderate exertion — CO₂ retention becomes possible even with maximum respiratory effort.

Air at 40 m already exceeds this threshold. Narcosis and CO₂ retention can coexist and are difficult to distinguish. Both cause confusion and poor judgment. CO₂ also directly potentiates narcosis: elevated PaCO₂ increases cerebral blood flow, delivering more dissolved nitrogen to the brain per unit time. A diver working hard at 40 m on air is simultaneously experiencing denser gas, greater respiratory dead space, and enhanced narcotic delivery. Skip breathing worsens all of this.

DAN Event Patterns

From DAN's cumulative incident data, inert gas narcosis rarely causes accidents on its own. It typically acts as a precursor or contributing factor in mishandled descents, task overload, and decompression errors. The recurring causal chain: increased depth drives elevated inert gas partial pressure; CNS effects degrade situational awareness and decision-making; operational error leads to excessive gas consumption, buoyancy loss, or missed decompression stops; physiological stress then compounds the risk.

DAN case compilations link accidents at depths above 30 m where nitrogen was the primary diluent. Inexperienced divers and those with rapid descent profiles show higher rates of control and judgment errors at these depths. Human factors dominate consistently: cognitive underestimation of risk, adaptation myths that lead to over-extension of depth, task narrowing where narcotised divers fixate on a single problem and stop monitoring gas and time, and peer pressure — DAN data link buddy-pair errors to a more experienced diver reassuring a novice against aborting a deep dive.

Practical Notes

Narcosis becomes measurable at approximately 20–30 m; executive function is already compromised at 30 m on air. Simple tests are unreliable — they test the last functions to fail. CFFF impairment persists 30+ minutes after surfacing. Gas density at 40 m on air exceeds the 6.2 g/L threshold; narcosis and hypercapnia can coexist and compound each other. The END formula lets you compare gas mixes; nitrox offers a meaningful narcosis reduction at depth. For complex or high-stakes dives beyond 40 m, use trimix to keep END within task-appropriate limits. If you feel off at depth, ascend 5–10 m, reduce task load, and fall back on over-learned procedures.

References

• Ahti PA, Wikgren J. Rapture of the deep: gas narcosis may impair decision-making in scuba divers. Diving Hyperb Med. 2023;53(4):306–312. doi:10.28920/dhm53.4.306-312
• Hemelryck W, Germonpré P, Papadopoulou V, Rozloznik M, Balestra C. Long term effects of recreational SCUBA diving on higher cognitive function. Scand J Med Sci Sports. 2013. PMID 22476770
• Lafère P, Balestra C, Hemelryck W, Germonpré P, Virgili F. Effect of recreational scuba diving on cognitive performance: An exploratory study. Diving Hyperb Med. 2019.
• Mitchell SJ, Doolette DJ. Recreational technical diving part 1: an introduction to the activities and the risks. Diving Hyperb Med. 2013;43(4):207–214
• Smith RA, Spiess M. The two faces of Eve: gaseous anaesthesia and inert gas narcosis. Diving Hyperb Med. 2010.
• Van Wijk CH, Meintjes WAJ. Cognitive and psychomotor changes during repetitive SCUBA dives. Undersea Hyperb Med. 2014.

Train With Me

Narcosis management and the transition to trimix are a core part of every deep technical course I teach. Enquire about training →