Do CO2, VOCs, and Cabin Pressure Cloud Your Thinking?

Moderately high CO2 readings can make a room feel like a cognitive emergency. The best short-term experiment in this evidence base offers a calmer conclusion: healthy adults did not show a broad, systematic decline in cognitive performance under simulated-flight conditions with CO2 up to 4,200 ppm, total volatile organic compounds (TVOCs) up to 2,100 micrograms/m3, and cabin-like pressure of 755 hPa.

That does not make stale, hot, crowded rooms a productivity hack. It means that a CO2 number alone does not reliably predict an immediate drop in thinking ability. Temperature, odors, pollutant mixtures, task type, and exposure duration can all change the result.

The largest trial tested the variables together

Herbig et al. (2026) randomly assigned 398 healthy adults aged 18 to 69 to 11 simulated-flight exposure groups. The experiment varied pure CO2 at 1,200, 2,750, and 4,200 ppm; atmospheric pressure at 940 or 755 hPa; and a selected VOC mixture at low, medium, or high concentrations, reaching TVOCs of 2,100 micrograms/m3.

Participants completed established tests across eight cognitive domains. Only concentration and figural working memory differed significantly between groups. Those differences did not worsen consistently as CO2, VOC concentration, or pressure conditions became more challenging, so the authors described them as non-systematic. The published abstract does not provide the full exposure table, exact VOC mixture, exposure duration, pairwise effect sizes, confidence intervals, or all detailed test results. That prevents anyone from extracting a reliable "brain-fog threshold" from this trial.

Still, its central result matters. This was a randomized experiment that manipulated several cabin-relevant conditions at once, rather than an observational study asking whether people in stuffier rooms happened to perform differently. Within this short exposure window and this healthy adult population, the study found no generalized cognitive collapse. Participants also rated air quality good to very good across groups, so perceived air quality did not neatly track the objectively less favorable conditions.

A reading of 2,000 ppm or 4,000 ppm may indicate limited ventilation or high occupancy. It does not, by itself, prove that everyone in the room has acutely impaired cognition.

Why CO2 is a useful signal but a poor solo villain

CO2 commonly rises when people share an inadequately ventilated space, which is why monitors have become popular. But it often travels with occupant-generated odors, other pollutants, heat, and low air exchange. A 2023 review of indoor CO2 notes that schools, offices, and transportation settings often exceed the commonly discussed 1,000 ppm value and that CO2 can co-vary with other pollutants. In many everyday settings, it works better as a ventilation and occupancy signal than as a stand-alone toxin meter.

A controlled vehicle experiment supports that distinction. In Wang et al. (2024), 25 participants experienced CO2 conditions of 800, 1,800, and 3,500 ppm while researchers assessed driving and cognitive performance. CO2 did not significantly change driving speed, acceleration, lateral control, sleepiness, or most task-load measures. Participants did report greater effort at 1,800 ppm than at 800 ppm. Body odor also affected N-back accuracy, although in this study it significantly increased accuracy rather than reducing it.

That is not a contradiction to common sense. It is a reminder that subjective effort, task performance, and environmental discomfort are separate outcomes. A person can feel that a task takes more work without showing worse performance on that task, and a statistically detectable change in one test does not establish a broad decline in real-world competence.

Heat can turn a modest VOC burden into a different exposure

The strongest reason not to overread the simulated-flight null result comes from a smaller controlled chamber study. Zhao et al. (2025) studied 71 university students under 23 C or 28 C and TVOC concentrations of 100 or 1,000 micrograms/m3. TVOC did not produce a clear main effect overall. Yet at 28 C with 1,000 micrograms/m3 TVOC, accuracy worsened significantly on six of 10 cognitive tasks, and average task error rose 5.2%.

The study does not show that every warm room with VOCs damages cognition. It shows an interaction: the same pollutant concentration can have a different effect under greater thermal stress. That is exactly the kind of result a single-number rule misses. A warm meeting room full of people, fragrances, cleaning-product residues, and stale air is not equivalent to a cool laboratory room with the same CO2 reading.

Aircraft research makes the mixture problem concrete. Yin et al. (2021) measured carbonyl compounds and other VOCs on 56 commercial flights, collecting 350 samples across eight aircraft models. Most highly detected long-chain VOC measurements fell between 0 and 10 micrograms/m3, while individual carbonyl averages ranged from 0.3 to 8.3 micrograms/m3 except acetone and acrolein, which averaged 20.7 micrograms/m3. These measurements cannot establish cognitive effects, but they show why a selected laboratory VOC mixture cannot stand in for every real cabin or room.

Lowering classroom CO2 did not raise short-term grades

Real-world classroom evidence also resists a simple threshold story. In a 2023 between-groups classroom experiment, 201 students provided 527 responses. Improved ventilation reduced CO2 from about 1,100 ppm to below 800 ppm, but it did not improve short-term academic performance. Another environmental intervention did worsen problem-solving ability.

The difference between this field study and the simulated-flight trial is informative rather than fatal to either result. The classroom study measured academic outcomes in an operating educational setting; the flight study used a standardized cognitive battery under controlled conditions. They also involved different populations, exposures, and tasks. Neither study supports the claim that modest CO2 reduction automatically produces an immediate, measurable gain on every cognitive outcome.

What better indoor air can still do

The case for improving indoor air does not depend on selling CO2 monitors as brain scanners. Ventilation can dilute occupant-generated pollutants and reduce airborne-infection risk. Source control can reduce emissions before they enter the room. Filtration can reduce particulate pollutants, depending on the device, placement, and how people use it. A 2024 systematic review and meta-analysis of portable air cleaners found that they can reduce indoor airborne pollutant concentrations, with results shaped by device type and use patterns.

For cabins, the tradeoffs include energy use and recirculation. In a randomized simulated-flight trial, Herbig et al. (2022) tested occupancy and ventilation regimes during a four-hour low-pressure simulated flight, with recirculation increasing to a maximum CO2 concentration of 4,200 ppm. Higher recirculation did not obviously worsen participants' short-term health and well-being measures. That study did not test cognition, but it adds context: cabin air management involves more than maximizing outside air at any cost.

What this evidence cannot answer

The anchor trial does not establish safety or equivalence for children, frail older adults, clinical populations, sleep-deprived travelers, stressed flight crews, or people doing safety-critical work. It also cannot settle what happens over a long-haul flight, repeated exposure across workdays, or cumulative fatigue. The lower-pressure condition was not an independent oxygen or hypoxia experiment, so the available report cannot separate pressure effects cleanly from the other cabin-like features.

Its VOC conclusion is also mixture-specific. Indoor VOCs can come from building materials, cleaning products, outdoor infiltration, and occupants; a given TVOC total does not reveal which chemicals are present or whether they have similar biological effects. TVOC is a useful aggregate measurement, not a chemical identity card.

The practical verdict

Optimize your workspace air, especially when it is hot, crowded, odorous, or visibly under-ventilated. Open windows when outdoor air is suitable, use mechanical ventilation well, address emission sources, manage temperature, and choose filtration for the pollutants you actually need to reduce. These steps support comfort, exposure control, and infection-risk reduction.

But skip the claim that moderate CO2 alone reliably causes acute cognitive damage. The largest randomized simulated-flight study found no systematic broad impairment through 4,200 ppm CO2 under its tested conditions. The more defensible view is less dramatic and more useful: performance responds to the whole room, and the whole room is usually more complicated than one sensor reading.