I am a professional Chemist whose practice includes occupational health and safety and risk assessment. I did my first indoor air assessment involving “sick building syndrome” in 2001 and have done countless indoor air assessments in the last two decades. For those unfamiliar with the topic, the big factor in sick building syndrome is elevated indoor carbon dioxide concentrations. As such, I was quite interested when I read the headline: Rising carbon dioxide levels will make us stupider – If allowed to soar unchecked, greenhouse-gas emissions will interfere with people’s ability to think at Nature.com.

I accessed the underlying paper in GeoHealth titled: Fossil fuel combustion is driving indoor CO₂ toward levels harmful to human cognition (Karnauskas, Miller and Schapiro, 2020 pre-print here) and was disappointed with what I found. The paper really could have benefited from a peer-review by someone better informed with the occupational health and safety literature on the topic of carbon dioxide’s effects on human cognition.

I previously had issues with outsiders getting confused about this topic and four years ago wrote a blog post discussing the state of the literature, at the time, on the topic of carbon dioxide toxicity and its effect on human cognition and performance.

Where the authors of the current article struggle is confusing the human response to acute exposure to elevated carbon dioxide concentrations with chronic exposures. A quick examination of the reference list for the paper shows that the studies cited deal with acute changes in carbon dioxide concentrations but the article is purportedly looking what would happen with a chronic increase in atmospheric carbon dioxide concentrations.

As any health and safety practitioner can explain, elevated carbon dioxide concentrations represent a health hazard. As I detail in my previous post:

• At high concentrations carbon dioxide is an asphyxiant that displaces oxygen and kills affected individuals.
• At moderately high concentrations it has both short-term and long-term toxic effects.
• At moderate concentrations it is tolerable as your body will adapt to the exposure.

For the purposes of this post we can ignore carbon dioxide as an asphyxiant (any concentration over about 30,000 ppm). At moderately high concentrations, carbon dioxide toxicity is based on the perturbation of the acid/base balance in your blood resulting in acidosis. The initial human response is cellular buffering that occurs within minutes-to-hours. In the continued presence of elevated carbon dioxide renal compensation occurs over around 35 days. What this means is your body will adapt to higher concentrations, over time.

Carbon dioxide is also a potent vasodilator and as anyone who has migraines knows, vasodilators are also associated with headaches. That is the context in which most individuals experiencing sick-building syndrome experience carbon dioxide toxicity. It is the short term (acute) exposure of elevated concentrations in poorly ventilated rooms.

The following references all discuss how elevated carbon dioxide concentrations (Erdmann, Steiner and Apte, 2002 Apte et al, 2000, Wargocki et al, 2000, Seppanen, Fisk and Mendall, 1999) effect human health. All point out that a proportion of the human population reacts poorly to daily variations in carbon dioxide concentrations. This is understandable since, as a vasodilator, it would be expected to have particular effects on people prone to migraines or headaches.

The reason students perform poorly on tests under these conditions is because their brain chemistry is being acutely effected by the vasodilator. People perform tasks less effectively when their heads are pounding and they are tired (the major consequences of exposure to vasodilators).

This is where the paper gets the science wrong. The authors infer that the poor performance, due to acute exposure, will somehow extrapolate to poor performance over the long term. Except, the human body adapts.

I had a long twitter exchange with one of the authors which concluded with her making this statement:

Our fundamental disagreement is that you take the adaptation to be fact, whereas I think there is a relatively small amount of evidence for it, so more work is needed.

This line was the one I found the most confounding of our exchange. The human ability to adapt to compounds like vasodilators and vasoconstrictors is well understood. The classic example of the latter is my daily coffee. Coffee is the other side of the coin to carbon dioxide but is a useful example because so many people are familiar with the effects of coffee on human physiology.

Any long-term coffee drinker knows that their body becomes adapted to coffee. You don’t quite feel right in the morning before your first cup and if you are a heavy drinker then going cold-turkey can be a real headache. I was a two pot-a-day drinker in grad school and after completing my thesis I tried to go cold-turkey. The result was horrible. It felt like my brain was going to melt out my ears, the pain was so intense. I talked to a medical professional who explained that my body had adapted to large doses of coffee and if I wanted to stop I had to wean myself off the beverage slowly over time.

Now unlike coffee drinkers, there aren’t many people who experience prolonged exposure to high concentrations of carbon dioxide. That being said there are two different classes of individuals who have undergone this very unusual experiment: astronauts and submariners.

Both astronauts and submariners spend prolonged periods of time in conditions where it is not possible to completely scrub carbon dioxide down to background concentrations. In the International Space Station (ISS) concentrations are typically in the 2000 ppm – 4000 ppm range and submariners typically encounter concentrations averaging 3500 ppm to 4100 ppm. What they found studying submariners is that during an 11-day cruise, submariners adapt to high CO2 levels, as evidenced by the significant dependence of respiratory disturbance index on CO2 during the final but not initial days of the cruise.

The strange thing about the Karnauskas, Miller and Schapiro paper is that they do acknowledge the existence of this literature based on this citation:

However, null or nonmonotonic effects have been found even in these demanding tasks for two special populations—submariners (Rodeheffer et al. 2018) and astronaut–like subjects (Scully et al. 2019). These results suggest that factors like increased experience with demanding cognitive tasks or physiological adaptation to increased ambient CO2 could potentially mitigate the harmful effects of CO2 on cognition.

In writing this they appear to miss the conclusion of the papers. As presented in Rodeheffer et al (Acute Exposure to Low-to-Moderate Carbon Dioxide Levels and Submariner Decision Making)

In conclusion, our findings failed to replicate the impaired decision-making performance reported by Allen et al. and Satish et al. during acute exposures to CO2 at 2500 ppm; however, our results are in concurrence with more recent research reporting null effects at low-to-moderate levels of CO2 on both the SMS test and on traditional measures of cognitive and neurobehavioral function. Overall, results suggest that there is no effect of acute CO2 exposure on submariner decision-making performance at levels routinely experienced during submarine operations.

As for Scully et al, 2019 (Effects of acute exposures to carbon dioxide on decision making and cognition in astronaut-like subjects) it reported:

at higher CO₂ concentrations performance was similar to or exceeded baseline for most measures. These outcomes, which conflict with those of other studies, likely indicate differing characteristics of the various subject populations and differences in the aggregation of unrecognized stressors, in addition to CO₂, are responsible for disparate outcomes among studies. Studies with longer exposure durations are needed to verify that cognitive impairment does not develop over time in crew-like subjects.

These two studies appear to contradict the Karnauskas, Miller and Schapiro article citing them. The cited articles clearly state that adapted populations show no detectable deleterious effects. Remember, submarines have operated for over a century and we have been operating in space for over 50 years. If the premise of the Karnauskas, Miller and Schapiro paper were correct then it is unlikely that submariners would be put in charge of nuclear weapons or astronauts would be able to safely operate their complex and incredibly dangerous ships.

If negative effects were to be observed then after 100 years of submarine operation we would likely have seen them by now.

If as suggested in the article “reduction in cognitive function score statistically significant, they were typically rather large—on the order of tens of percent decrease in performance per ~400 ppm CO2 increase” then wouldn’t that mean an astronaut operating at 4000 ppm over months at a time would result in a loss of almost 100% in cognitive function?

As I noted earlier in this blog, human cognition and performance are affected by rapid changes in carbon dioxide concentrations over short periods of time. Not the elevated concentrations themselves. This paper completely misses this fact. It looks at acute changes in carbon dioxide concentrations and infers that chronic changes in carbon dioxide concentrations would have the same effect on cognition. In doing so this article completely ignores that humans adapt to chronic conditions such as the long term changes associated with climate change. So, no rising carbon dioxide concentrations will not make us stupider.