Satish · Environmental health perspectives 2012 · double-blind randomized crossover chamber study · n=22

Is CO2 an indoor pollutant? Direct effects of low-to-moderate CO2 concentrations on human decision-making performance.

Cited 1021 times in the scientific literature.

Level 2 - randomized trial

Randomized, double-blind crossover exposure trial

PubMed 23008272 · doi:10.1289/ehp.1104789 · record verified 2026-08-29

What was done

Twenty-two participants in six groups were exposed to three carbon dioxide (CO2) concentrations (600, 1,000, and 2,500 ppm) in an office-like environmental chamber. Each participant underwent three 2.5-hour exposure sessions on a single day, with exposure order balanced across groups. Ventilation rate and temperature were kept constant, with higher concentrations achieved by injecting ultrapure CO2. Participants and the test administrator were blinded to CO2 concentrations. Under each condition, participants completed a computer-based decision-making performance test and subjective questionnaires regarding health symptoms and perceived air quality. Data were analyzed using analysis of variance (ANOVA) models.

What was found

Compared with the 600 ppm baseline, exposure to 1,000 ppm CO2 resulted in moderate, statistically significant performance decrements on 6 of 9 decision-making scales. At 2,500 ppm CO2, large and statistically significant reductions occurred on 7 of 9 decision-making scales (raw score ratios ranging from 0.06 to 0.56), while performance on the focused activity scale significantly increased.

Why it matters

This study suggests that indoor CO2 directly impairs human cognitive and decision-making performance at concentrations commonly found in poorly ventilated buildings, rather than merely acting as a passive surrogate marker for other indoor pollutants.

Limits

The sample size was very small (n = 22), and all three 2.5-hour exposure conditions occurred on a single day, leaving potential carryover or fatigue effects uncertain. The abstract does not provide participant demographic characteristics, exact effect sizes for the 1,000 ppm condition, or findings from the symptom and air quality questionnaires, and laboratory computer tasks may not fully capture real-world occupational performance.

Cited by