Direct impact of short-term exposure to pure carbon dioxide levels on cognitive performance

Keywords: cognitive performance, carbon dioxide, students, environmental chamber

Abstract

To generate further evidence on the effects of pure CO₂ on cognitive performance, an experiment was conducted in an environmentally controlled chamber. Sixty-nine healthy university students were exposed individually for seventy minutes, in three separate sessions, to three CO₂ conditions of 600, 1500 and 2100 ppm (crossover design). A validated neurobehavioral BARS test battery was used to assess participants’ cognitive performance. Results indicated that the cognitive performance of university students, measured by the BARS test battery, was not adversely affected by pure CO₂ below 2100 ppm. The findings are consistent with some prior studies, indicating that pure CO₂ within the guideline range does not imply any harm and could be treated primarily as a proxy for ventilation rates and indoor air quality in the indoor environments.

Introduction

Human exposure to indoor carbon dioxide has increased over the years due to climate change, given the growth in atmospheric CO₂ concentration [1], whilst, on the other hand, ventilation rates have been drastically reduced for energy-saving reasons [2]. Raised concerns about the impacts of elevated CO₂ concentrations on performance in the educational environment have drawn lots of attention, as schools normally face the problems of increased occupancies and decreased ventilation rates. In the UK, the Department for Education and Skills provides a guidance standard document Building Bulletin 101 (BB101) for educational buildings [3], which recommends an average concentration of CO₂ should not exceed 1500 ppm and 2100 ppm with a minimum ventilation rate of 3L/s-p. Average levels of CO₂ concentrations normally range from 600 to 1000 ppm in the educational environment, sometimes may surpass 2000 ppm and even reach a peak level of 4000 ppm [4], [5]. Cognitive performance and learning outcomes [6]–[8] showed decrements with inadequate air quality (characterized by CO₂ levels) in schools. In these studies, poor air quality due to insufficient ventilation rates might cause impairment in performance, and CO₂ was seen as a proxy for ventilation rates but not a potential pollutant. Nonetheless, adjusting ventilation rates during the experiment would affect the concentrations of other indoor pollutants like bioeffluents and contribute to cognitive performance decrement.

However, rather than being a proxy of ventilation effectiveness and an indicator of air quality, carbon dioxide itself started to gain attention in a limited number of studies about its direct impact on humans in the indoor environment. Regarding the results of cognitive performance under elevated pure CO₂ levels in previous studies (Table 1), the findings were inconsistent. Significant adverse effects of pure CO₂ on cognitive performance have been reported by some lab studies [9]–[16], and cognitive performance such as decision-making [9], [10] have been found at levels as low as 1000 ppm in the chamber and laboratory, while other studies found no statistically significant effects on cognitive performance during exposures [17]–[21]. The conflicting outcomes between the studies may be due to the diverse cognitive tests used in the experiment, differences in study populations, and disparity in the experimental procedure. The inconsistent results from previous studies and the potential socioeconomic impact due to cognitive performance decrements in built environment stimulated this research project, conducted in an environmentally controlled chamber, to generate further evidence on the effects of pure CO₂ on cognitive performance, using a systematically structured valid test battery.

Table 1. Summary of previous studies exploring the effects of pure CO₂ on cognitive performance in built environment.

(a The number in the bracket shows the sample size of the whole experiment or specific session; b Effects of exposures to CO₂ on response speed, * means significant difference between the two exposure levels, (+) means response speed increased at higher CO₂ levels, (-) means response speed decreased at higher CO₂ levels; c Effects of exposures to CO₂ on the accuracy, * means significant difference between the two exposure levels, (+) means accuracy increased at higher CO₂ levels, (-) means accuracy decreased at higher CO₂ levels)

Methods

Facilities

The experiment was conducted in a stainless-steel environmentally controlled chamber (Figure 1), measuring 4.4m wide × 4.6m deep × 3.0m high inside, with an internal volume of approximately 60m³. Through the chamber controller (Figure 2), the researcher accomplished precise control of the chamber environment, achieving defined temperature, humidity, CO₂ concentration levels and ventilation rates. Outdoor air was drawn through a HEPA filter from the building ventilation system and expelled into the chamber through a diffuser. Chemically pure carbon dioxide (99.8%) was automatically drawn from a cylinder and well mixed with outdoor air to reach the desired test levels in the chamber.

Figure 1. Image of the chamber and experiment setup.

Figure 2. The screen of the Watlow F4T Controller outside the chamber.

Experimental procedure

Sixty-nine participants were recruited from university students, including thirty-seven females and thirty-two males. Participants visited the environmentally controlled chamber at UCL Here East three times, each time exposed alone to one CO₂ condition and lasted 70 minutes (Figure 3). The interval between two experiment sessions was at least four weeks, to minimise the learning effects of cognitive tests. Participants were advised to ensure adequate sleep, avoid intense physical activity for at least 12 hours prior to the experiment, avoid drinks and non-essential medications containing caffeine and refrain from strong-scent perfume. Under a fixed high ventilation rate of 108m³/h, CO₂ was reduced to the background levels at the baseline condition, and higher CO₂ conditions were achieved with injection of pure CO₂. The temperature (23˚C) and relative humidity (50%) were kept constant in all three conditions. The participant first gave written and informed consent to the participation, then was suggested to do some quiet non-work-related activities to adapt to the chamber environment, for about 20 minutes. When the time was up, the participant put on the noise-cancelling headphones and started the BARS test on the laptop, lasting about 40 minutes. After the cognitive test, the participant filled in a post-assessment questionnaire at the end of the experiment.

Figure 3. Diagram of the experimental protocol.

BARS test battery

As a computerised assessment system designed to evaluate the neurobehavioral function of individuals, the BARS test battery [22] has been mainly used to study the effects of chemical exposure on performance [23], [24]. In this study, ten tests (Table 2) were selected to assess participants’ cognitive performance during exposure to different CO₂ concentration levels: match to sample, continuous performance test, symbol digit test, tapping, simple reaction test, reversal learning test, selective attention test, digit span, serial digit learning and progressive ratio test.

Table 2. Cognitive domains tested in the BARS battery.

*Parameters that randomly changed within a defined range each time to generate parallel tests

Results

Measured conditions of the chamber environment

Table 3 presents the conditions measured in the environmental chamber during exposures. The average CO₂ concentration levels in the baseline, intermediate and high conditions were 633, 1520 and 2120 ppm, described as 600, 1500 and 2100 ppm in the paper. The results showed that CO₂ levels were effectively controlled within 64 ppm of the desired exposure levels in all conditions.

Table 3. Measured conditions in the environmental chamber (mean ± standard deviation)

Cognitive test results

Due to the repeated measurements, univariable linear mixed-effect models were used to assess the association between cognitive test results (response times and error rates) and CO₂ concentrations. Factors that were significantly associated with at least five cognitive performance outcomes across the three conditions were then included in multivariable mixed-effect models to correct for confounding. After adjusting for potential confounders, only a few of the cognitive performance outcomes demonstrated significant effects (Table 4). Regarding response times, only two found statistically significant decreases (Figure 4). For the Selective Attention test and the Progressive Ratio Test, the response times significantly decreased at 1500 ppm and 2100 ppm, compared to the baseline condition. Regarding the error rates of cognitive tasks, no significant effect on accuracy was found in any of the ten BARS tests.

Table 4. Univariable and multivariable associations of conditions with response times and error rates in the ten BARS tests.

*Model adjusted for gender, age, first language, weekday, test durations, time slots, perceived air quality (before BARS test), perceived noise level and perceived difficulty level

Figure 4. Response times of Progressive Ratio test and Selective Attention test at three CO₂ exposure levels of 600, 1500 and 2100 ppm. *p < 0.05, **p < 0.01.

Discussion

This study indicates that exposures to pure elevated CO₂ levels up to 2100 ppm were not associated with detrimental changes in individuals’ cognitive performance in eight out of ten tests, while Selective Attention and Progressive Ratio tests reported significantly improved performance with reduced response time at 1500 and 2100 ppm, compared to a baseline of 600 ppm. Our results deviated from some previous studies that indicated detrimental effects, and were consistent with those that found no effects.Compared with most studies which found adverse effects, our study used relatively lower CO₂ levels, aiming to examine the effects at low-to-moderate levels close to routine scenarios. Some studies [12]–[14], [16] testing higher concentrations exceeding 3000 ppm tended to be more likely to exhibit the effects of CO₂ on individuals. Disparities in the results might be partly due to the different population groups. This study was consistent with the two studies which found no effects on university students [18], [20], contrary to Allen et al. which focused on office workers [10] and pilots [11]. One possible explanation for the discrepancies among the findings could be the different cognitive assessment methods employed. Allen et al. and Satish et al. both found decrement in decision-making performance with the SMS battery, which was described as a sensitive cognitive function assessment tool, but limitation also exists, as Rodeheffer et al. [25] mentioned.

Regarding the two tests which reported improved speed at higher CO₂ levels, Zhang et al. [26] found that the beta relative power of EEG significantly increased at higher concentrations, indicating that elevated CO₂ levels could be associated with higher arousal. Zhang et al. [18], [19] postulated a weak indication of arousal increased at higher CO₂ levels, based on the slightly increased ETCO₂ and reduced performance of Tsai-partington test. Arousal could play a crucial role in performance, but it remains unclear why some tests were affected with other tests showed no effect. Therefore, the improvement of performance measured by response times in the two tests could be due to higher arousal under higher CO₂ levels. Another explanation of the significant results of response time in two tests might be increases in alpha error due to multiple comparisons [27] of ten tests. Significant results from numerous tests sometimes could be false positives due to chance.

Conclusion

This study examined whether different pure CO₂ concentrations below 2100 ppm impact cognitive performance, independent of ventilation rates. With only two out of ten tests showing significant increases in response times and no significant results in accuracy under higher CO₂ levels, the findings provide empirical evidence that the cognitive performance of university students, measured by the BARS test battery, was not adversely affected by pure CO₂ below 2100 ppm, consistent with the current guidelines and some previous studies. More importantly, not only that some of the results from prior studies which reported decrements in cognitive performance could not be confirmed in this study, the contrary was found. This adds to the uncertain nature of evidence in this field, although more research is needed to confirm this finding which could be due to arousal effects or to numerous testings across several variables. The results indicate that for regulatory purposes CO₂ levels below 2100 ppm could be treated primarily as a proxy for ventilation rates and indoor air quality, while concerns still exist when levels rise above that to the occupational exposure limit of 5000 ppm.

References

[1]     R. J. Lowe et al., “Possible future impacts of elevated levels of atmospheric CO₂ on human cognitive performance and on the design and operation of ventilation systems in buildings,” Build. Serv. Eng. Res. Technol., vol. 39, no. 6, pp. 698–711, Nov. 2018.

[2]     P. Wargocki, “Saving energy for ventilation by careful selection of building,” Symp. A Q. J. Mod. Foreign Lit., no. January 2008, pp. 489–496, 2008.

[3]     Education & Skills Funding Agency, Building Bulletin 101: Guidelines on ventilation,thermal comfort and indoor air quality in schools (Version 1). 2018.

[4]     W. J. Fisk, “The ventilation problem in schools: literature review,” Indoor Air, vol. 27, no. 6, pp. 1039–1051, 2017.

[5]     D. A. Coley and A. Beisteiner, “Carbon Dioxide Levels and Ventilation Rates in Schools,” Int. J. Vent., vol. 1, no. 1, pp. 45–52, 2002.

[6]     Z. Bakó-Biró, D. J. Clements-Croome, N. Kochhar, H. B. Awbi, and M. J. Williams, “Ventilation rates in schools and pupils’ performance,” Build. Environ., vol. 48, pp. 215–223, 2012.

[7]     P. Wargocki, J. A. Porras-Salazar, S. Contreras-Espinoza, and W. Bahnfleth, “The relationships between classroom air quality and children’s performance in school,” Build. Environ., vol. 173, no. December 2019, p. 106749, 2020.

[8]     D. A. Coley, R. Greeves, and B. K. Saxby, “The effect of low ventilation rates on the cognitive function of a primary school class,” Int. J. Vent., vol. 6, no. 2, pp. 107–112, 2007.

[9]     U. Satish et al., “Is CO₂ an Indoor Pollutant ? Direct Effects of Low-to-Moderate CO₂ Concentrations on Human Decision-Making Performance,” Environ. Health Perspect., vol. 120, no. 12, pp. 1671–1677, 2012.

[10]   J. G. Allen et al., “Associations of cognitive function scores with carbon dioxide, ventilation, and volatile organic compound exposures in office workers: A controlled exposure study of green and conventional office environments,” Environ. Health Perspect., vol. 124, no. 6, pp. 805–812, 2016.

[11]   J. G. Allen et al., “Airplane pilot flight performance on 21 maneuvers in a flight simulator under varying carbon dioxide concentrations,” J. Expo. Sci. Environ. Epidemiol., vol. 29, no. 4, pp. 457–468, 2019.

[12]   L. Kajtár and L. Herczeg, “Influence of carbon-dioxide concentration on human well-being and intensity of mental work,” Idojaras, vol. 116, no. 2, pp. 145–169, 2012.

[13]   J. Zhang et al., “The effects of elevated carbon dioxide concentration and mental workload on task performance in an enclosed environmental chamber,” Build. Environ., vol. 178, no. February, p. 106938, 2020.

[14]   L. Pang et al., “The effects of carbon dioxide exposure concentrations on human vigilance and sentiment in an enclosed workplace environment,” Indoor Air, vol. 31, no. 2, pp. 467–479, 2021.

[15]   X. Cao, P. Li, J. Zhang, and L. Pang, “Associations of Human Cognitive Abilities with Elevated Carbon Dioxide Concentrations in an Enclosed Chamber,” Atmosphere (Basel)., vol. 13, no. 891, pp. 1–17, 2022.

[16]   Z. Tu, Y. Li, S. Geng, K. Zhou, R. Wang, and X. Dong, “Human responses to high levels of carbon dioxide and air temperature,” Indoor Air, vol. 31, no. 3, pp. 872–886, 2021.

[17]   J. Maniscalco et al., “Physiological responses, self-reported health effects, and cognitive performance during exposure to carbon dioxide at 20 000 ppm,” Indoor Air, vol. 32, no. 1, pp. 1–15, 2022.

[18]   X. Zhang, P. Wargocki, Z. Lian, and C. Thyregod, “Effects of exposure to carbon dioxide and bioeffluents on perceived air quality, self-assessed acute health symptoms, and cognitive performance,” Indoor Air, vol. 27, no. 1, pp. 47–64, 2017.

[19]   X. Zhang, P. Wargocki, and Z. Lian, “Human responses to carbon dioxide, a follow-up study at recommended exposure limits in non-industrial environments,” Build. Environ., vol. 100, pp. 162–171, 2016.

[20]   W. Liu, W. Zhong, and P. Wargocki, “Performance, acute health symptoms and physiological responses during exposure to high air temperature and carbon dioxide concentration,” Build. Environ., vol. 114, pp. 96–105, 2017.

[21]   S. Snow et al., “Exploring the physiological, neurophysiological and cognitive performance effects of elevated carbon dioxide concentrations indoors,” Build. Environ., vol. 156, no. March, pp. 243–252, 2019.

[22]   D. S. Rohlman, L. S. Gimenes, D. A. Eckerman, S. K. Kang, F. M. Farahat, and W. K. Anger, “Development of the Behavioral Assessment and Research System (BARS) to detect and characterize neurotoxicity in humans,” Neurotoxicology, vol. 24, no. 4–5, pp. 523–531, 2003.

[23]   S. K. Kang, D. S. Rohlman, M. Y. Lee, H. S. Lee, S. Y. Chung, and W. K. Anger, “Neurobehavioral performance in workers exposed to toluene,” Environ. Toxicol. Pharmacol., vol. 19, no. 3, pp. 645–650, 2005.

[24]   J. Rothlein, D. Rohlman, M. Lasarev, J. Phillips, J. Muniz, and L. A. McCauley, “Organophosphate pesticide exposure and neurobehavioral performance in agricultural and nonagricultural Hispanic workers,” Environ. Health Perspect., vol. 114, no. 5, pp. 691–696, 2006.

[25]   C. D. Rodeheffer, S. Chabal, J. M. Clarke, and D. M. Fothergill, “Acute exposure to low-to-moderate carbon dioxide levels and submariner decision making,” Aerosp. Med. Hum. Perform., vol. 89, no. 6, pp. 520–525, 2018.

[26]   J. Zhang, X. Cao, X. Wang, L. Pang, J. Liang, and L. Zhang, “Physiological responses to elevated carbon dioxide concentration and mental workload during performing MATB tasks,” Build. Environ., vol. 195, no. February, p. 107752, 2021.

[27]   R. J. Feise, “Do multiple outcome measures require p-value adjustment?,” BMC Med. Res. Methodol., vol. 2, no. 8, 2002.