The impact of climate change on children’s cognitive performance in English school building stock

Keywords: Schools, Retrofit, Climate Change, Cognitive Performance

Research background

In the UK, there are around 9 million children in schools, spending most of their learning time in classrooms. By learning knowledge and skills across different subjects, children develop their cognitive function that fosters critical thinking, problem-solving abilities, and a broader understanding of the world schools [1]. In classrooms, this process is affected by many factors, such as curriculum design, quality of instruction, and classroom environment (including social and physical environment) [2]. Among these, the physical environment plays a crucial role in children’s cognitive performance in classrooms. For example, high indoor temperature increases the burden of human’s thermoregulatory system and suppress brain neural activity [3], often leading to headache, fatigue and distraction, and therefore, impair the cognitive performance of children. Thus, maintaining good classroom environment is essential to avoid cognitive function impairment [4].

Climate change is leading to hotter and drier summers in the UK, yet many school buildings were constructed without consideration for the potential impact [5]. In addition, schools often have high occupancy densities and intensive use of teaching devices, resulting in significant internal heat gains. On-site surveys have indicated that many classrooms lack adequate natural ventilation and are not equipped with mechanical cooling systems [6-8], making them vulnerable to overheating. However, empirical evidence remains limited regarding the extent to climate change may impair cognitive performance of school-aged children in England.

Previous experimental studies have attempted to explore the exposure - response relationship between indoor environmental parameters to cognitive performance. However, these studies derived the relationship based on a small sample of participants, limiting the generalizability of their findings to broader populations. Some studies [9, 10]have developed the exposure-response function linking cognitive performance to indoor temperature/air quality by meta-analysis. By combining data from multiple studies, these analyses increase statistical power and enable the application of the resulting functions to larger populations.

This study investigates the impact of climate change on children’s cognitive performance in English secondary schools, using representative schools from three regions: Thames Valley (Southern England), West Pennines (Central England), and the Borders (Northern England). Future indoor temperatures were predicted using EnergyPlus with Chartered Institution of Building Services Engineers (CIBSE) weather files. Cognitive performance was then estimated using established temperature-performance functional relationships. The study also evaluates two mitigation strategies: improved ventilation and air-conditioning by comparing their effectiveness in reducing cognitive performance loss caused by climate change.

Methodology design

School building models

There are around 3400 secondary schools located 13 climate regions in England [11]. Due to the high computational demand of modelling each school individually, this study adopted an archetype-based modelling approach, developing a small number of representative models to reflect the broader school building stock [12]. The archetype approach first categorised English school buildings by construction era, as each period is associated with characteristic geometry and insulation levels. ‘Seed models’ were then developed based on these typical features (Table 1). These models were further refined to account for regional variations in average building height and floor area using Property Data Survey Program (PDSP) datasets. For schools in each of the three selected regions - Thames Valley (N = 230), West Pennines (N = 75), and Borders (N = 32) - 15 archetype models were generated by scaling the ‘Seed models’ accordingly (Figure 1).

Table 1. The built form and built-up characteristics of each ‘Seed’ model (GF: Ground Floor. EW: External Wall. R: Roof).

Figure 1. The school archetype models for the three representative regions.

Each archetype includes a main building (representing the majority of school space) and an additional building (representing remaining facilities). Operational parameters were derived from National Calculation Method (NCM) and BB101 school design guidelines. Simulations were run from May to September, excluding weekends and holidays.

Cognitive performance prediction

Cognitive performance was estimated using the by the function developed by Worgocki et al., 2019 [9].

RP = 0.2269 t² − 13.441 t + 277.84                (1)

Where t represents the indoor temperature, and RP represents relative performance at a specific temperature. This function is valid for children aged 9-18 and for indoor temperatures between 20°C and 28°C. performance at and below 20°C was normalised to 100%, based on Wargocki and Wyon [13]. for temperatures above 28°C, it was capped at the level observed at 28°C (79.4%) due to limited empirical evidence.

To quantify the impact of elevated temperatures, Cognitive Performance Loss (CPL) was used as a Key Performance Indicator (KPI), defined as:

CPLt= 100% − RP                                                  (2)

CPL was calculated based on predicted indoor temperature on an hourly basis using EnergyPlus, with hourly CPL values ranging from 0% to 20.6% across all temperature levels. These hourly CPL values were then aggregated at the school or regional level to reflect the overall performance.

Weather files

The weather files chosen for this study were derived from CIBSE weather data sets, which are the standardized weather data in the building industry in the UK. Future climate scenarios were represented using data for three periods: 2020s (2010–2039), 2050s (2040–2069), and 2080s (2070–2099). Design Summer Year (DSY) files for London, Manchester, and Newcastle were selected to represent the climate conditions of the Thames Valley, West Pennines, and Borders regions, respectively, and were used as climate inputs for the school building models.

Results

The impact of climate change on cognitive performance

Figure 2 illustrates the distribution of hourly CPL for children across 15 school archetypes, with each sub-graph corresponding to one archetype. Within each sub-graph, three boxplots represent the hourly CPLs under three future climate periods for a given school type. For each school in a given climate period, hourly CPLs were area-weighted averages across all classrooms, with a total of CPL values for the simulated period.

A clear upward trend in CPL is observed across all archetypes, with boxplots becoming progressively flatter and median CPLs (orange lines) increasing from the 2020s to the 2080s, showing a consistent shift toward higher cognitive performance loss as outdoor temperatures rise. This pattern is particularly pronounced in Southern England schools, where by the 2080s, CPLs reach 20.6% in most hours during warmer months under London climates. In comparison, schools in Central and Northern England display relatively better conditions in the 2020s. For instance, in Central England schools, the median CPL in the 2020s is 16.0%, and in Northern England, it is even lower, at approximately 13.0%. However, the rising temperatures over the decades drive significant increases in CPL across these regions as well. By the 2080s, the median CPL for Central England schools reach approximately 19.0%, while Northern England schools approach a median of 18.0%. Despite these increases, Northern schools remain slightly less impacted than their Southern and Central counterparts, benefiting from cooler baseline temperatures. Nevertheless, the narrowing interquartile ranges across all regions indicate that the majority of children, regardless of location, will experience conditions leading to near-maximum CPL in the 2080s.

Figure 2. The distribution of hourly CPLs of pupils during warmer months in Southern, Central and Northern England schools under future climates.

To assess whether CPL distributions differ significantly among school archetypes within the same region and climate period, Kruskal–Wallis tests were conducted. As shown in Table 2, none of the comparisons yielded statistically significant differences (p > 0.05), indicating that within-region variations among archetypes are minimal. Therefore, future local climates are likely to exert broadly uniform effects on children’ cognitive performance.

Table 2 Results of K-W test for schools in each region in 2020s, 2050s and 2080s

The CPLs of secondary school children by climate regions were explored by aggregating the hourly CPLs of all school archetypes in each region. To be more specific, there are 5 school archetypes in each region, and the study calculated hourly CPL for a total of 2652 hours. Figure 3 shows the frequency distributions of all school hours in the three regions in future climates. A categorisation approach was proposed to better compare and visualize the overall CPLs in the following analyses: No loss (CPL = 0%), Minimal loss (0% < CPL ≤ 5%), Moderate loss (5% < CPL < 20.6%) and Severe loss (20.6%).

Figure 3. The frequency distributions of hourly CPLs within different levels in English schools per region.

The results indicate that outdoor climates have different impacts on schools at the regional level:

Southern England: Schools in this region have minimal hours categorized as 'No loss' and 'No significant loss' in future climates. There are around 59.8% of hours in which children experience severe cognitive performance loss (20.6%) under 2020s climate scenario, and the hours categorized as ‘Severe loss’ rise to 81.1% in 2080s climates, both of which are much higher than those in the other two regions.

Central England: for schools in this region, the majority of hours (78.6%) fall under the category of 'Moderate loss' in all future climate periods. However, due to an increase in hours with severe CPLs (38.1%), the frequency of hours at 'moderate loss,' 'no significant loss,' and 'no loss' will decrease under climate change.

Northern England: for schools in this region, 11.3% of hours experience 'no loss,' while only 7.5% of hours are classified as 'severe loss' in current climate. As the climate warms, the percentage of hours with 'severe loss' is projected to rise to 26.1%, while the percentages of hours categorized as 'no loss' (1.6%) and 'no significant loss' (7.7%) decrease in the 2080s.

Ventilative cooling to mitigate climate change impact

The impact of increased ventilation rates in classrooms on cognitive performance were examined. Two ventilation rates were tested: 8 ℓ/s-p and 15 ℓ/s-p, in comparison with the baseline ventilation rate (5 ℓ/s-p).

Within each climate period, increasing the ventilation rates has demonstrated positive impacts on schools (Figure 4 and Figure 5), resulting in a higher proportion of hours with CPLs categorized as 'no loss' and 'no significant loss' at 15 ℓ/s-p compared to 8 ℓ/s-p. However, average CPLs at higher ventilation rates rise, and the difference in median CPLs between the baseline and a higher ventilation rate diminishes across all regions over the future climatic periods. For instance, in Southern England schools, the median CPLs increase from 14.1% in 2020s to 18.0% in 2080s when the ventilation rate is set at 15 ℓ/s-p. In Central England schools, the median CPLs rise from 1.9% to 10.0%, while in Northern England schools, they increase from 0.0% to 7.1%. This implies that in 2020s, higher ventilation rates prove more effective in improving cognitive performance, but their impact weakens in the 2050s and 2080s due to the rising outdoor temperatures.

In summary, ventilative cooling has positive effects on the cognitive performance of children in all three regions, while its effectiveness will be limited in 2050s and 2080s because of the warmer climates.

Figure 4. The frequency distributions of hourly CPLs within different levels in English schools per region at ventilation rate of 8 ℓ/s-p.

Figure 5. The frequency distributions of hourly CPLs within different levels in English schools per region at ventilation rate of 15 ℓ/s-p.

The potential of set-point temperature control on the cognitive performance of children by introducing air conditioning were examined (Table 3). To ensure that the classrooms remain within satisfactory levels, CIBSE Guide A recommends the summer operative temperatures in teaching space should be kept from 21°C to 25°C. Therefore, 21°C and 25°C, as the border recommended values, were chosen for the analyses in this section. The air conditioning is assumed to operate when the classroom temperature is above the set-point during the occupied hours in all schools. In these two scenarios, the ventilation rates are maintained at baseline ventilation rate (5 ℓ/s – person) to meet minimum fresh air supply requirements for the children.

Table 3. The CPL of children in English schools in changing climate scenarios, per climate region, due to different ventilation rates.

It should be noted that median CPLs are different at the same set point in three regions in 2020s and 2050s, with schools in Central and Northern England having median CPLs of 1.3% and 2.9% less than schools in Southern England. This is because average outdoor temperatures projected in future DSYs in Central and Northern regions remain relatively low (less than 15°C) during May, June and September, and therefore the classroom temperature may not reach the 25°C set point temperature during these months. Since the classroom remained ventilated below 25°C, ventilation can bring cooler outdoor air into the classrooms, resulting in a relatively cool environment during the non-air conditioning period, and therefore, decrease in overall cognitive performance loss.

Figure 6 The frequency distributions of hourly CPLs at different levels in England per region in future climates at set-point temperature of 21°C.

Figure 7 The frequency distributions of hourly CPLs at different levels in England per region in future climates at set-point temperature of 25°C.

Table 4 The CPL of children and cooling load of schools in England in changing climates per climate region.

As expected, setting the temperature at 21°C demonstrates a significant improvement in cognitive performance, CPLs categorized as 'Moderate Loss' and 'severe loss' are eliminated, resulting in an average CPL decrease to 4.4% across all regions during all future climate periods. Table 4 also provides insights into the corresponding cooling loads of schools per region at the two set-point temperatures. The cooling load calculation weighted the total floor area of each school archetype within a region. Cooling loads set at 21°C will have that are 20 - 22 kWh/m² higher in Southern England schools, and 16 - 20 kWh/m² higher in Central England schools, and 12 - 16 kWh/m² in Northern England schools in the future, compared to those set at 25°C. Southern England schools exhibit greater cooling loads compared to those in Central and Northern England. By the 2080s, cooling loads are projected to reach 54 kWh/m² in Southern England schools and 25 kWh/m² in Northern England schools. By setting lower classroom temperatures, schools can significantly reduce cognitive performance loss, but at the expense of increased cooling demands, particularly in regions with warmer climates.

Discussion and conclusion

This study proposes a quantitative framework to evaluate the impact of climate change on cognitive performance in English secondary schools, using cognitive performance loss (CPL) as a KPI. Representative schools in Southern, Central, and Northern England were assessed under future climate scenarios. Results suggest that warming climates are likely to exacerbate CPL, with variations driven by local climate conditions, while differences among schools in the same region are minimal. Ventilative cooling is shown to be effective in mitigating CPL in the near future; however, its efficacy diminishes under mid- to late-century climate projections. In contrast, the adoption of air conditioning becomes increasingly advantageous in the far future, despite its higher energy demand. The proposed framework aims to provides evidence to climate risks faced by children and insights on possible optimisation pathways for English school buildings. However, as the exposure-response relationship between indoor conditions and cognition is still evolving, the findings should be interpreted as indicative projections requiring further validation. The methodology is flexible and can be refined in line with future psychological research.

References

[1]     Vilia, P.N., et al., Academic achievement in physics-chemistry: The predictive effect of attitudes and reasoning abilities. Frontiers in psychology, 2017. 8: p. 1064.

[2]     Zhang, F., R. de Dear, and P. Hancock, Effects of moderate thermal environments on cognitive performance: A multidisciplinary review. Applied Energy, 2019. 236: p. 760-777. https://doi.org/10.1016/j.apenergy.2018.12.005.

[3]     Fischer, S., et al., Emerging effects of temperature on human cognition, affect, and behaviour. Biological Psychology, 2024. 189: p. 108791. https://doi.org/10.1016/j.biopsycho.2024.108791.

[4]     Wang, D., et al., Experimental investigation of the effect of indoor air temperature on students' learning performance under the summer conditions in China. Building and Environment, 2018. 140: p. 140-152. https://doi.org/10.1016/j.buildenv.2018.05.022.

[5]     Moazami, A., et al., Impacts of future weather data typology on building energy performance – Investigating long-term patterns of climate change and extreme weather conditions. Applied Energy, 2019. 238: p. 696-720. https://doi.org/10.1016/j.apenergy.2019.01.085.

[6]     Mumovic, D., et al., Winter indoor air quality, thermal comfort and acoustic performance of newly built secondary schools in England. Building and Environment, 2009. 44(7): p. 1466-1477. https://doi.org/10.1016/j.buildenv.2008.06.014.

[7]     Zhang, Y., et al., A field survey on the indoor environmental quality of the UK primary school classroom. British Journal of School Nursing, 2016. 11(10): p. 492-496.

[8]     Teli, D., et al., Thermal Performance Evaluation of School Buildings using a Children-based Adaptive Comfort Model. Procedia Environmental Sciences, 2017. 38: p. 844-851. https://doi.org/10.1016/j.proenv.2017.03.170.

[9]     Wargocki, P., J.A. Porras-Salazar, and S. Contreras-Espinoza, The relationship between classroom temperature and children’s performance in school. Building and Environment, 2019. 157: p. 197-204. https://doi.org/10.1016/j.buildenv.2019.04.046.

[10]   Wargocki, P., et al., The relationships between classroom air quality and children’s performance in school. Building and Environment, 2020. 173: p. 106749. https://doi.org/10.1016/j.buildenv.2020.106749.

[11]   Schools, pupils and their characteristics. 2022 Nov10th 2022]

[12]   Schwartz, Y., et al., Developing a Data-driven school building stock energy and indoor environmental quality modelling method. Energy and Buildings, 2021. 249: p. 111249. https://doi.org/10.1016/j.enbuild.2021.111249.

[13]   Wargocki, P. and D.P. Wyon, Providing better thermal and air quality conditions in school classrooms would be cost-effective. Building and Environment, 2013. 59: p. 581-589. https://doi.org/10.1016/j.buildenv.2012.10.007.