Cognitive and athletic performances in tropical climate: The effect of time of day

ABSTRACT

We determined whether the time of day when students exercise in a tropical climate and the environmental conditions in which they attend theory classes (air conditioning vs. tropical climate) have an impact on athletic performance and physiological, psychological and perceptual parameters. Twenty-nine students took part in four experimental sessions consisting of outdoor exercises and theory classes from 7am-1pm, according to a randomized and counterbalanced order. Cognitive and exercise performances, perceptual responses, core temperature and heart rate were assessed in the air-conditioned and tropical climate conditions. Cognitive performance was lower in the tropical vs air-conditioned environment, including lower concentration, higher inattention and lower processing speed during the theory classes. During theory classes in a tropical climate, heart rate and core temperature were higher. Average concentration scores (i.e. mean: 107.0 vs. 234.0), and inattention (i.e. mean: 96.0 vs. 32.5) at end-morning (i.e. 11am-1pm) were negatively affected by tropical climate compared to air conditioning when it was preceded by physical exercise at 9am-11am, but not at 7am-9am. There was a momentary sensation of fatigue, with higher scores at 11am-1 pm than at 7am-9am, in both conditions. Core temperatures were higher during exercise performance at 9am-11 am (38.6°C ±4.9°C) than at 7am-9 am (38.3°C ±5.2°C), but there was no difference for exercise performance, heart rate. Climate conditions must be taken into account for (i) cognitive performance and physiological parameters in tropical climate, and (ii) the time of day for exercise when theory classes take place in late morning and young student-athletes must perform cognitively well.

Introduction

Climate change is imposing ever-increasing temperatures and incidences of extreme weather events, such as heat waves. The heat stress caused by rising temperatures can have negative consequences for workers, and athletes, sometimes causing heat-related psychological and/or physiological disorders [1–4]. Indeed, it is now well established that exercise performance and health can be negatively impacted by heat [5]. Some studies have reported that environmental stress can alter cognitive functioning (complex tasks such as vigilance, reaction time, sustained attention) [6,7] and also exacerbate mental fatigue [8]. In a study of simple and complex cognitive tasks (One Touch Stockings of Cambridge: OTS-4 and OTS-6, respectively) in hot (50°C and 50% relative humidity: rH) and temperature (25°C and 50% rH) conditions, Gaoua et al. found that the performance of complex cognitive tasks decreased in the heat due to hyperthermia, while the completion of simpler tasks was not affected [9]. Schlader et al. tested the hypothesis that motor task performance in a hot environment would be impaired when performed simultaneously with a cognitive task (a test of working memory). The authors observed that motor performance was reduced following exercise in the heat compared to in a temperate condition. In addition, cognitive task performance after exercise was also reduced in the hot condition [10,11], as is particularly the case in tropical climate [8]. Nevertheless, other studies have reported no deterioration in cognitive performance [12], although these differences may be due to methodological discrepancies between the studies [6,13,14].

In tropical climates (TC, hot and humid), the rise in the average environmental temperature might increase the physiological limits to thermoregulate [15]. Several studies have shown that cognitive performance is affected in people who work and live year-round in tropical climate [16–18]. For example, in a recent study carried out with acclimated subjects, cognitive performance (attention task) was examined under three conditions (i.e. 26°C/70% rH; 39°C/50% rH, and 39°C/70% rH) in a climate chamber [19]. The results showed that at a relative humidity of 70%, increasing the temperature from 26°C to 39°C resulted in a significant drop in cognitive performance. However, when the relative humidity was reduced from 70% to 50%, cognitive test performances at 39°C significantly increased [19]. These results were confirmed by Coudevylle et al., who compared two ambient conditions (tropical climate: using a heater, and air-conditioned: temperate condition) artificially created in a closed room. The authors reported that a hot and humid environmental condition (i.e. 30°C and 70% rH) had a negative influence on the cognitive performances (processing speed, inattention, and concentration) of students living in the West Indies compared to temperate condition (22.5°C and 47% rH) [18].

The deleterious effect of tropical climate has also been observed on aerobic sports performance, particularly in young athletes living and training in Guadeloupe [5,20–22]. Recently, Jenkins et al. [23] investigated the independent effects of ambient temperature and humidity on performance and thermal, cardiovascular and perceptual responses during endurance exercise. The authors reported that, when these factors were examined independently, high air temperature increased heat stress and impaired aerobic performances, and this to a greater extent when combined with high relative humidity.

Several observational studies have also shown that heat stress can negatively affect productivity at work [24,25] and school results (learning and performance on cognitive tasks) [26,27]. For example, in assessing the impact of air conditioning on the cognitive functioning of students living in air-conditioned versus non-air-conditioned buildings during heat waves, the authors reported reduced cognitive performance in participants housed in the non-air-conditioned buildings [28]. These results concur with those of a meta-analysis of 18 studies, which showed that learners’ performances on psychological tests increased by an average of 20% when classroom temperatures were lowered from 30°C to 20°C [29], highlighting the negative effects of high ambient temperatures.

When thermal stress is high, human activity is negatively affected, particularly when complex cognitive tasks or aerobic exercise are performed [7,20,30]. These authors suggested that for optimal cognitive and athletic functioning, it would be preferable to ensure thermally comfortable environmental conditions, particularly for students living in tropical climate. For example, in the French West Indies, students in a sports sciences faculty often participate in physical activities early (i.e. 7am-9am) or mid- (i.e. 9am-11am) morning to avoid excessive heat stress in the late morning (i.e. 11am-1pm) when they attend theory classes in rooms that are most often air-conditioned. The aim of this study was to determine whether the time of day when students engage in exercise in ecological tropical climate conditions and the environmental conditions in which they attend theory classes (air conditioning vs. tropical climate) have an impact on athletic performance and physiological, psychological and perceptual parameters. We hypothesized that athletic and cognitive performances are negatively affected in tropical climate, particularly in the late morning.

Materials and methods

Participants

The data and standard deviations (±SD) from a previous study using a similar protocol enabled us to perform an a priori power analysis using G*Power 3.1 software. Sample size calculations were based on the efficiency of detecting a difference in d2T scores between two conditions (tropical and neutral climate). This analysis (with effect size = 0.25, α = 0.05, and 1-β power = 80%) repeated-measures ANOVAs indicated a minimum sample size of 24 participants. As a precaution, we planned to include at least 30 students to compensate for possible problems such as dropouts, participant absences from classes, and loss of session data in order to have a sufficient probability of analyzing complete data for 24 students.

Twenty-nine students (mean age: 21.8 ± 1.0 years old; body height: 1.7 ± 0.0 m; and body mass: 76.8 ± 16.7 kg), including six females, volunteered to participate in the study. The inclusion criteria were to be enrolled at the Sports Sciences Faculty, to have lived in Guadeloupe for more than 6 months, and to be able and willing to take part in an intervention including physical exercise and cognitive tests. All participants were acclimated to tropical climate, lived in the West Indies for at least 6 months and regularly took part in athletic activities (12.0 ± 2.1 h/week). All had given their informed written consent, were considered to be in good health, and presented a medical certificate stating that they had no contraindications for athletic activities (compulsory for enrollment each university year). This study was granted approval by the local ethics committee of the University (ACTES URp5–4–2024–07) and was conducted in accordance with the Declaration of Helsinki.

Procedure

The research protocol, which was implemented during regular lessons, was designed to avoid imposing any constraints on the timetable and to fit in with the various sports sciences classes (alternating theory and exercise classes). Two of the physical exercise classes were held at the beginning of the morning and the other two at mid-morning (Figure 1).

Figure 1.

Time course of the four experimental sessions during class days (exercise and theory) in a tropical climate: TC (in red) and in air conditioning: AC (in blue). Continuous measurements are as follows: WBGT: wet bulb globe temperature (means and standard deviations), Polar M400 heart rate monitor, CORE sensor (continuous measures in each session). The measurements taken at the beginning and end of the session are as follows: fluid ingestion, weight loss. The measurements carried out at the end of the theory courses are as follows: cognitive and perceptual responses.

Participants completed four experimental sessions composed of outdoor exercise and theory classes on separate days (7 days apart) in a randomized and counterbalanced order. While aerobic exercise was only performed in tropical climate, the theory classes took place twice in tropical climate (without air conditioning and directly exposed to the tropical heat and humidity of the moment) and twice in air conditioning. The four experimental sessions were conducted at the same time of day between 7 am and 1pm. Before the start of the study, the resting heart rate (HR) was measured for each student, as was body composition using a multifrequency body composition analyzer (InBody 720). The data on muscle mass, percentage of fat mass, lean mass per body segment and water status are reported. During the four sessions, participants had to wear an HR monitor (Polar M400) and a CORE Sensor. They were asked to arrive 30 minutes before the start of each session and to have refrained from consuming alcohol, drugs or caffeine for 24 hours before the start of each experimental session in order to avoid adverse effects on measured parameters.

Measurements

Exercise: Exercise performances were assessed during each physical exercise class. This involved running a 4-km foot race in tropical climate on a 200-m track marked out on a football pitch. Participants had to perform the 4-km run as quickly as possible in each session. The running time of each participant was measured using the Polar M400 HR monitor (confirmed by a stopwatch by two nonparticipating examiners) and exported.

Cognitive performance: Simple and complex cognitive task performances were assessed by means of the Bells test and the d2T, respectively, at the end of each theory class in the two environmental conditions. The test duration was 5 minutes. The cognitive tests were carried out by hand (handwritten) by the students at each session.

The Bells test [31] is a simple cancelation task requiring selective attention. It consists of 315 stimuli (i.e. 35 bells and 280 distractions) randomly distributed on a sheet of A4 paper. Participants had to find the maximum number of bells in 30 seconds.

The d2T [32] is a widely used test that assesses sustained and selective attention, processing speed and concentration. It consists of a single sheet of paper with 14 rows of letters (d’s and p’s). The instructions are to search each row of letters, consisting of 47 items, and to mark each letter d that has two marks above or below the letter, while also refraining from responding to seductively similar stimuli (e.g. a p with two marks). Participants were instructed to indicate the letters as quickly and accurately as possible, with a time limit of 20 seconds per row. Three standard scores – the concentration performance score, the error score, and the speed score – were derived from the test. The concentration performance score is the absolute number of detected targets minus the number of errors, thus reflecting both speed and accuracy. The error score reflects the percentage of incorrectly processed items, due to either omission or commission errors, while the speed score represents the absolute number of detected targets.

Perceptual responses: Thermal sensation (1: very cold to 7: very hot) [33], thermal comfort (1: very comfortable to 7: very uncomfortable) [34] and sensation of fatigue (1: not at all tired to 7: totally tired) were adapted from Hooper et al. [35] and recorded using visual analogue scales at the end the theory classes (in both environmental conditions). Rate of perceived exertion (RPE; 6: no exertion to 20: maximum exertion) was assessed using the Borg scale [36] immediately after the exercise performance.

Physiological measures: HR was measured using Polar H7 [37,38] belts coupled to Polar M400 monitors. HR was recorded automatically as soon as the device was triggered. HR was recorded continuously during each exercise class and each theory class (in all environmental conditions). Rest and recovery HR were also recorded. The stopwatch, integrated into the HR monitor of the Polar M400 watch, was triggered from the start of the warm-up and stopped at the end of the cool-down.

Core temperature (Tco): The CORE Sensor was used to assess students’ core temperature throughout the theory classes and during the physical exercise classes before/during/after exercise, and to avoid any risk of malignant hyperthermia (hot, humid environment). In 2020, the Swiss company GreenTeg launched the CORE Sensor, a compact, lightweight, rechargeable and relatively inexpensive portable sensor that estimates core temperature based on data related to skin temperature, heat flow and heart rate. The aim of this noninvasive sensor is to assess core body temperature, which is transmitted in real time directly to a smartwatch, smartphone or cycling computer. Several publications have reported on its advantages and disadvantages [39–43]. For example, Vedel et al. reported on its reliability compared to the rectal sensor (mean bias between the two tests was 0.02°C) in men performing two 60-minute tests in the laboratory [43]. The limits of agreement for the whole workout are − 0.3°C and 0.4°C. In our study, the CORE Sensor was used to assess core temperature during exercise and theory classes. It was placed 30 minutes before the start of the protocol on the torso/chest approximately 20 cm below the armpit, as recommended by the manufacturer, using a HR monitor strap. The data stored on the device was downloaded to the CORE application (laptop, computer) for statistical analysis.

Fluid ingestion (water) and loss were monitored. Bottles of water at room temperature (1.5 L of drinking water) were distributed to the students at each experimental session (in the morning before the start of classes). The students’ water loss over the course of the day was assessed by changes in body mass (weighed at the start and end of the protocol), as well as the quantity of water consumed.

Environmental conditions: The experimental sessions were carried out under two environmental conditions during the regular theory lessons (in the classroom):

1. Air conditioning (i.e. room with air conditioning at 22°C and 50% rH)

2. Tropical climate (i.e. room without air conditioning and directly exposed to tropical heat and humidity during the day). All measurements were made continuously. The environmental conditions were assessed by the Delta OHM HD32.2 kit, which measured the WBGT. An additional thermohygrometer (Fisherbrand; hygrometry precision ±5% rH; temperature precision ±0.1°C or 3.4°F) measured ambient temperature and rH.

Statistical analysis

Due to signal losses during the experimental sessions, Tco was only collected in 14 participants. Thus, the Tco data were analyzed and are presented as n = 14. In addition, due to student absences or injuries, certain data from exercise performances (i.e. 4-km run) were not taken into account. These data were therefore analyzed based on N = 24 participants. All other data were analyzed and are presented as n = 29.

The Shapiro-Wilk test was used to check the normality of distribution. Thus, analysis of variance (ANOVA) with repeated measures was conducted, including interactions between factors. When Mauchly’s test identified a violation of the sphericity assumption, a Greenhouse – Geisser correction was applied. In post-hoc comparisons, p values were Bonferroni-corrected for the number of comparisons. When the assumption of normality of distribution was violated, non-parametric analyses were conducted to determine any differences. Accordingly, a Friedman ANOVA and a Wilcoxon test were used. Variables and statistics are presented as means (SD), frequencies (n) or medians (quartiles) where appropriate.

Repeated-measures ANOVAs were performed on the following dependent variables: Bells test score, the four d2T-based measures of attention (i.e. processing speed, inattention, impulsivity and concentration scores), thermal sensation, thermal comfort, and sensation of fatigue, with the moment factor (time of day, i.e. early-morning vs. mid-morning) and condition (tropical climate vs. air conditioning) as the within-participant factors. Additional ANOVAs on the dependent variables from 11 am to 1 pm were performed taking into account the time of exercise performance: early- (i.e. at 7am) or mid- (i.e. at 9am) morning. The duration of the 4-km run, core temperature, HR, and RPE of each of the experimental sessions were analyzed with repeated-measures ANOVAs with the time factor (i.e. early-morning vs. mid-morning) as the between-factor. Last, repeated-measures ANOVAs were performed for the quantity of water ingested (liters) and the percentage weight loss (%) between each experimental session as an intermediate factor (Supplementary file). All analyses were performed using IBM SPSS Statistics 23. A p-value of < 0.05 was considered statistically significant.

Results

Theory classes when exercise was performed mid-morning (i.e. 9am-11am)

Physiological data

HR was affected by time [F(1,22) = 16.34, p = 0.001, ηp2 = 0.42], condition [F(1,22) = 59.93, p < 0.001, ηp2 = 0.73], and the condition×time interaction [F(1,22) = 8.55, p = 0.008, ηp2 = 0.28]. Tco was not significantly affected by time [F(1,13) = 0.10, p = 0.74, ηp2 = 0.008], but repeated measures ANOVA revealed a significant effect of condition [F(1,13) = 162.03, p < 0.001, ηp2 = 0.92] and the condition×time interaction [F(1,13) = 7.91, p = 0.015, ηp2 = 0.37]. Overall, the post-hoc analyses revealed that HR and Tco were higher in tropical climate than in air conditioning. Values (means and SD) of HR (lower panel C) and Tco (upper panel A) with significant post-hoc tests are presented in Figure 2.

Figure 2.

A and B: core temperature (Tco) (values and mean ± SD): upper panel A, when exercise was performed mid-morning (i.e. 9am-11am) and upper panel B, when exercise was performed early in the morning (i.e. 7am-9am). C and D: heart rate (HR) (values and mean ± SD): lower panel C when exercise was performed mid-morning (i.e. 9am-11am) and lower panel D, when exercise was performed early in the morning (i.e. 7am-9am).

Brackets annotated with † indicates a significant difference between the given pairs (condition effect) and * indicates a significant difference between the given pairs (time effect) (all ps < 0.05).

Perceptual data

A Friedman ANOVA indicated significant differences for thermal comfort (Friedman χ2 = 57.72; df = 3; p < 0.001), thermal sensation (Friedman χ2 = 66.24; df = 3; p < 0.001) and sensation of fatigue (Friedman χ2 = 31.23; df = 3; p < 0.001). Values (medians and quartiles) of thermal comfort, thermal sensation and sensation of fatigue with significant post-hoc tests (where appropriate) are presented in Table 1. Details of pairwise comparisons are shown in Supplementary Data.

Table 1.

Values of environmental conditions (WBGT and rH) in air-conditioned and tropical conditions, as well as sensation of fatigue, thermal comfort (from −3: very uncomfortable to + 3: very comfortable), thermal sensation (from −3: very cold to + 3: very hot) of students in theory classes when exercise was performed mid-morning (i.e. 9am-11am), or when exercise was performed early-morning (i.e. 7am-9am). Variables and statistics are presented as means (SD), or medians (quartiles) where appropriate. Brackets annotated with # indicate a significant difference between the given pairs (condition×time interaction); † indicates a significant difference between the given pairs (condition effect), and * indicates a significant difference between the given pairs (time effect) (all ps < 0.05).

	when exercise was performed mid-morning (i.e. 9am-11am)				when exercise was performed early-morning (i.e. 7am-9am)
Theory classes	7am-9am		11am-1pm		9am-11am		11am-1pm
Environmental conditions WBGT in (°C) rH (%)	Tropical climate 26.7 ± 0.5 (82.0 ± 0.0)	Air-conditioned 22.7 ± 0.4 (76.4 ± 2.8)	Tropical climate 30.0 ± 0.6 (86.3 ± 0.0)	Air-conditioned 21.8 ± 0.2 (61.3 ± 1.2)	Air-conditioned 21.4 ± 0.7 (60.8 ± 0.0)	Tropical climate 29.4 ± 0.4 (93.1 ± 0.0)	Air-conditioned 21.1 ± 0.2 (56.7 ± 0.0)	Tropical climate 30.1 ± 0.4 (95.3 ± 0.0)
Sensation of fatigue	4.0 (2.2–5.0) †	3.0 (2.0–3.0)	6.0 (5.0–6.0) #†*	5.0 (3.0–5.0) *	4.0 (3.0–5.0)	6.0 (5.0–6.0) #†	4.0 (3.0–5.0)	5.0 (5.0–5.0) #*
Thermal comfort	−1.0 (−2.0- −1.0) #†	2.0 (1.0–2.0)	−2.0 (−2.0- −1.0) #†	1.5 (1.0–2.0)	1.0 (0.0–2.0)	−2.0 (−2.0- −0.5) #†	1.0 (0.0–2.0)	−1.0 (−2.0- −1.0) #†
Thermal sensation	1.5 (1.0–2.0) #†	0.0 (−1.0–0.0)	2.0 (1.0–2.0) #†	0.0 (0.0–0.0)	0.0 (−0.5–0.0)	2.0 (1.0–2.0) #†	0.0 (−1.0–0.0)	1.0 (1.0–2.0) #†

d2 test variables

A Friedman ANOVA indicated significant differences for d2T concentration (Friedman χ2 = 42.85; df = 3; p < 0.001), d2T inattention (Friedman χ2 = 38.64; df = 3; p < 0.001), and d2T processing speed (Friedman χ2 = 38.22; df = 3; p < 0.001) during theory classes. However, no significant difference was found for d2T impulsivity (Friedman χ2 = 0.81; df = 3; p = 0.8). Values (medians and quartiles) and significant results for post-hoc tests (where appropriate) are presented in Figure 3. Details of pairwise comparisons are shown in Supplementary Data.

Figure 3.

Medians and quartiles are presented for d2T concentration (upper panel A), d2T inattention (upper panel B), d2T processing speed (lower panel C), d2T impulsivity (lower panel D) and the Bells test (lower panel E) when exercise was performed mid-morning (i.e. 9am-11am).

Brackets annotated with † indicates a significant difference between the given pairs (condition effect) and * indicates a significant difference between the given pairs (time effect) (all ps < 0.05).

Bells test

For the Bells test, an effect time-of-day was observed [F(1, 26) = 13.1, p < 0.001, ηp2 = 0.33], although the post-hoc analyses did not indicate significant differences (all ps > 0.05). There was no significant effect of condition [F(1, 26) = 0.005, p = 0.94, ηp2 = 0.005] or the condition×time interaction [F(1, 26) = 0.08, p = 0.77, ηp2 = 0.08]. Values (means and SD) of the Bells test (lower panel E) are presented in Figure 3.

Theory classes when exercise was performed early morning (i.e. 7am-9am)

Physiological data

HR was affected by time [F(1,23) = 30.72, p < 0.001, ηp2 = 0.57], condition [F(1,23) = 15.44, p < 0.001, ηp2 = 0.40], and the condition×time interaction [F(1,23) = 10.90, p = 0.003, ηp2 = 0.32]. Overall, the post-hoc analyses revealed that HR was higher in tropical climate than in air conditioning. Notably regarding the time effect, HR was higher at 9am-11 am than at 11am-1 pm in air conditioning (p < 0.001); however, there was no difference in tropical climate from 9am-11 am compared to 11am-1 pm (p = 0.9) (Figure 2, lower panel D).

For Tco, no significant effect of time [F(1,13) = 0.45, p = 0.83, ηp2 = 0.003] or the condition×time interaction [F(1,13) = 1.31, p = 0.27, ηp2 = 0.09] was revealed; however, the effect of condition was significant [F(1,13) = 122.61, p < 0.001, ηp2 = 0.90]. The post-hoc analyses showed Tco was higher at 9am-11 am and 11am-1 pm in tropical climate than in air conditioning (all ps < 0.001) (Figure 2, lower panel B). Values (means and SD) of HR (lower panel D) and Tco (upper panel B) with significant post-hoc tests (where appropriate) are presented in Figure 2.

Perceptual data

A Friedman ANOVA indicated significant differences for thermal comfort (Friedman χ2 = 50.43; df = 3; p < 0.001), thermal sensation (Friedman χ2 = 55.19; df = 3; p < 0.001) and sensation of fatigue (Friedman χ2 = 14.79; df = 3; p = 0.002) during theory classes. Values (medians and quartiles) of thermal comfort, thermal sensation and sensation of fatigue with significant post-hoc tests are presented in Table 1. Details of pairwise comparisons are shown in Supplementary Data.

d2 test variables

A Friedman ANOVA indicated no significant differences for d2T concentration (Friedman χ2 = 7.19; df = 3; p = 0.06), d2T inattention (Friedman χ2 = 7.19; df = 3; p = 0.06), d2T processing speed (Friedman χ2 = 6.72; df = 3; p = 0.08), and d2T impulsivity (Friedman χ2 = 4.39; df = 3; p = 0.2). Values (medians and quartiles) of d2T concentration (upper panel A), d2T inattention (upper panel B), d2T processing speed (lower panel C) and d2T impulsivity (lower panel D) are presented in Figure 4.

Figure 4.

Medians and quartiles are presented for d2T concentration (upper panel A), d2T inattention (upper panel B), d2T processing speed (lower panel C), d2T impulsivity (lower panel D) and the Bells test (lower panel E) when exercise was performed mid-morning (i.e. 7am-9am).

Bells test

A Friedman ANOVA indicated no significant differences for the Bells test (Friedman χ2 = 2.30; df = 3; p = 0.5). Values (medians and quartiles) of the Bells test (lower panel E) are presented in Figure 4.

Exercise

The values measured for running, HR, RPE, Tco and the significant post-hoc tests during the exercise classes (from 9am-11 am and 7am-9am) are presented in Figure 5. The post-hoc analyses revealed higher Tco in run 1 (from 9am-11am) than in run 3 (T = 5.00, Z = 2.98, p = 0.002) and run 4 (T = 5.00, Z = 2.98, p = 0.002) (all from 7am-9am). Moreover, the post-hoc analyses revealed higher Tco in run 2 (from 9am-11am) than in run 3 (T = 14.00, Z = 2.41, p = 0.015) and run 4 (T = 20.00, Z = 2.04, p = 0.041) (all from 7am-9am). There was no main effect for exercise performance, HR or RPE (p > 0.05).

Figure 5.

All data represent measurements from each session (run 1 to 4) in tropical climate conditions. A and B: run duration (upper panel: A, values and mean ± SD) and heart rate (upper panel: B, values and mean ± SD). C and D: core temperature (Tco) (lower panel: C, mean ± SD) and rate of perceived exertion (RPE) (lower panel: D, mean ± SD). Brackets annotated with # indicate a significant difference between the given pairs (time×moment of day interaction): race 1 (from 9am-11am) compared to race 3 and race 4 (all from 7am-9am) and race 2 (from 9am-11am) compared to race 3 and race 4 (all from 7am-9am) (all ps < 0.05).

Discussion

The aim of this study was to determine whether the time of day when students exercise in tropical climate (ecological) and the environmental conditions in which they attend theory classes (air conditioning vs. tropical climate) have an impact on athletic performance and physiological, psychological and perceptual parameters. Tropical climate had a negative impact on the students’ cognitive performances and the psychological and physiological factors assessed during the theory classes. Moreover, cognitive performance at end-morning (i.e. 11am-1pm) was negatively affected in tropical climate when preceded by exercise performance between 9am-11am. No differences between exercise performance, RPE or HR were found for the exercise performed in tropical climate; however, a higher Tco was measured during exercise performance from 9am-11am.

The results of the current study are in line with the literature and highlight the negative impact of tropical climate on cognitive performance, especially for complex tasks [26,27,30,44]. Indeed, the students obtained lower concentration and processing speed scores for the d2T, as well as higher inattention d2T scores, in tropical climate than in air-conditioned conditions at 7am-9 am and 11am-1pm, which are similar to the results in Coudevylle et al.’s study [18] (except for the impulsivity d2T score). Unlike the complex task, the results for the simpler task, the Bells test, showed no significant difference dependent on the environmental conditions. These results corroborate those of previous studies on this topic with, for example, the study of Gaoua et al. showing that simple cognitive tasks were not negatively affected by heat stress [6,45,46].

Interestingly, cognitive performance was negatively affected at the beginning of the day rather than at the end of the morning, whatever the climate (Table 1). These results for tropical climate are questionable, since in our study the heat stress at 11am-1 pm (30.0°C ±86.3% rH) was greater than at 7am-9 am (26.7°C ±82.0% rH). They also seem to contradict those of Tian et al., who reported a drop in cognitive performance when the temperature rises from 26°C to 39°C (with 70% rH for both cases) [19]. This might be explained by the students’ lower motivation to perform complex cognitive tasks early in the morning, in addition to the tropical climate. More specifically, the very low concentration score in tropical climate at 7am-9 am and the large difference with the air-conditioned condition (Table 1) are major discrepancies with the findings of other studies [18,47].

To date, we are not aware of any studies of cognitive performance as a function of the time of day in the tropical climate; however, the speed of response to a task and the ability to perceive stimuli seem to be faster in the afternoon than in the morning, according to Wright et al. [48], which could explain our results. On the other hand, contrary to our previous results, there was no significant difference in the d2T variables between the theory classes at 9am-11 am and 11 am-1 pm in tropical climate and air conditioning (Table 1).

In parallel, we assume that the cognitive decline we observed was linked to higher thermal sensation and a reduction in thermal comfort, as reported in several studies [8,19,30,46]. The sensation of fatigue of these students, who were passive in the classroom, was greater in tropical climate than in air conditioning at 7am-9 am and 11am-1pm, similar to the results of Malcolm et al. [49] and others [19,50]. In addition to the condition effect on fatigue, our results showed an effect time-of-day, with higher scores at 11am-1 pm than at 7am-9am, in both conditions. These results are not surprising given the cumulative cognitive (learning) and athletic (performance) demands that these students face during the day. Here, in addition to the theory class from 7am-9am, the students put in athletic performances from 9am-11 am (tropical climate in ecological condition), which could accentuate the sensation of fatigue at the end of the morning.

However, when exercise was performed early in the morning (i.e. 7am-9am), students reported a higher sensation of fatigue at 9am-11 am than at 11am-1pm, and only in tropical climate. This suggests that regardless of the time of day of exercise (i.e. early-morning vs. mid-morning) in tropical climate, the sensation of fatigue is higher among students when they then attend a theory class. As a secondary finding, the analyses showed that the cognitive attentional performances (concentration, inattention and processing speed) at end-morning (i.e. 11am-1pm) were negatively affected by tropical climate, but only when the theory classes were preceded by the exercise performed in tropical climate at 9am-11am, and not at 7am-9am. Our results are consistent with the systematic review by Donnan et al. [51], who reported that after exposure to a hot environment, cognitive performance decrements were observed, particularly when prolonged doses of vigorous exercise occurred beforehand.

Overall, our results support the following: (i) the time of day should be taken into account in the timetables when scheduling exercise in the heat for students taking theory classes requiring cognitive performance in tropical climate, and (ii) post-exercise cognitive performance in a tropical climate may depend on the experimental design used (e.g. type of environment, post-exercise delay, type of cognitive task) [52].

Except for 9am-11am, our results showed that the average student HR was overall higher in the theory classes in tropical climate, in accordance with the literature [19,49]. We assume that this physiological tension could be attributed to a higher Tco [53] and may have adversely affected cognitive performance [6,8]. Our results echo those of Gaoua et al., who exposed subjects for 4 hours to two (hot and neutral) environments and found a significantly higher body temperature in the heat at the start and end of exposure compared with in the neutral environment [54]. It seems that a slight increase in Tco can improve cognitive performance [8] but that a decrease in cognitive performance is observed when Tco exceeds ∼39.0°C, affecting more complex cognitive tasks [8,55]. Although heat-induced cognitive impairment depends on the extent of heat stress (Tco), as well as the complexity of the task to be performed, the body temperatures observed in our study were lower than those found in the literature.

We report that running performance was not affected by time of day in tropical climate, specifically between early morning (i.e. 7am-9am) and mid-morning (i.e. 9am-11am) performances. The studies that have compared performance according to time of day have used methodologies that differed from ours (i.e. time of day, types of exercise) [56,57], making comparison difficult. The acclimation of the subjects to the hot and/or humid environment could be one possible explanation may relate to, since they regularly practiced athletic activities (12.0 ± 2.1 h/week) in tropical climate. The literature reports a decrease in exercise performance in a hot environment following pre-exercise heat stress exposure and mental fatigue [58,59], which was not the case in our study. We believe that the thermal stress during the theory class (26.7°C, 82.0% rH), as well as the duration of the cognitive tasks, was not sufficient to induce mental fatigue and a deleterious effect on subsequent aerobic performance.

The higher Tco reported during the run at 9am-11 am can be explained by differences in environmental conditions, notably a higher temperature in 9am-11 am vs. 7am-9am, and by the diurnal increase in central temperature and the higher ambient temperature throughout the day [60]. Racinais et al. suggested that an increase in body temperature, prior to prolonged exercise, may alter heat storage capacity during exercise [60]. However, we assume that the increase in Tco was not sufficient to induce higher physiological stress (HR) and reduced performance. RPE was not modified by the exercise performance at different times of day, suggesting that even at slightly lower temperatures, particularly very early in the morning, the arduousness of an aerobic effort remains significant in tropical climate. Regarding drink intake and weight loss (Supplementary Figures 6A and 6B, respectively), none of the variables differed between experimental sessions. The consumption of ad-libitum water probably helped to maintain body weight at pre-session levels, and energy intake was not controlled and may therefore have influenced the weight results.

Strengths

This study has two notable strengths. First, it was conducted in a natural environment, whereas to date the effects of chronic exposure to high heat and humidity (e.g. tropical climate) on cognitive and exercise performances have been assessed using simulated environments (i.e. climate chambers to artificially recreate a tropical climate) with only a few hours of exposure to this climate [18,19]. Moreover, very few studies have reported on indoor and outdoor tropical climate exposures, in real work or school studies, because hygrometry is difficult to manipulate. Second, the experimental sessions all took place in the same time slots, in line with the students’ practical and theory classes, to avoid the effect of the normal fatigue, which occurs later in the day.

Limitations

This study also had limitations. First, the d2T cognitive task took only 5 minutes to complete, whereas the theory classes lasted for 2 hours. Also, we did not conduct baseline measurements for the students’ cognitive tests, and our results may be explained by day-to-day variability. In addition, we were confronted with a loss of signal during the protocol, and the accuracy of the of Tco assessment using the CORE Sensor may be questioned compared to a telemetric measurement via ingestible pills (BodyCap, Caen, France). Notably, the increase in Tco may be underestimated in hot and tropical climates, as reported in the literature [41,42].

Moreover, this study did not collect information on water intake at the end of each 2-hour class (theory and exercise), and therefore we were unable to examine how individual water intake is related to hydration levels and cognitive performance, although it has been reported that exercise-induced dehydration can negatively impact cognitive performance in young people [61]. The lack of a sufficient number of female subjects in this study is also a limitation, as the menstrual cycle modifies core temperature, with an increase of between 0.3 and 0.5°C during the luteal phase of exercise [62], making women more susceptible to hyperthermia [63].

Conclusion

In students acclimated to tropical climate, cognitive performance was negatively impacted in the theory classes in hot and wet as opposed to air-conditioned conditions. The same was observed for the physiological (HR, Tco) and perceptual (sensation of fatigue, thermal comfort and thermal sensation) parameters. During exercise performance, Tco was higher in mid-morning than in early-morning, suggesting that exercise performance should preferably be programmed early in the morning (i.e. 7am-9am). These results have important implications because they support the idea that the climate conditions must be taken into account for both cognitive performance and the physiological parameters of athletes living in tropical climate all year round. Moreover, particular attention must be paid to the time of day for exercise when young student-athletes have theory classes scheduled for late morning and therefore need to perform cognitively well. Given that students may have exercise classes from 11am-1pm, future studies should examine this hypothesis.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Abbreviations

HR

Heart rate

Rh

Relative humidity

RPE

Rate of perceived exertion

TC

Tropical climate

Tco

Core temperature

WBGT

Wet bulb globe temperature

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/23328940.2025.2535046