Effects of morning versus evening high-intensity exercise on electroencephalographic and autonomic responses in middle-aged women

Abstract

[Purpose]

The time of day at which exercise is performed influences physiological responses and performance, and thus its health benefits. This study investigated whether electroencephalographic (EEG) and autonomic nervous system responses to morning versus evening exercise differed in middle-aged women.

[Methods]

Thirty women aged 50-60 years were randomly assigned to a morning or evening exercise group. Participants were asked to sleep for seven to eight hours before testing. The morning group exercised at 6:00 am and the evening group at 6:00 pm. EEG-derived indices included the theta-to-beta ratio (TBR) and spectral edge frequency at 90% (SEF-90). Photoplethysmography-derived autonomic indices comprised the HRV index and HRV-derived sympathetic-related and parasympathetic-related indices, measured at rest before and immediately after exercise. Data were analyzed using a two-way mixed-design analysis of variance. For significant interactions, simple effects and change-score analyses were performed.

[Results]

TBR and SEF-90 showed significant group × time interactions, with greater changes from pre- to post-exercise in the morning group. TBR increased and SEF-90 decreased following exercise in both groups, with greater changes in the morning group. The HRV index showed only a significant main effect of time. However, HRV-derived sympathetic-related and parasympathetic-related indices demonstrated significant group × time interactions, with greater autonomic responses in the morning group, as measured by change scores.

[Conclusion]

EEG- and HRV-derived neurophysiological response patterns to high-intensity exercise may vary with exercise timing. Morning high-intensity exercise may elicit greater neurophysiological responses than evening exercise, highlighting the importance of exercise timing in training strategies and recovery.

INTRODUCTION

Exercise research is essential for establishing evidence-based guidelines that improve quality of life and reduce health risks. Within this field, the timing of exercise is increasingly recognized as an important factor influencing physiological responses and performance. The human body is governed by circadian rhythms that follow an approximately 24-hour cycle and modulate body temperature, hormonal secretion, autonomic nervous system activity, and metabolic function. Variations in circadian rhythms can directly influence exercise performance and its resulting benefits [1-3]. Various components of physical performance, including muscular strength, anaerobic power, and endurance capacity, are reportedly greater in the afternoon than in the morning [4-7]. These diurnal variations are frequently attributable to physiological factors such as a higher core body temperature, enhanced neuromuscular activation, and improved metabolic efficiency later in the day.

Studies have also begun to interpret exercise performance not only as a function of peripheral physiological responses but also in terms of central nervous system (CNS) regulation [8,9]. Specifically, during high-intensity exercise, the CNS plays a critical role not only in regulating muscular activity but also in sustaining performance, perceiving fatigue, and modulating exercise intensity [8-11]. CNS responses can be assessed non-invasively using electroencephalography (EEG), and frequency-domain analysis provides insight into cortical arousal levels and central fatigue [12]. Among EEG-derived indices, theta (4-8 Hz) and beta (13-30 Hz) frequency bands are closely associated with fatigue and arousal, respectively, and the theta-to-beta ratio (TBR) has been widely used as an indicator of neural arousal and fatigue-related EEG states [12,13]. The spectral edge frequency at 90% (SEF-90) also reflects EEG spectral power distribution, with lower frequencies generally interpreted as reduced cortical activation and increased central fatigue [13,14]. Given their sensitivity to changes in cortical arousal and fatigue, EEGbased indices can be used to investigate neural responses to factors that may influence the baseline neurophysiological state, such as time of day.

Previous EEG studies of high-intensity exercise have reported increased theta activity and altered spectral distribution, particularly in frontal regions, following exercise. These neurophysiological changes are commonly interpreted as indicators of greater fatigue-related EEG responses and lower cortical arousal [15-17]. These findings suggest that neural responses to identical exercise vary according to the baseline neurophysiological state, highlighting the importance of these conditions when interpreting EEG responses. Consistent with this, recent studies have shown that EEG responses following high-intensity exercise differ according to an individual’s pre-exercise baseline state. For example, Jung et al. [18] reported that TBR significantly increased and SEF-90 decreased following high-intensity exercise under sleep restriction, but not under normal sleep conditions. Such findings imply that neurophysiological responses to exercise are modulated by baseline physiology, which can produce a shift toward lower EEG frequencies. This shift is often referred to as “spectral slowing” under conditions of reduced physiological readiness.

Together, these observations indicate that factors influencing the baseline state of the CNS, such as sleep status, affect post-exercise EEG responses. By extension, time of day, as a circadian determinant, may modulate neurophysiological responses to exercise.

EEG responses exhibit circadian variation [19,20]. Cortical arousal and brain activity fluctuate throughout the day, with overall EEG activity and power gradually increasing over time. Low-frequency activity is relatively low in the morning, generally increasing with prolonged wakefulness, whereas beta activity, associated with cortical activation, is typically higher in the afternoon and evening [21,22]. This suggests that even under identical external stimuli, the baseline neurophysiological state varies according to time of day, potentially leading to distinct EEG responses to high-intensity exercise in the morning versus the evening.

To date, most studies have examined either diurnal variations in exercise performance or general EEG responses to high-intensity exercise. Although time of day is known to affect physiological and performance outcomes, few studies have directly examined how EEG responses to high-intensity exercise vary across the day. Additionally, limited research has integrated EEG and autonomic nervous system indices to jointly evaluate neurophysiological responses. To address this gap, we hypothesized that time of day may influence not only exercise performance but also neurophysiological response patterns derived from EEG and HRV. As baseline neurophysiological states and post-exercise EEG responses are influenced by factors such as the stage of sleep, time of day may likewise contribute to this variability through circadian regulation. As direct empirical evidence for this possibility remains scarce, our study explored differences in EEG- and HRV-derived neurophysiological response patterns following high-intensity exercise performed in the morning versus the evening under controlled sleep conditions. Therefore, we aimed to provide preliminary insight into time-of-day-dependent EEG- and HRV-derived neurophysiological response patterns, extending the current peripheral-focused perspective in exercise physiology toward a more integrative neurophysiological approach.

METHODS

Participants

We recruited 30 middle-aged women aged 50-60 years from local community centers and fitness facilities in Ilsan, Gyeonggi-do, Republic of Korea, between January and February 2026. Participants were randomly assigned to either a morning exercise group (AM-EX; n = 15), exercising at 6:00 am, or an evening exercise group (PM-EX; n = 15), exercising at 6:00 pm. The sample size was estimated using G*Power software (version 3.1.9.7). Because few prior studies have directly examined time-of-day-dependent EEG responses following high-intensity exercise in middle-aged women, the sample size estimation was based on effect sizes reported in previous exploratory studies investigating exercise-related EEG responses under different physiological conditions [18]. Assuming a significance level of 0.05 and a statistical power of 0.80, a total sample size of 30 participants (15 per group) was considered sufficient for this exploratory study. However, given the limited prior evidence and the preliminary nature of the present study, the sample size estimation should be interpreted cautiously, and future studies with larger samples and repeated-measures designs are warranted. Our inclusion criteria required that participants had no history of neurological or cardiovascular disease, could perform high-intensity exercise, fully understood the study procedures, and voluntarily agreed to take part. The exclusion criteria included a diagnosis of neurological or cardiovascular disease, any medical condition limiting participation in high-intensity exercise, and resting blood pressure or heart rate outside the normal range. Menopausal status and the use of hormone-related medications were also assessed, and those undergoing hormone therapy were excluded. Thus, all enrolled participants were physically healthy and free from conditions that would contraindicate high-intensity exercise.

Before participation, all participants received a detailed explanation of the study’s purpose and procedures and provided written informed consent. All procedures were conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. Table 1 presents the participants’ characteristics.

Table 1.

Participant characteristics

Characteristic	Groups		p
	AM-EX (n = 15)	PM-EX (n = 15)
Age (years)	57.20 ± 2.23	55.68 ± 2.74	0.119
Height (m)	1.59 ± 0.15	1.61 ± 0.09	0.209
Weight (kg)	63.78 ± 1.33	65.14 ± 1.89	0.106
Body fat (%)	29.97 ± 1.57	28.92 ± 2.07	0.105
BMI (kg/m²)	25.72 ± 1.02	25.19 ± 0.97	0.138

Values are presented as means ± standard deviation. AM-EX: morning exercise group; PM-EX: evening exercise group. BMI: body mass index.

Experimental Design and Time-of-Day Conditions

We used a between-subject design to compare physiological responses by time of day. After being randomly assigned to a morning or evening exercise group, each participant completed a single experimental session under that time-of-day condition. All measurements were conducted under standardized laboratory conditions. To minimize potential confounding, participants were asked to abstain from alcohol, smoking, and medication during testing. They were also advised to refrain from any other structured exercise programs during the study.

Figure 1. Experimental procedure.

Sleep Control

To control for the influence of sleep, all participants were asked to obtain approximately seven to eight hours of sleep the night before the experiment. Sleep duration was verified using a sleep diary or an electronic device log. Participants were asked to maintain their usual daily routines but to abstain from caffeine for at least 24 hours before the test. We also asked them to avoid any other vigorous physical activity or daytime napping. However, chronotype, habitual wake-up time, and objective sleep quality were not formally assessed in this study.

Body Composition Assessment

Height (cm) and body weight (kg) were measured using a stadiometer and scale (BSM330, Biospace, Korea), and body fat (%) was assessed using a bioelectrical impedance analysis device (InBody 720, Biospace, Korea). To ensure accuracy, participants were asked to fast for at least 12 hours before the test, void their bladder immediately before measurement, and remove all metal objects. All measurements were taken in a standardized posture.

EEG and Autonomic Nervous System Measurements

To assess EEG- and photoplethysmography (PPG)-derived neurophysiological response patterns, we simultaneously collected EEG and PPG signals using a portable integrated device (Omnifit Mindcare, Seoul, Republic of Korea). The primary outcome variables comprised EEG-derived indices—the TBR and SEF-90—as well as PPG-derived autonomic indices, including the HRV index, HRV-derived sympathetic-related index, and parasympathetic-related index.

We took measurements after the participants had rested for at least 10 minutes in a seated position. Baseline recordings were obtained while participants sat with their eyes closed, with EEG and PPG signals recorded simultaneously for one minute. Immediately after exercise, post-exercise measurements were recorded again for one minute under the same conditions. A one-minute measurement was adopted based on previous studies evaluating ultra-short-term physiological and HRV responses under resting and exercise-related conditions [18]. These studies have suggested that ultra-short HRV measurements provide acceptable estimates of short-term autonomic responses when standardized measurement conditions and repeated comparisons are maintained.

All measurements were conducted in a quiet, temperature-controlled environment (22-24°C) to minimize external stimuli. Before testing, we carefully checked the electrode contact quality and recorded only after stable signals were confirmed.

We recorded EEG signals from the prefrontal cortex (Fp1 and Fp2) following the international 10-20 system. The sampling frequency and frequency range were set at 250 Hz and 3-41 Hz, respectively, based on the technical specifications of the device. We then performed signal quality control using the device’s built-in automated artifact rejection and quality assurance algorithms, with no additional offline preprocessing. The measurement and processing procedures followed a prior study using the same device [18].

The TBR was calculated as the ratio of relative power between the theta (4-8 Hz) and beta (13-30 Hz) frequency bands. SEF-90 was identified as the frequency below which 90% of the total EEG spectral power was contained.

PPG signals reflect peripheral changes in blood flow and were used to derive the HRV index and autonomic-related indices. HRV-based parameters were used to compute HRV-derived sympathetic-related and parasympathetic-related indices, which were interpreted as indirect physiological indices rather than direct measures of autonomic neural activity. All indices were automatically calculated using the device’s embedded proprietary algorithms. Therefore, the detailed computational procedures and specific HRV parameter weighting used to derive these indices could not be independently verified. As noted, we used one-minute recordings to assess short-term physiological responses under resting conditions, repeating the measurements under identical conditions to compare pre- and post-exercise responses and time-of-day differences.

High-Intensity Exercise Protocol

Participants performed high-intensity exercise using the treadmill-based Bruce protocol. This protocol begins at 1.7 mph with a 10% incline, with speed and grade increasing incrementally every three minutes until the participant reaches volitional exhaustion. Exercise intensity was prescribed according to each participant’s maximal exercise capacity.

Heart rate was monitored continuously throughout exercise using a heart rate monitor. The rating of perceived exertion was assessed using the Borg 6-20 scale to confirm that the exercise reached a high intensity.

Statistical Analysis

Data are presented as means ± standard deviations. Normality was assessed using the Shapiro-Wilk test, and variance homogeneity was evaluated using Levene’s test. We conducted a two-way mixed-design analysis of variance (ANOVA) to examine the effects of group (time of day) and time (pre vs. post). When a significant interaction was found, we conducted simple-effects analyses using paired and independent t-tests. To further account for potential baseline differences, we calculated change scores (Δ = post − pre) and compared these between groups. All statistical analyses were conducted using SPSS 31.0 (IBM Corp., Armonk, NY, USA), and statistical significance was set at α = 0.05. Given the exploratory nature and relatively small sample size of the present study, no additional baseline-adjusted approaches, such as analysis of covariance or linear mixed-effects modeling, were performed. Therefore, the findings should be interpreted cautiously, particularly in terms of baseline differences between groups.

RESULTS

Changes in TBR and SEF-90

The two-way mixed-design ANOVA for the TBR revealed a significant group × time interaction effect [F (1, 28) = 7.65, p = 0.010], as well as a significant main effect of time [F (1, 28) = 26.28, p < 0.001] (Table 2). The paired t-tests revealed that the TBR significantly increased from pre- to post-exercise in the morning and evening groups (p < 0.05). The independent t-tests at each time point also showed significant differences between groups at pre- (p = 0.015) and post-exercise (p = 0.038). Given the significant baseline difference in TBR between groups, between-group comparisons should be interpreted with caution, and greater emphasis should be placed on pre-post change scores. The change in the TBR (post-pre) was also significantly greater in the morning than in the evening group (p < 0.05).

Table 2.

Changes in EEG variables (TBR and SEF-90) before and after high-intensity exercise in the AM and PM groups

Variable	Group	Pre	Post	Source	F	p
TBR (θ/β ratio)	AM­EX	3.40 ± 0.21	4.18 ± 0.31	Group	1.05	0.288
	PM­EX	3.61 ± 0.23	3.95 ± 0.26	Time	26.28	<0.001***
				Group × time	7.65	0.010*
SEF-90 (Hz)	AM­EX	15.33 ± 0.65	14.52 ± 0.57	Group	4.04	0.054
	PM­EX	15.71 ± 0.56	15.21 ± 0.45	Time	17.78	<0.001***
				Group × time	9.99	0.004**

Values are presented as means ± standard deviations. TBR: theta-to-beta ratio; SEF-90: spectral edge frequency at 90%. AM-EX: morning exercise group; PM-EX: evening exercise group. Group indicates the main effect of the group. Time indicates the main effect of time. Group × time indicates the interaction effect. The p-values correspond to the main effects of group and time and the group × time interaction obtained from the two-way mixed-design analysis of variance.

^*p < 0.05,

^**p < 0.01,

^***p < 0.001.

For SEF-90, the two-way mixed-design ANOVA also revealed a significant group × time interaction effect [F (1, 28) = 9.99, p = 0.004], and a significant main effect of time [F (1, 28) = 17.78, p < 0.001]. The paired t-tests showed that SEF-90 significantly decreased from pre- to post-exercise in both groups (p < 0.001), and the independent t-tests indicated a significant between-group difference post-exercise (p = 0.002). The change in SEF-90 (post-pre) differed significantly between groups (p = 0.004).

Changes in the HRV Index and HRV-Derived Sympathetic-Related and Parasympathetic-Related Indices

The two-way mixed-design ANOVA for the HRV index revealed no significant group × time interaction effect [F (1, 28) = 2.44, p = 0.130]. In contrast, a significant main effect of time [F (1, 28) = 32.28, p < 0.001] was observed (Table 3). The paired t-tests revealed that the HRV index significantly decreased from pre- to post-exercise in the morning and evening groups (p < 0.001). However, no significant between-group differences were found at either time point or in the change scores (post-pre).

Table 3.

Changes in HRV index and HRV-derived sympathetic-related and parasympathetic-related indices before and after high-intensity exercise in the AM and PM groups

Variable	Group	Pre	Post	Source	F	p
HRV index	AM­EX	11.37 ± 1.09	7.61 ± 1.15	Group	3.05	0.088
	PM­EX	11.58 ± 1.02	8.53 ± 1.02	Time	32.28	< 0.001***
				Group × time	2.44	0.130
HRV-derived sympathetic-related index	AM­EX	4.87 ± 0.69	6.35 ± 0.72	Group	1.61	0.215
	PM­EX	4.61 ± 0.93	5.76 ± 0.93	Time	37.70	< 0.001***
				Group × time	10.11	0.004**
HRV-derived parasympathetic-related index	AM­EX	5.71 ± 0.74	4.29 ± 0.68	Group	1.02	0.321
	PM­EX	5.87 ± 0.78	4.93 ± 0.78	Time	26.78	< 0.001***
				Group × time	9.27	0.005**

Values are presented as means ± standard deviations. AM-EX: morning exercise group; PM-EX: evening exercise group. HRV: heart rate variability. Group indicates the main effect of the group. Time indicates the main effect of time. Group × time indicates the interaction effect. The p-values correspond to the main effects of group and time and the group × time interaction obtained from the two-way mixed-design analysis of variance.

^*p < 0.05,

^**p < 0.01,

^***p < 0.001.

For the HRV-derived sympathetic-related index, the twoway mixed-design ANOVA revealed a significant group × time interaction effect [F (1, 28) = 10.11, p = 0.004], along with a significant main effect of time [F (1, 28) = 37.70, p < 0.001]. The paired t-tests demonstrated that the HRV-derived sympathetic-related index significantly increased from pre- to post-exercise in both groups (p < 0.001). However, the independent t-tests showed no significant between-group differences at either time point, although the increase in the HRV-derived sympathetic-related index (post-pre) was significantly greater in the morning group than in the evening group (p = 0.004).

For the HRV-derived parasympathetic-related index, the two-way mixed-design ANOVA also revealed a significant group × time interaction effect [F (1, 28) = 9.27, p = 0.005] and a significant main effect of time [F (1, 28) = 26.78, p < 0.001]. The paired t-tests revealed that the HRV-derived parasympathetic-related index significantly decreased from pre- to post-exercise in both groups (p < 0.01). No significant between-group differences were observed at either time point. However, the decrease in the HRV-derived parasympathetic-related index (post-pre) was significantly greater in the morning group than in the evening group (p < 0.01).

DISCUSSION

Time of day is a key factor influencing physiological responses and exercise performance. However, previous research has primarily focused on performance outcomes or peripheral physiological responses, with few studies examining how EEG- and HRV-derived neurophysiological response patterns to identical exercise stimuli may vary by time of day. Moreover, few have integrated EEG and autonomic indices to evaluate neurophysiological responses. To address this gap, we investigated differences in EEG- and HRV-derived neurophysiological response patterns following high-intensity exercise performed in the morning and evening under controlled sleep conditions.

Our main findings indicate significant group × time interactions in the EEG indices, including TBR and SEF-90, with greater pre-post changes in the morning group. In contrast, the HRV index showed only a main effect of time, whereas HRV-derived sympathetic-related and parasympathetic-related indices demonstrated greater changes in the morning group, as measured by change scores. These findings suggest that our results should be interpreted on the basis of pre-post changes and interaction effects rather than on the absolute values at a single time point. Because a significant baseline difference in TBR was observed between the groups, the between-group differences in absolute TBR values should be interpreted with caution. Therefore, greater emphasis should be placed on the pre-post changes and the significant group × time interaction when interpreting the present findings.

Specifically, TBR increased, and SEF-90 decreased following exercise, with more pronounced changes observed in the morning group. The significant main effects of time indicate consistent EEG changes in response to acute high-intensity exercise irrespective of timing, whereas the significant group × time interactions suggest that the magnitude of these changes differed between the morning and evening exercise conditions. The increase in TBR reflects a relative shift toward greater theta and reduced beta activity, associated with reduced cortical arousal and fatigue-related EEG responses. Despite few direct investigations of TBR during exercise, existing EEG studies consistently report changes in individual frequency bands (e.g., theta, alpha, and beta) and electrocortical responses in frontal and motor cortical regions following an increase in exercise intensity or exhaustive exercise [23-25]. These changes have been interpreted as reflecting arousal, fatigue accumulation, and neurophysiological regulatory demands during and after exercise.

The observed decrease in SEF-90 reflects a shift in EEG spectral distribution to lower-frequency components, possibly associated with “spectral slowing.” Such low-frequency activity is generally interpreted as reflecting reduced arousal and an altered neurophysiological state. Fatigue has been associated with increased theta and low-frequency activity, reflecting reduced neural efficiency and diminished cognitive processing capacity [26,27]. However, given that SEF-90 represents a summary index of spectral distribution rather than a direct measure of specific neural mechanisms, we interpret it as reflecting a pattern of neural response rather than a definitive mechanistic change.

Notably, the EEG responses were not fixed but state-dependent, being shaped by the pre-exercise neurophysiological state. Our finding that the magnitude of the EEG changes differed according to time of day, despite an identical exercise stimulus, suggests that post-exercise neural responses were influenced not only by exercise intensity but also by the participants’ baseline physiological states. This interpretation is consistent with previous studies highlighting the role of the CNS in regulating exercise performance and fatigue, as well as evidence indicating that cortical activation and EEG responses vary with arousal level and circadian phase [8,28,29].

EEG activity is also known to vary throughout the day as a function of circadian rhythms and time awake. EEG power tends to increase over the daytime relative to the early morning, and higher-frequency components peak later in the day, whereas subjective sleepiness is often higher in the morning [22,30]. These signals also change dynamically over time, reflecting fluctuations in cortical activation and arousal states [31]. Thus, our findings provide a physiological basis for understanding how identical external stimuli can elicit different neural responses depending on time of day.

From this perspective, the greater increase in TBR and decrease in SEF-90 in the morning group may indicate that high-intensity exercise performed under conditions of relatively low baseline arousal induces more pronounced EEG spectral changes. Although these findings may reflect timeof-day-dependent neural response patterns associated with altered cortical arousal states, the nature of the EEG-derived indices limits direct mechanistic interpretation. Therefore, our findings are more appropriately understood as time-ofday-dependent differences in neural response patterns rather than direct evidence of specific physiological mechanisms.

Regarding autonomic responses, the HRV index decreased significantly after exercise, with no significant interaction effect, indicating a consistent response to high-intensity exercise in both groups. This finding is consistent with previous studies that reported reduced HRV and decreased parasympathetic activity following high-intensity or exhaustive exercise32,33. In contrast, HRV-derived sympathetic-related and parasympathetic-related indices showed greater changes in the morning group based on change scores, suggesting that the magnitude of the autonomic response varies with time of day. However, these indices were derived from HRV-based algorithms and therefore represent indirect rather than direct measures of autonomic neural activity.

During high-intensity exercise, parasympathetic withdrawal and relative sympathetic dominance typically occur, followed by gradual reactivation of parasympathetic activity and stabilization of sympathetic activity during recovery. These processes are closely linked to cardiovascular recovery, including heart-rate and HRV dynamics [34-36]. Their magnitude and rate may be influenced not only by individual physiological status but also by time-of-day baseline conditions [37,38]. Reductions in HRV following exercise have been recognized as key physiological markers of autonomic regulation and may reflect increased recovery demands and a physiological stress response under conditions of relative sympathetic activity [32,33,39].

The EEG- and HRV-derived neurophysiological response patterns we observed are unlikely to operate independently; thus, they may be integrated and regulated through shared physiological baseline states. Cortical arousal is also closely linked to autonomic regulation, and their interaction plays a critical role in exercise performance and recovery. Thus, differences in baseline neurophysiological states at different times of day may influence autonomic responses, contributing to an integrated physiological response pattern.

In summary, our results indicate that time of day relates to exercise performance and is an important physiological factor influencing EEG- and HRV-derived neurophysiological response patterns. The finding that greater neurophysiological responses followed morning exercise suggests that time of day should be considered when developing exercise strategies and recovery management. As such, our results provide preliminary insight into a more integrative neurophysiological perspective of exercise responses. Despite these findings, several limitations of the present study should be considered. First, we derived the EEG and autonomic indices from device-embedded algorithms; therefore, these represent indirect measures, limiting mechanistic interpretations. Additionally, ultra-short-term recordings may limit the physiological precision and stability of HRV-related measurements compared with conventional longer-duration recordings. Second, the single-session design did not account for potential adaptations to repeated exercise. Third, because a significant baseline difference was observed for TBR, this variable should be interpreted primarily based on pre-post response patterns and interaction effects rather than absolute values. Fourth, chronotype, habitual wake-up time, sleep quality, and potential sleep inertia effects were not formally assessed and may have influenced the observed morning responses. Fifth, the study population was limited to middle-aged women, thereby restricting the generalizability of the findings. Finally, as the study employed a between-subject rather than a crossover design, individual differences in circadian responses could not be directly assessed.

This study aimed to compare time-of-day differences in EEG- and HRV-derived neurophysiological response patterns following high-intensity exercise under controlled sleep conditions. The results demonstrated significant group × time interaction effects for EEG-derived indices (TBR and SEF-90), with greater changes observed in the morning exercise group. Among the autonomic variables, the HRV index showed a significant main effect of time, whereas HRV-derived sympathetic-related and parasympathetic-related indices exhibited greater changes in the morning group when assessed using change scores. These findings suggest that neurophysiological responses to high-intensity exercise should be interpreted primarily in terms of pre-post changes and interaction effects rather than absolute values at a single time point. The enhanced EEG- and HRV-derived neurophysiological response patterns observed in the morning condition suggest that baseline physiological states before exercise influence post-exercise neurophysiological responses. Time of day appears to be an important physiological determinant of exercise performance and EEG- and HRV-derived neurophysiological response patterns. These findings highlight the importance of considering exercise timing when developing training strategies and recovery management, providing preliminary insight into a more integrative neurophysiological perspective of exercise responses. However, this study used a single-session, between-subject design with a sample limited to a specific population, and some physiological indices were derived from indirect measurements. Therefore, caution is warranted in interpreting the findings. Future studies should incorporate repeated-measures designs and more diverse populations to better clarify the generalizability and applicability of the present results.