Effects of transcranial pulsed current stimulation on sleep quality in Student-Athletes with different types of sleep disorders

Abstract

Objective

To investigate the intervention effects of transcranial pulsed current stimulation (tPCS) on sleep quality in Student-Athlete with different types of sleep disorders.

Methods

Thirty Student-Athlete with morning-type (MT) sleep disorders and 30 with evening-type (ET) sleep disorders were recruited, aged 18–22 years, including 45 males and 15 females. Participants within each disorder type were equally divided into an experimental group (EXP, n = 15) and a control group (CON, n = 15) using a random number table. All groups received intervention for 28 consecutive days at a fixed time (stimulation intensity: 1.5 mA; duration: 30 min). The EXP received active stimulation, while the CON received sham stimulation. Heart Rate Variability (HRV) parameters and Pittsburgh Sleep Quality Index (PSQI) scores were collected from all participants before and after the intervention.

Results

(1) The ET group had lower baseline PSQI scores and HRV parameters than the MT group. (2) Following the tPCS intervention, the EXP exhibited significant decreases in PSQI (P = 0.007, = 0.231) scores and Low Frequency (LF) power, along with significant increases in the Root Mean Square of Successive Differences (RMSSD, P < 0.001, = 0.633), Standard Deviation of NN intervals (SDNN, P < 0.001, = 0.399), and High Frequency (HF, P < 0.001, = 0.541) power. These changes were also significantly different compared to the CON (P < 0.05). (3) Post-intervention, the MT experimental group showed greater improvement in PSQI scores, whereas the ET experimental group exhibited a more significant increase in HRV indices.

Conclusion

(1) A 4-week tPCS intervention can improve sleep quality and increase parasympathetically-mediated HRV indices in Student-Athlete with sleep disorders. (2) Differential tPCS intervention time windows exist for Student-Athlete with sleep disorders based on their chronotype. (3) tPCS intervention administered between 18:00 and 19:00 yields greater improvement in sleep quality for the MT group compared to the ET group.

Introduction

Sleep is not a passive state of rest but rather an active and crucial biological process fundamental to maintaining human physiological homeostasis and cognitive function. High-quality sleep is indispensable for eliminating fatigue, restoring energy, and ensuring the stable functioning of the nervous, cardiovascular, endocrine-metabolic, and immune systems [1]. However, the contemporary fast-paced lifestyle and high levels of stress have severely disrupted sleep. Consequently, sleep insufficiency and sleep disorders have emerged as a prominent public health problem [2]. Data from the World Health Organization (WHO) indicate that 27% of the global population is affected by sleep problems to varying degrees. Some studies indicate that the insomnia prevalence rate among Chinese university students ranges from 9.4% to 38.2%, whereas a Chinese sleep quality report suggests that the proportion of students with poor sleep quality is as high as 45% [3, 4].

For athletes, the importance of sleep is further amplified. Sleep is not only a means of recovery, but also an indispensable pillar as important as training and nutrition. It is the key link of recovery, adaptation and improvement of competitive performance [5, 6]. It is also the most direct and important physiological and psychological method for athletes to recover and adapt to training stimulation [7]. However, high intensity training, tight competition schedule, frequent travel and huge psychological pressure make athletes a high-risk group of people with insufficient sleep [5–8]. Research shows that athletes are usually difficult to get the recommended 7–9 h of sleep per night, and in order to achieve the best recovery effect, they may need up to 9–10 h of sleep [6]. For Student-Athlete, they play a dual role and must seek a balance between heavy academic pursuit and high-intensity sports training, which further magnifies the importance of sleep [9]. However, the sleep problems of this group are very serious. Research shows that Student-Athlete generally have problems of insufficient sleep and poor sleep quality, and the incidence of sleep disorders is much higher than that of the general population [8, 10]. Different research reports show that the proportion of poor sleep quality of Student-Athlete is between 42% and 65% [11, 12]. A survey of Student-Athlete during the season showed that their average sleep per night was only 6.27 h, far lower than the recommended 8 h [12]. This common sleep problem stems from multiple and unique stressors, of which academic pressure and training pressure are the main factors causing sleep disorders. The study clearly points out that there is a significant positive correlation between weekly learning time and poor sleep quality [13], while there is a strong negative correlation between high academic pressure and sleep quality [14]. At the same time, training load, competition arrangement (especially early morning or late-night competition), frequent travel and pre competition anxiety have seriously interfered with their establishment of regular sleep rhythm [15, 16]. These factors contribute to poor sleep quality of Student-Athlete, which poses a serious challenge to their physical and mental health.

“Excessive arousal hypothesis” is one of the mainstream physiological explanations for low sleep quality, which believes that the brain of people with poor sleep quality is in a state of excessive excitement for a long time [17]. This state is closely related to the functional imbalance of the autonomic nervous system (ANS), especially the overactive sympathetic nervous system responsible for the “fight or flight” response, while the parasympathetic nervous system responsible for “rest and digestion” has insufficient activity [18]. Therefore, intervention measures that can effectively regulate the balance of autonomic nervous system and reduce sympathetic hyperactivity may theoretically become effective means to improve sleep quality. In recent years, Transcranial Pulsed Current Stimulation (tPCS) has emerged as a promising non-invasive brain stimulation technique, owing to its superior efficacy in modulating the autonomic nervous system and the distinct advantages of pulsed current in minimizing charge accumulation and enhancing skin safety during long-term interventions [19, 20]. A study on athletes found that athletes’ application of tPCS after moderate intensity training can significantly promote the recovery of autonomic nervous system function and enhance the regulation ability of parasympathetic nerve [21]. Under stress conditions, tPCS intervention increased heart rate variability (HRV), which showed that tPCS could change the nervous system to a parasympathetic dominated state, and achieve better autonomic nerve control [22], which had the potential to improve sleep quality. The type of individual Chronotype can be divided into morning type, intermediate type and evening type, which has a significant impact on the quality of daily sleep [23]. However, existing research has seldom considered the relationship between chronotype, intervention modalities, and sleep quality. Therefore, this study aims to investigate the effects of tPCS on sleep quality in Student-Athlete with sleep disorders and different chronotypes. Collectively, this study proposes the following hypotheses:

1. tPCS intervention can improve the sleep quality of Student-Athlete with different types of sleep disorders.

2. tPCS intervention has different effects on different types of sleep disorders.

Participants and methods

Ethics

All procedures were performed in accordance with the ethical standards of the Declaration of Helsinki. Prior to participation, all individuals provided written informed consent. The study protocol was reviewed and approved by the Institutional Ethics Committee of Nantong University (Approval No. 1; 16 November 2020). Clinical trial number: not applicable.

Participants

This study initially recruited 304 Student-Athlete. Participants were excluded based on preliminary screening if they reported: frequent smoking, alcohol consumption, or caffeinated beverage intake within the past month; pain-related injuries; or conditions known to affect sleep quality or HRV, such as depression or insomnia.

Student-Athlete with a Pittsburgh Sleep Quality Index (PSQI) score greater than 5 were included in the study. Participants were screened using the Morningness-Eveningness Questionnaire (MEQ), and those with scores ranging from 4 to 11 and 18–25 were selected. Subsequently, morning-type (MT) and evening-type (ET) athletes were randomly assigned to groups using the random number table method (simple randomization). Following the 4-week tPCS intervention, 15 participants per group completed the entire experiment. The detailed procedure is illustrated in Fig. 1.

Fig. 1.

Flowchart of participant screening

Prior to the experiment, participants received a detailed explanation of the tPCS procedure, including its principles, operation, potential effects, and safety. The experimental protocol (pre-test preparations, procedural steps, and post-test requirements) was also described. After comprehension, all participants provided written informed consent, confirming voluntary participation and adherence to the study’s protocols.

Sample size was calculated using G*Power 3.1. The sample size was calculated based on the study’s primary outcome: the PSQI total score, referring to previous research on the efficacy of non-invasive brain stimulation for sleep quality [24, 25]. The effect size of 0.4 was set. An alpha (α) error probability of 0.05 and a statistical power (1-β error probability) of 0.95 were used. A repeated-measures analysis of variance (ANOVA), targeting the primary interaction effect, indicated a required total sample size of 44 participants. Considering a potential attrition rate of 15–20%, the planned total recruitment sample size was 56 participants (14 per group). Thus, the final sample of 15 participants per group in this study met the requirements (Table 1).

Table 1.

Baseline characteristics of participants

	Group	Sex (M/F)	Age (years)	Height (cm)	Weight (kg)	Training duration (years)
MT	Experimental	11/4	18.73 ± 0.8	171.3 ± 1.86	75.21 ± 4.51	3.53 ± 0.52
	Control	11/4	20.07 ± 0.8	174.64 ± 2.3	83.05 ± 6.08	4.47 ± 0.52
ET	Experimental	11/4	20.73 ± 0.88	174.7 ± 2.68	83.67 ± 6.54	5.07 ± 0.7
	Control	12/3	21.6 ± 0.51	175.25 ± 2.47	84.7 ± 6.42	5.53 ± 0.52

Experimental procedure

Participants in both the MT and ET groups were equally assigned to either the experimental or control group using the random number table metho. One day before the experiment (pre-test) and one day after the experiment (post-test), all participants completed HRV data collection and the PSQI questionnaire upon waking (07:00). Participants received one tPCS intervention session daily from 18:00 to 19:00 for a total of 28 days. Ultimately, 60 participants (15 per group) successfully completed the entire study protocol.

On the day before the intervention (pre-test), participants were brought to the laboratory by their team leader. They rested quietly for 5 min, after which the principal investigator explained the experimental procedure, objectives, and precautions. Data collectors then recorded participants’ basic information, followed by the administration of the HRV test and the PSQI questionnaire. The HRV test duration exceeded 5 min. Participants then proceeded with the 4-week tPCS intervention. One day after the intervention concluded, all groups repeated the HRV test and the PSQI questionnaire. The experimental flowchart is shown in Fig. 2.

Fig. 2.

Experimental testing procedure flowchart

Given the diurnal fluctuations of HRV, all tests were conducted in the morning after waking. Testing occurred in a comfortable, temperature-controlled (25 °C) room with dim lighting and no noise interference. Participants rested quietly in a supine position for 20 min before the HRV test was conducted, during which they remained supine and breathed normally.

Intervention protocol

This study was a double-blind experiment, both the participants and the data collectors were blinded to the group assignments. Utilizing a 2 (group) × 2 (intervention type) × 2 (time) three-factor mixed experimental design. Standardized tPCS procedures were implemented, and consistent experimental conditions were maintained. The tPCS device was operated by skilled research personnel. In consideration of the participants’ biological rhythms and daily routines, participants received tPCS stimulation daily between 18:00–19:00 during the four-week experimental period [26, 27]. A single tPCS electrode was placed over the prefrontal cortex area, and two electrodes were placed near the bilateral mastoids, secured with an elastic bandage (Fig. 2). The stimulation intensity was 1.5 mA, and the duration was 30 min. For the EXP, the current intensity was ramped up from 0 mA to the target intensity over 30 s at the beginning of the stimulation, and ramped down to 0 mA over 30 s at the end. All procedures were completed by the same staff member. The CON received the same stimulation duration, intensity, and ramp-up process, but the current was ramped down to 0 mA 30 s after the initial ramp-up. All participants remained quiet during stimulation, avoiding extraneous stimuli. All procedures adhered to the safety consensus reached at the Copenhagen meeting [20]. The tPCS operation for each participant was performed by their designated research staff member. If any adverse reactions (e.g., pain, dizziness) occurred during the intervention, the session was terminated immediately.

Experimental tools

Transcranial pulsed current stimulation (tPCS) device

Intervention was conducted using a tPCS device developed by our research group (China National Key R&D Program for the Winter Olympics, No: 2019YFF0301604). The basic parameters of the tPCS were as follows: pulse waveform: square wave, biphasic pulse; duty cycle: 29.7%; pulse frequency: 60–80 Hz (Stimulation-sequence-locked dynamic frequency sweep stimulation mode); current intensity adjustment range: 0–2 mA; maximum voltage: 20 V; and electrode sizes: one (5*9) cm² electrode and two (5*5) cm² electrodes. This product passed national safety certification (Report Nos: CHTSM21040049, CHTSM21040050, CHTSM21040056).

Heart rate variability monitor

Heart Rate Variability (HRV) related parameters were measured using a Finnish-made HRV monitor (Model: Firstbeat Sports) with a sampling frequency of 250 Hz.

Pittsburgh sleep quality index

The Chinese version of the Pittsburgh Sleep Quality Index (PSQI) was used to assess subjective sleep quality. Higher scores indicate poorer subjective sleep quality. A score greater than 5 was used as the cut-off value to distinguish between good sleepers and poor sleepers, with a sensitivity of 89.6% and a specificity of 86.5%. The internal consistency was represented by a Cronbach’s alpha coefficient of 0.83 [26].

Morningness-eveningness questionnaire

The Morningness-Eveningness Questionnaire (MEQ) is a tool widely used to assess the natural tendencies of sleep-wake circadian rhythms. The Chinese version of the MEQ-5 has demonstrated good reliability and validity among Chinese secondary school students, with a Cronbach’s alpha coefficient of 0.74. The scale consists of 5 items—including rising time, fatigue time, optimal sleep time, peak performance time, and self-assessed chronotype—to evaluate an individual’s long-term chronotype. The total score for the scale ranges from 4 to 25. In this study, scores were classified as: MT (18–25 points), intermediate-type (12–17 points), and ET (4–11 points). The Cronbach’s alpha for this study was 0.68 [30].

Testing indicators

HRV parameters

The HRV parameters tested included time-domain indicators: the Standard Deviation of Normal-to-Normal intervals (SDNN) and the Root Mean Square of Successive Differences in NN intervals (RMSSD); and frequency-domain indicators: High Frequency (HF) and Low Frequency (LF).

Sleep quality measure

The PSQI is a self-report questionnaire used to assess sleep quality and disturbances over the past 4 weeks. It consists of 18 items aggregated into seven components: subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleep medication, and daytime dysfunction. Participants are instructed to answer each item on a scale of 0 to 3. The scores from each component are summed to yield a global score for subjective sleep quality (range 0–21).

Data analysis

The collected R-R interval data were exported via Polar software and transferred to a computer, then imported into Kubios HRV Standard analysis software. Artifacts were corrected using a ‘medium’ threshold setting (0.25 s) [28]. Detrending was performed using the smoothness priors method, with a smoothing parameter of 500 and a cutoff frequency of 0.035 Hz [29]. The data were then imported into Kubios HRV Standard analysis software for Fourier transform. A 5-minute signal sample was selected for frequency-domain and time-domain analysis. After performing a Fast Fourier Transform (FFT), the HRV results were categorized into different indices [28, 29].

The frequency-domain indicators included: High Frequency (HF) (0.15–0.4 Hz), reflecting vagal tension; and Low frequency (LF) (0.04–0.15 Hz), reflects a mixture of both sympathetic and parasympathetic activity; The time-domain indicators included: the Root Mean Square of Successive Differences in NN intervals (RMSSD), representing vagal tension; and the Standard Deviation of Normal-to-Normal intervals (SDNN), representing total HRV, Given the controversy and ambiguity surrounding the LF/HF ratio, this study decided not to include it as a metric [30].

Statistical analysis

All data were statistically analyzed using SPSS version 27.0. First, all data were assessed for normal distribution. The Mann-Whitney U test was applied to data that did not conform to a normal distribution. A 2 (group: MT, ET) × 2 (intervention: active tPCS, sham tPCS) × 2 (time: pre-test, post-test) three-factor repeated measures ANOVA was conducted on the HRV indices and PSQI scores. ‘Group’ and ‘intervention’ were treated as between-subject factors, and ‘time’ was treated as the within-subject factor. A P-value < 0.05 was considered statistically significant, and a P-value < 0.01 was considered highly significant. All data are expressed as mean ± standard error (SE).

Results

Pre-intervention baseline data

Prior to the tPCS intervention, inter-group homogeneity tests were conducted on the HRV and PSQI scores for participants within each of the two sleep types (i.e., comparing experimental vs. control groups separately for MT and ET). The results indicated no significant differences between the groups at baseline. The specific findings were: PSQI (MT: t = 0.705, p = 0.486; ET: t = 0.723, p = 0.476); RMSSD (MT: t = 0.383, p = 0.504; ET: t = -0.766, p = 0.450); SDNN (MT: t = -0.311, p = 0.758; ET: t = -0.971, p = 0.340); HF (MT: t = 0.782, p = 0.441; ET: t = -0.672, p = 0.507); and LF (MT: t = 0.677, p = 0.504; ET: t = 1.185, p = 0.246). Therefore, the groups were considered homogeneous.

Changes in PSQI indicators

Following the intervention, participants in the EXP of both sleep types exhibited a downward trend in PSQI total scores. Analysis of variance (ANOVA) indicated a significant interaction effects were observed for Intervention type× Time [F(1, 56) = 8.415, P = 0.007, = 0. 0.231] and Group × Intervention type [F(1, 56) = 4.486, P = 0.043, = 0.138], along with significant main effects for Intervention type(See Table 2). Significant main effects were observed for intervention type (See Table 2). Simple effect analysis revealed that post-intervention, the PSQI difference in the MT group were higher than in the ET group. Specifically, the EXP exhibited significantly lower scores than the CON in both MT group [95% CI (2.576, 4.491), P < 0.001] and ET group [95% CI (1.042, 2.957), P < 0.001]. Within the EXP, the MT group scored significantly lower than the ET group [95% CI (0.216, 2.451), P = 0.021]. (See Fig. 3; Table 3).

Table 2.

Analysis of variance results for PSQI in Student-Athletes after tPCS

Indicator		df	F	P	Partial η2
PSQI	Group	56	2.423	0.131	0.080
	Intervention type		10.802	0.003	0.278
	Time		0.286	0.597	0.010
	Group * Intervention type		4.486	0.043	0.138
	Group * Time		0.472	0.498	0.017
	Intervention type * Time		8.415	0.007	0.231
	Group * Intervention type * Time		0.210	0.650	0.007

Fig. 3.

Effects of tPCS intervention on HRV and PSQI in Student-Athletes

Table 3.

Effects of tPCS intervention on PSQI in Student-Athletes

Group	Intervention type	Pre-intervention	Post-intervention	Difference	Confidence intervals	Cohen’s d	Percentage
MT	Experimental	9.533	5.600	-3.933	[-4.711, -3.155]	-3.382	-41.259%
	Control	9.133	9.200	0.067	[-0.436, 0.570]	0.065	0.730%
ET	Experimental	9.333	6.933	-2.400	[-3.178, -1.622]	-1.392	-25.714%
	Control	8.933	8.667	-0.267	[-0.703, 0.303]	-0.285	-2.985%

Note:

Changes in HRV parameters

Time-domain indicators

Following the intervention, participants in the EXP showed an increasing trend in both RMSSD and SDNN scores. ANOVA indicated no significant three-way interaction for either index. However, a significant Intervention type × Time interaction was found for both RMSSD [F(1, 56) = 48.424, P < 0.001, = 0.633] and SDNN [F(1, 56) = 18.561, P < 0.001, = 0.399]. Significant main effects for intervention type and time were also observed (see Table 4).

Table 4.

Analysis of variance results for HRV indicators in Student-Athletes after tPCS

Indicator		df	F	P	Partial η2
RMSSD	Group	56	5.776	0.023	0.171
	Intervention type		34.065	<0.001	0.549
	Time		14.885	<0.001	0.347
	Group * Intervention type		0.288	0.595	0.010
	Group * Time		0.309	0.583	0.011
	Intervention type * Time		48.224	<0.001	0.633
	Group * Intervention type * Time		0.838	0.368	0.029
SDNN	Group	56	6.883	0.014	0.197
	Intervention type		24.092	<0.001	0.462
	Time		25.717	<0.001	0.479
	Group * Intervention type		0.022	0.883	0.001
	Group * Time		0.665	0.422	0.023
	Intervention type * Time		18.561	<0.001	0.399
	Group * Intervention type * Time		0.184	0.672	0.007
HF	Group	56	4.389	0.045	0.136
	Intervention type		38.139	<0.001	0.577
	Time		46.015	0.000	0.622
	Group * Intervention type		0.200	0.658	0.007
	Group * Time		6.312	0.018	0.184
	Intervention type * Time		33.023	<0.001	0.541
	Group * Intervention type * Time		5.479	0.027	0.164
LF	Group	56	4.028	0.054	0.126
	Intervention type		10.314	0.003	0.269
	Time		89.466	0.000	0.762
	Group * Intervention type		0.712	0.406	0.025
	Group * Time		32.565	<0.001	0.538
	Intervention type * Time		58.314	<0.001	0.676
	Group * Intervention type * Time		5.156	0.031	0.156

Simple effect analysis revealed that post-intervention, the RMSSD and SDNN differences in the MT group were lower than those in the ET group. EXP showed significantly higher RMSSD and SDNN scores than the CON across both Chronotypes. RMSSD: Significant differences were observed in both the MT group [95% CI (12.513, 36.900), P < 0.001] and the ET group [95% CI (-37.000, -10.898), P < 0.001]. SDNN: Significant differences were observed in both the MT group [95% CI (-42.457, -12.624), P < 0.001] and the ET group [95% CI (-44.949, 15.138), P < 0.001].

However, within the EXP, no significant difference in scores was found between the MT and ET groups for RMSSD [95% CI (-19.618, 6.022), P = 0.127] or SDNN [95% CI (-24.462, 9.091), P = 0.356] (See Fig. 3; Table 5).

Table 5.

Effects of tPCS intervention on HRV in Student-Athletes

Group	Intervention type	Indicator	Pre-interventio	Post-intervention	Difference	Confidence intervals	Cohen’s d	Percentage
MT	Experimental	RMSSD	47.514	68.171	20.657	[10.898, 30.416]	1.012	43.476%
		SDNN	61.395	87.774	26.378	[12.419, 40.338]	0.923	42.965%
		HF	917.267	1093.733	176.467	[145.246, 207.686]	3.035	19.238%
		LF	1947.600	1664.733	-282.867	[-381.003, -184.730]	-3.496	-14.524%
	Control	RMSSD	45.208	43.464	-1.744	[-11.759, 8.271]	-0.084	-3.858%
		SDNN	63.218	60.222	-2.996	[-16.123, 10.130]	-0.118	-4.740%
		HF	842.533	860.267	17.733	[-111.938, 147.405]	0.111	2.105%
		LF	1843.667	1927.467	83.800	[-63.761, 231.361]	0.191	4.545%
ET	Experimental	RMSSD	34.132	61.373	27.241	[17.481, 37.000]	1.675	79.810%
		SDNN	45.987	80.089	34.102	[20.141, 48.061]	1.42	74.156%
		HF	615.467	1015.267	399.800	[368.579, 431.020]	6.675	64.959%
		LF	2451.933	1711.133	-740.800	[-838.936, -642.663]	-2.968	-30.213%
	Control	RMSSD	37.503	35.541	-1.962	[-8.053, 4.129]	-0.395	-5.232%
		SDNN	51.808	50.046	-1.762	[-14.888, 11.364]	-0.435	-3.401%
		HF	670.667	693.533	22.867	[-106.805, 152.538]	0.580	3.410%
		LF	2249.400	2185.600	-63.800	[-211.361, 83.761]	-0.350	-2.836%

Frequency-domain indicators

Following the intervention, participants in the EXP showed an increasing trend in HF scores, while decreasing trend in LF scores. Unlike the time-domain indices, a significant Group × Intervention type × Time interaction was observed for both HF [F(1, 56) = 5.479, P = 0.027, = 0.164] and LF [F(1, 56) = 5.156, P = 0.031, = 0.156]. For both indices, significant interaction effects were observed for Intervention type× Time and Group × Time, along with significant main effects for Intervention type and Time (see Table 4).

Simple effect analysis revealed that post-intervention, the HF and LF differences in the MT group were lower than those in the ET group. HF scores in the EXP were significantly higher than those in the CON across both chronotypes, while LF scores were significantly lower. HF: Post-intervention scores were significantly higher in the EXP compared to the CON for both the MT group [95% CI (-365.072, -101.862), P = 0.001] and the ET group [95% CI (-453.338, -190.128), P < 0.001]. LF: Post-intervention scores were significantly lower in the EXP compared to the CON for both the MT group [95% CI (71.200, 454.266, P = 0.009) and the ET group [95% CI (282.934, 666.000), P < 0.001].

However, within the EXP, no significant difference in scores was found between the MT and ET groups for HF [95% CI (-480.595, 123.005), P = 0.417] or LF [95% CI (-298.001, 390.801), P = 0.785] (See Fig. 3; Table 5).

Discussion

This study explored the intervention effects of tPCS on Student-Athletes with MT and ET sleep disorders, as well as the potential underlying physiological mechanisms. The findings demonstrated that tPCS effectively improved the sleep quality of Student-Athletes; however, the improvement was more pronounced in MT Student-Athletes compared to their ET counterparts. This finding not only confirms the potential of tPCS in improving sleep but, more importantly, highlights that chronotype, as an individual biological characteristic, is a significant variable influencing the efficacy of tPCS in improving sleep quality. This discovery provides crucial empirical support not only for the intervention of sleep disorders but also for the development of personalized interventions and sleep management strategies for athletes.

Effects of tPCS on subjective sleep quality in different sleep disorder chronotypes

In this study, the participants’ baseline PSQI scores averaged around 9, which is markedly higher than the normal threshold, reflecting the prevalent sleep problems among Student-Athletes. This high prevalence is associated with multiple factors, such as shortened sleep duration due to morning training, academic pressure, and pre-sleep hyperarousal caused by competition anxiety [30]. For MT Student-Athletes, their biological rhythms are relatively synchronized with the university schedule. They possess a normal biological clock phase, and their physiological systems are in a relatively “synchronized” and “efficient” state when coping with academics and training [31]. However, this study found that the tPCS intervention exhibited differential effects on sleep disorders between the two chronotypes: the improvement margin in PSQI for the ET experimental group (25.714%) was significantly smaller than that for the MT experimental group (41.259%). We speculate that this stems from the complexity of ET sleep disorders, which is not merely a physiological phase delay but also a form of persistent stress triggered by the conflict between the biological clock and the social environment. First, this difference may stem from the subjective nature of the PSQI scale and the persistence of “Social Jet Lag”. The PSQI is a subjective assessment of sleep quality, with scores closely related to psychological perceptions and daytime function rather than purely reflecting physiological sleep parameters [32]. The endogenous period of the biological clock in ET individuals often exceeds 24 h, leading to a physiological phase delay of 2–6 h [33, 34]. However, university courses and training schedules (e.g., morning training) force them to violate their biological clock [35], a conflict defined as Social Jet Lag [15]. Social Jet Lag not only causes sleep deprivation but is itself a persistent psychological stressor, inducing psychological distress such as anxiety and daytime fatigue [36, 37]. The limited efficacy of the tPCS intervention for ET sleep disorders in this study lies in its inability to eliminate the rigid external environmental pressures that cause Social Jet Lag. Consequently, even if the tPCS intervention improved the physiological aspects of nighttime sleep in ET participants (e.g., autonomic nervous system), they still faced the immense pressure of “forced early rising” each morning. This persistent environment-biology conflict weakened their overall subjective perception of sleep quality improvement, ultimately resulting in a limited reduction in their PSQI scores [38]. Second, this difference may also stem from the physiological limitations of tPCS. The biological root of ET sleep disorders lies in the functional characteristics of a deep brain structure [39]. A fundamental adjustment of the biological clock phase requires modulation of the suprachiasmatic nucleus (SCN). However, tPCS primarily acts on the cerebral cortex and has a limited direct impact on the SCN [40]. It is therefore unlikely to fundamentally correct the SCN-driven biological clock phase delay within 4 weeks [41]. Although indirect neural projections exist between the prefrontal cortex and the SCN [42], this top-down modulation may be insufficient to counteract the robust endogenous rhythms of ET individuals.

Therefore, the final results indicated that tPCS improved the PSQI scores of both MT and ET Student-Athletes; however, the magnitude of PSQI improvement in the ET experimental group was smaller than that in the MT group.

Effects of tPCS on the autonomic nervous system in different sleep disorder chronotypes

HRV reflects the dynamic balance between the sympathetic and parasympathetic nervous systems [43]. Analysis of variance (ANOVA) indicated that for HRV, the main effects of group, intervention type, and time were significant, and the group × intervention type × time interaction effect was also significant. The HRV data from this study not only confirm the intervention effect of tPCS but, more importantly, reveal fundamental differences in autonomic nervous system (ANS) regulation between Student-Athletes of different chronotypes.

Pre-intervention data showed that the baseline parasympathetic indices (RMSSD, HF) of the ET group were significantly lower than those of the MT group. This suggests that the ANS of ET participants was already in a state of long-term imbalanced stress, a finding consistent with previous research [44, 45]. The academic and training pressures superimposed on the “Social Jet Lag” context may amplify ANS dysfunction in ET Student-Athletes. Following the tPCS intervention, the ET’s parasympathetic indices (e.g., RMSSD, HF) were significantly enhanced, whereas the CON showed no significant changes. More importantly, post-intervention, the RMSSD, SDNN, and HF values of the MT group were higher than those of the ET group; conversely, their LF was lower than the ET group. This confirms that tPCS has a stronger activating effect on the parasympathetic nervous system in the MT group. In contrast, ET individuals typically exhibit higher sympathetic tone and lower parasympathetic tone [46]. Although the tPCS intervention was able to improve this imbalanced state to some extent, the magnitude and durability of its effect were relatively limited. The reason may be that long-term circadian misalignment can cause the ANS to lose this flexibility, making it difficult to effectively switch to a parasympathetic-dominant recovery mode even during rest. The modulation of the ANS by tPCS is considered the core mechanism for its improvement of sleep. The Central Autonomic Network (CAN) is a complex network composed of a series of interconnected structures in the brain (e.g., prefrontal cortex, insula, amygdala), responsible for the top-down regulation of ANS activity [43]. The tPCS intervention thus shifts the balance point of the ANS toward the “rest-and-digest” mode. This result is highly consistent with the “hyperarousal hypothesis” of insomnia [46]. In fact, the ANS and sleep are bidirectionally regulated: sleep disorders lead to sympathetic over-activation, and ANS imbalance, in turn, disturbs sleep, forming a vicious cycle [18]. The key utility of the tPCS intervention lies in breaking this cycle. It restores ANS balance by enhancing parasympathetic activity, thereby initiating a virtuous cycle: tPCS stimulation enhances parasympathetic activity → sleep quality improves → autonomic nervous function is further optimized → sleep continues to improve [47].

Although the core mechanisms of MT and ET sleep disorders differ, both manifest ANS dysfunction, albeit with different specific patterns. MT disorder is primarily characterized by sustained sympathetic over-activation and insufficient parasympathetic activity, consistent with a hyperarousal state [48]. ET disorder, in contrast, exhibits an abnormal ANS circadian rhythm, including delayed parasympathetic activation at night and insufficient sympathetic activation upon waking [49]. The HRV data from this study support this view; the ET group’s baseline RMSSD, SDNN, and HF were significantly lower than the MT group’s, indicating a more severe ANS imbalance in the ET group. tPCS stimulates the prefrontal cortex, enhancing its modulation of the central autonomic network, a pathway that is effective in both sleep disorder types [50]. Although the ET group’s biological clock phase issue was not fundamentally resolved, their autonomic nervous function markedly improved following the tPCS intervention. While these results highlight the efficacy of tPCS, the potential influence of individual demographic characteristics, such as age and weight, cannot be entirely ruled out. Existing literature indicates that HRV indices, such as RMSSD and SDNN, decline with age and are negatively correlated with increased Body Mass Index; Furthermore, younger individuals typically exhibit greater neuroplasticity, which may render their autonomic nervous system more sensitive to the neuromodulatory effects of tPCS [51, 52]. Although the physiological stability characteristic of the 18–22-year-old student-athlete cohort in this study may mitigate these effects, these demographic factors could theoretically still modulate individual responses to tPCS intervention. Analyzing these potential influencing factors contributes to a deeper understanding of the intervention effects of tPCS on autonomic regulation. This also suggests a viable direction for future research: specifically, conducting subgroup analyses based on individual attributes when investigating the impact of tPCS on the ANS across different sleep disorder chronotypes.

In summary, tPCS is an effective intervention for improving both subjective sleep and objective physiological recovery in Student-Athletes with sleep disorders. This study confirmed that the efficacy of tPCS for MT sleep disorder Student-Athletes is superior to that for ET. The superior effect in the MT group may stem from their better circadian synchronization, ANS flexibility, and neurotransmitter homeostatic regulation. Conversely, the efficacy in the ET group was relatively limited, suggesting that the persistent physiological and psychological stress caused by “Social Jet Lag” collectively restricts the upper limit of the tPCS intervention’s effectiveness.

Differential analysis of intervention effects across sleep disorder chronotypes

In fact, an intrinsic link exists between the improvement in HRV parameters and the reduction in PSQI scores. High-quality sleep relies on the adaptive modulation of the ANS across different sleep stages [18]. Sleep disorders are frequently accompanied by ANS imbalance, characterized by insufficient nocturnal parasympathetic activity and sustained sympathetic hyper-activation [53]. In the present study, although the tPCS intervention simultaneously improved ANS function and PSQI scores in both groups, thereby enhancing sleep quality, it revealed a central paradox: the magnitude of HRV improvement in the ET group was significantly greater than that in the MT group, whereas their PSQI improvement was, conversely, smaller. This section aims to explore the potential mechanisms underlying this desynchronization between the improvement of physiological indicators and subjective perception.

The PSQI assesses subjective sleep quality, encompassing multiple dimensions such as sleep latency, sleep efficiency, and sleep disturbances [32]. The core problem in ET sleep disorder is biological clock phase delay; thus, even if ANS function improves, the sleep phase mismatch persists. In contrast, MT sleep disorder primarily manifests as hyperarousal and autonomic imbalance. The tPCS intervention, by modulating cortical excitability and autonomic function, directly targets this pathophysiology, thereby yielding a greater magnitude of PSQI improvement than in the ET group. Dissimilar to the PSQI changes, the ET group exhibited a greater magnitude of improvement in HRV indicators following the tPCS intervention. This phenomenon can be explained by the “ceiling effect”: a lower baseline level allows for greater room for improvement [54]. The MT group’s baseline HRV was already relatively high, limiting the potential for further enhancement; the ET group had a lower baseline and thus more room for improvement, although their absolute post-intervention values may still be lower than the MT group. Beyond baseline differences, this paradox in the ET group can be further explained through the lens of “physiological recovery versus perceived recovery.” Although the tPCS intervention successfully modulated the parasympathetic nervous system by stimulating the prefrontal cortex, significantly increasing RMSSD and HF indices, the subjective PSQI is influenced not only by physiological drive but also by cognitive load, anxiety, and dysfunctional beliefs about sleep [38, 55]. For ET student-athletes, “social jet lag” creates a misalignment between their internal biological arousal timing and daytime daily schedules. Consequently, despite the objective restoration of parasympathetic tone, they still experience significant sleep inertia and negative emotional experiences when forced to rise early. This persistent sleep dysfunction fosters a “cognitive bias” that desensitizes their subjective evaluation of sleep improvement, leading to an evaluative discrepancy between physiological recovery and psychological perception. As a result, they tend to rate their overall sleep quality lower even as physiological sleep parameters improve [56]. This phenomenon signifies a “recovery lag,” wherein the physiological recovery of the ANS precedes the cognitive perception of sleep quality. While tPCS effectively improves sleep quality at the physiological level, the “perceived recovery” of ET individuals remains masked by the sleep disturbances inherent in circadian misalignment [55].

In recent years, growing evidence indicates that the effects of non-invasive brain stimulation techniques are not constant throughout the day but are highly dependent on the brain’s physiological state at the time of stimulation [57]. Cortical excitability, neurotransmitter levels, and the hormonal milieu (e.g., cortisol and melatonin) all follow strict 24-hour rhythmic fluctuations [58]. These fluctuations collectively create certain “optimal windows” during which the brain is most sensitive and susceptible to externally induced neuroplastic changes. Conversely, during “suboptimal windows,” the brain may resist or even show adverse effects to the same stimulation. This concept, known as the “circadian–homeostatic ‘window of efficacy’,” integrates evidence from chronobiology, sleep research, and non-invasive brain stimulation, positing that non-invasive brain stimulation can only produce reliable, positive effects when applied within specific time windows [59]. This study set the intervention time at 18:00–19:00, a time point that holds entirely different biological significance for MT and ET Student-Athletes [60]. For MT individuals with sleep disorders, 18:00–19:00 corresponds to their biological evening. At this time, their circadian system’s wake-promoting signals are waning, cortisol levels (promoting wakefulness) have dropped to a diurnal low, and melatonin (initiating sleep) may be starting to secrete [61]. The physiological system is “prepared” to enter a state of rest and recovery. Applying a tPCS stimulus aimed at enhancing parasympathetic activity and promoting relaxation at this juncture means the external stimulation aligns with the internal biological rhythm. This “in-phase” alignment synergistically enhances the physiological drive toward sleep, producing a maximal effect. However, for ET individuals with sleep disorders, the same 18:00–19:00 window is often their “wake-maintenance zone,” when they are most alert and at their peak cognitive performance [62]. Their cortical excitability may be at or near its diurnal peak [57]. Studies have found that ET individuals typically exhibit higher sympathetic tone and lower HRV during the daytime, especially in the evening [44]. Applying a “sedating” and “relaxing” tPCS stimulus at this moment of peak physiological arousal constitutes a “Chrono-physiological Mismatch.” The external inhibitory signal conflicts with and antagonizes the strong internal excitatory drive. This temporal mismatch partially explains the differential intervention effects observed between the two sleep disorder types, despite identical tPCS parameters. For instance, Frase et al. found that applying anodal tDCS (aimed at increasing cortical excitability) to awake subjects before sleep did not promote sleep; instead, it significantly reduced total sleep time and increased arousal metrics—a classic example of a mismatch between external stimulation and the brain’s internal state [63]. Synthesizing the above, this study proposes a core explanatory framework: the efficacy of tPCS is constrained by a “Window of Efficacy for Neuromodulation,” which is jointly determined by circadian rhythms and sleep homeostasis. The differential effects observed in this study are, in essence, a direct reflection of the “match” (for MT) versus “mismatch” (for ET) between the fixed intervention time (18:00–19:00) and the intrinsic biological rhythms of the two Student-Athlete chronotypes. For MT Student-Athletes, the standard tPCS protocol (18:00–19:00, 1.5 mA, 30 min) demonstrated good efficacy and can be considered a first-line intervention strategy. However, for ET Student-Athletes, the intervention window may need to be adjusted to better match their physiological characteristics. For example, considering that the melatonin peak in ET individuals is typically delayed by 2–3 h, adjusting the tPCS intervention time to 20:00–21:00 might be more appropriate [33]. Furthermore, appropriately increasing the stimulation intensity (e.g., to 2.0 mA) or extending the duration (e.g., to 45 min) might also help enhance efficacy. However, such chronotype-based personalized adjustment strategies require further validation in future research.

The superior intervention effect of tPCS on MT Student-Athletes may also be related to prefrontal neural circuits. MT individuals typically exhibit stronger prefrontal cortex functional connectivity and higher executive control capacity, which provides a more favorable structural basis for the neuromodulatory effects of tPCS [64]. From a neuroanatomical perspective, the prefrontal-limbic circuit in MT individuals shows greater functional integration, particularly in the regulation of the HPA axis [65]. tPCS enhances prefrontal cortex function, strengthening the top-down inhibition of the HPA axis, shifting the autonomic balance toward parasympathetic dominance and sympathetic inhibition, lowering basal cortisol levels, and regulating its diurnal secretion pattern [66]. The Cortisol Awakening Response in MT individuals is also more regular and more responsive to external interventions, which may explain why tPCS produces greater sleep improvement in these individuals [67]. Furthermore, the pineal melatonin secretion pattern in MT individuals exhibits a higher diurnal amplitude and a more stable phase relationship. tPCS, by modulating the synchronized activity of suprachiasmatic nucleus neurons via the frontal lobe, can further enhance the circadian rhythmicity of melatonin secretion, thereby improving sleep quality [33]. In contrast, ET individuals, due to long-term “Social Jet Lag,” already have an ANS in an imbalanced state of high sympathetic activation and low flexibility; their melatonin secretion patterns also typically show lower diurnal amplitude and a delayed phase, which renders the modulatory effect of tPCS relatively limited [68].

In conclusion, although the four-week tPCS intervention improved the overall subjective sleep quality and HRV of Student-Athletes, the magnitude of this improvement was not uniform; rather, it was dependent on the individual’s chronotype. This partially validates our hypothesis that tPCS can serve as an effective intervention to alleviate the sleep problems in Student-Athletes caused by high-intensity training and academic pressure. The intervention’s effect on the ET group should not be interpreted as simply “less effective,” but rather as an “intervention protocol mismatch”. This implies that the continued application of an intervention that is misaligned with an individual’s biological rhythms may not be optimal. Furthermore, from a clinical application perspective, the intervention outcome for the MT group was positive, potentially enabling them to overcome their sleep disorders. Conversely, the ET group experienced limited improvement in sleep quality post-intervention, revealing a difference in the practical application value of tPCS. The emergence of this phenomenon shifts our interpretation of the results from “the intervention has differential efficacy” to “the intervention protocol may be inappropriate,” which deepens the significance of our study.

Study limitations

1. Optimization of the Intervention Protocol: This study implemented a standardized tPCS protocol involving fixed intensity and timing for all participants. However, considering the distinct circadian phases and neurobiological profiles associated with different chronotypes, a standardized protocol may not produce optimal intervention effects. This is particularly evident for evening-type individuals, who demonstrated limited responses in the current study; for this group, protocols incorporating a delayed intervention window (post-20:00) or enhanced stimulation intensity (> 1.5 mA) should be prioritized. Future research should design multi-arm randomized controlled trials to investigate personalized stimulation schemes, adjusting the intervention window or parameters to better align with individual biological rhythms and enhance the practical efficacy of tPCS.

2. Analysis of Confounding Factors: Although randomization ensured the statistical homogeneity of the primary outcomes at baseline, numerical differences in factors such as sex, age, weight, and training duration persisted across the groups. Theoretically, these factors serve as regulatory elements for physiological outcomes [69, 70]. Future large-scale randomized controlled trials should explicitly incorporate variables that may influence autonomic tone or subjective sleep perception—including sex, training years, age, training load, and competition period—as covariates or through subgroup analyses to further isolate the intervention effects of tPCS.

Conclusion

This study demonstrates that:

1. A 4-week tPCS intervention can improve sleep quality and increase parasympathetically-mediated HRV indices in Student-Athlete with sleep disorders.

2. Differential tPCS intervention time windows exist for Student-Athlete with sleep disorders based on their chronotype.

3. tPCS intervention administered between 18:00 and 19:00 yields greater improvement in sleep quality for the MT group compared to the ET group.

Abbreviations

tPCS

Transcranial pulsed current stimulation

MT

Morning-type

ET

Evening-type

HRV

Heart Rate Variability

PSQI

Pittsburgh Sleep Quality Index

RMSSD

Root Mean Square of Successive Differences

SDNN

Standard Deviation of NN intervals

HF

High Frequency

LF

Low Frequency

PFC

Prefrontal cortex

vmPFC

Ventromedial prefrontal cortex

dlPFC

Dorsolateral prefrontal cortex

HPA

Hypothalamic-pituitary-adrenal

TES

Transcranial Electrical Stimulation

ANS

Autonomic nervous system

Author contributions

Qingchang Wu: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Supervision, Visualization, Writing – original draft, Writing – review & editing. Weidong Zhu: Data curation, Investigation. Writing – review & editing. Hongfei Xu: Data curation, Investigation. Visualization. Jie Li: Data curation, Investigation. Visualization. Jian Liu: Data curation, Investigation. Supervision, Project administration. Tiancheng Yu: Data curation, Investigation, Supervision, Writing – review & editing.

Funding

No Funding.

Data availability

The datasets generated and/or analyzed during the current study are not publicly available due to privacy concerns and the sensitivity of the information, but are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study has been approved by the Ethics Committee of Nantong University (Approval No. 1; 16 November 2020). Informed consent was obtained from all subjects included in the study. All methods of this study were performed following relevant guidelines and regulations.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.