The effects of transcranial direct current stimulation techniques to modulate resting-state EEG and reduce SOL: A randomized controlled trial

Abstract

Objective

To investigate the effects of repetitive bilateral transcranial direct current stimulation (tDCS) on sleep of latency (SOL) and resting-state electroencephalography (EEG) in individuals.

Methods

Twenty-eight young adults, aged 21–25 years, with a mean of 22.96 ± 0.87 years, were recruited and randomly grouped into an experimental group(13 people) and a control group (15 people). Subjects in the experimental group received five awake-phase resting-state tDCS over a one-week period, with anodes placed bilaterally in the dorsolateral prefrontal cortex (F3, F4) and cathode in the left upper arm. The current intensity was 1.5 mA, and the stimulation time was 15 min. The control subjects received pseudo-stimulation, with only the beginning and end of the stimulation for 30 s to receive a gradual rise and fall of the current. The intermediate current intensity was 0 mA, and the rest of the treatment parameters and processes were the same as those in the experimental group. All subjects filled out a sleep diary every day during the experiment.

Results

Compared with the pre-intervention period, in the subjective scale results, the sleep of latency factor score in the subjective sleep quality assessment scores of the subjects in the experimental group was significantly lower (p = 0.001); in the EEG results, the theta band power in the midline regions of the brains of the subjects in the experimental group (especially in the central Cz location and the prefrontal cortex) was significantly elevated, whereas the control group did not show such a difference.

Conclusion

Transcranial direct current stimulation can significantly reduce the sleep of latency of individuals and enhance theta band power to promote changes in the brain from wakefulness to drowsiness, thus enhancing sleep quality. This enhancement may be due to enhancement of inhibitory executive function. In the future, neuromodulation technology is expected to be applied to insomnia patients, especially those whose main symptom is difficulty falling asleep.

Introduction

Sleep is one of the most basic physiological requirements of the human body to maintain normal life activities. The average amount of sleep required by a healthy adult is 7–9 h a day (Baranwal et al., 2023). Good quality sleep is not only important for restoring physical energy but also for maintaining brain function and performing normal daytime activities (Sen & Tai, 2023). Sleep deprivation affects cognitive function and can even lead to physical and psychological disorders such as depression, schizophrenia, multiple sclerosis and osteoporosis (Lo et al., 2016; Mignot, 2008; Sindi et al., 2020), especially brain disorders, such as Parkinson's, epilepsy and dementia (Iranzo, 2016; Malhotra, 2017; Wang et al., 2024). However, the prevalence of sleep disorders (e.g., difficulty falling asleep, poor sleep continuity) is high in today's society. An online cross-sectional survey of a Chinese population showed that >30 % of 40,000 people reported sleep problems, such as insomnia and daytime sleepiness (Zheng et al., 2024). Sleep of latency (SOL) is an important indicator of sleep function and represents the ease with which the body passes from wakefulness to sleep. The SOL of a normal person is typically within 30 min Ohayon et al. (2016). Electroencephalography (EEG)band segments are also a commonly used diagnostic modality for sleep status, and theta band segments are closely related to sleep. Dynamic changes in the frequency and power of the theta band in rats, both during sleep and wakefulness, reflect normal behavior, which provides evidence for the validity of EEG measurements of sleep (Leung, 1984). Previous studies have found that during nocturnal sleep, theta band segments vary across sleep periods and brain regions, indicating that theta band and sleep quality are highly correlated (Karakaş, 2020; Marawar et al., 2016; Tinguely et al., 2006; Vijayan et al., 2017). Under the regulation of sleep pressure, the sleep homeostasis regulation system has a strong ability to regulate the body's ability to obtain the appropriate amount of sleep according to sleep demand. It is like a balance that maintains the equilibrium between the awake state and the sleep state. And Finelli et al. (2000) suggest that theta activity in waking and slow-wave activity in sleep are markers of a common homeostatic sleep process. In rat experiments, researchers found that the hippocampus of rats showed strong dynamic oscillatory coupling at theta frequencies during wakefulness or rapid eye movement sleep (Montgomery et al., 2008). Sleep is often associated with cognitive activities and resting-state theta rhythms have been demonstrated to play an important role in cognitive control processes (Pscherer et al., 2020). In different intensities and types of cognitive tasks, the activated theta oscillatory network is different (Cooper et al., 2014), the magnitude of theta band increase is also different (Nigbur et al., 2011), and the above-mentioned changes in the task-state theta band may need to be based on the resting-state theta band (Pscherer et al., 2022). SOL can be measured by different subjective and objective methods, and EEG is a more accurate reflection of sleep quality. SOL and power of theta band are two very important observations in people with sleep disorders.

Existing methods to improve sleep mainly include medication, cognitive-behavioral therapy for insomnia (CBT-I) (Riemann et al., 2017), sleep hygiene education, light therapy, meditation, and acupuncture. However, all of the above methods have the drawbacks of long treatment duration, slow effect, and high side effects (Hall-Porter et al., 2014; Morin et al., 2009). Therefore, it is particularly important to discover new and better treatments. Noninvasive brain stimulation (NIBS) refers to the use of electric or magnetic fields to stimulate the cerebral cortex to improve brain function, with the advantages of non-invasive, painless and easy to operate. NIBS mainly includes transcranial electrical stimulation (TES) and transcranial magnetic stimulation (TMS). With the development of NIBS, transcranial direct current stimulation (tDCS), the most widely used form of TES, has gradually been applied to the treatment of sleep disorders and is becoming increasingly widely used in the clinic (Antal et al., 2017; Rossi et al., 2009). tDCS refers to the use of two electrodes (anode and cathode) placed on the scalp to apply a constant and weak direct current (generally in the range of 0.5–2 mA) through the head, thereby modulating target cortical neurons. This modulates the membrane potential of the target cortical neurons, leading to depolarization or hyperpolarization of the neurons (Yavari et al., 2017), with the anodic electrode increasing the excitability of the lower cortical regions and cathodic electrode, thereby decreasing the excitability (Nitsche et al., 2003; Nitsche & Paulus, 2000, 2001). tDCS produces neuroplasticity effects with a duration ranging from a few minutes to 24 h, depending on the protocol used (Stagg et al., 2018). Numerous studies have shown that tDCS is safe when administered to humans for up to 20 min at a time (Reckow et al., 2018; Villamar et al., 2013).

The current study found that tDCS is effective in improving sleep quality and that this effectiveness is reflected in different aspects of subjective drowsiness, objective sleep metrics, EEG data and so on. A study of 20 healthy female youths showed that 5-Hz frontal oscillatory anodic tDCS resulted in effective changes in subjective sleepiness and spontaneous low-frequency EEG activity (D’Atri et al., 2016). Charest et al. (2019) also found a prolonging effect of tDCS on total subjective sleep duration. It has also been shown that tDCS has a significant effect on sleep efficiency (Saebipour et al., 2015). However, subjective sleep scores did not improve significantly, and a double-blind study of 90 patients with comorbid insomnia and major depressive disorder (MDD), who were subjected to 20 sessions of left-anodal and right-cathodal tDCS stimulation of the dorsolateral prefrontal cortex (DLPFC) site over a 4-week period, found that compared with the pseudo-stimulation group, the experimental group showed a greater improvement in depression and anxiety scores, greater increases in polysomnography (PSG)-recorded total sleep time(TST), and both objective and subjective sleep efficiency showed a stronger improvement (Zhou et al., 2020). Some studies have also found that tDCS did not improve sleep (Frase et al., 2019), which may be due to the heterogeneity of the different experimental protocols, but it also reflects, to some extent, that there are no clear criteria for tDCS protocols.

Wysokiński et al. (2023) showed that stimulation of the bilateral DLPFC regions induced an increase in theta, alpha, and beta band power. Amatachaya et al. (2015) also found that anodal stimulation significantly increased frontal cortex delta and theta power. It has also been reported that cathodic stimulation can increase delta and theta power in the frontal cortex (Ardolino et al., 2005; Donaldson et al., 2019; Mangia et al., 2014). However, some studies did not find the emergence of such an ameliorative effect, and no changes in resting-state EEG occurred before and after tDCS (Ghafoor et al., 2021; Gordon et al., 2018). However, more studies have arrived at the opposite conclusion. Two experiments showed that anodic tDCS to the right inferior frontal gyrus significantly and selectively reduced theta power in this region (Jacobson et al., 2011; Wirth et al., 2011). A study by Moliadze et al. (2015) showed that anodic stimulation of electrode position C3 increased the delta and theta powers; however, Donaldson et al. (2019) found a decrease in theta power under the same conditions. This shows that, thus far, there is no definitive conclusion on the alterations of resting-state EEG by tDCS.

The young population is a category of people with frequent sleep problems. A considerable number of studies have shown that sleep problems among young people in various countries are highlighted by the difficulty in falling asleep, that is, long sleep of latency (Lack, 1986; Siddiqui et al., 2016; Sweileh et al., 2011). A Meta-analysis of Chinese university students showed that more than a quarter of >80,000 people surveyed reported a sleep of latency longer than 30 min Li et al. (2017). However, no studies have addressed the effect of tDCS on sleep of latency. Although there are different conclusions about the changes in resting-state EEG before and after tDCS, there are also no reports of combining it with sleep detection indices. Therefore, we designed a randomized controlled trial to investigate the effect of tDCS on sleep improvement in a group of young adults. tDCS was performed five times during a one-week period, and changes in sleep quality and resting EEG, especially in the theta band, were observed before and after one week.

Materials and methods

Participants

We recruited 28 healthy adults, aged 21–25 years, mean (22.96 ± 0.87) years, in March 2023 at a school in Anhui, China. All participants signed an informed consent form to participate in the experiment, with the following recruitment criteria: (1) 16 years of age and older, (2) right-handed, and (3) Pittsburgh sleep quality index (PSQI) total score ≥5 or self-reported problems with sleep. The exclusion criteria were as follows:(1) history of psychiatric or neurologically related illness or major medical illness; (2) one or more concussions in the past year; (3) pregnant women; (4) taking psychoactive medications; (5) suffering from migraines, epilepsy, or other disorders; (6) participants who were dissatisfied with the study; (7) having pacemakers or other metallic implants in their body; and (8) suffering from eczema of the skin.

Participants were randomly assigned to either the experimental group (receiving active tDCS) or the control group (pseudo-stimulation), and each participant did not know which group they were assigned to or whether they had received tDCS. Allocation concealment continued until the end of the study. The participant recruitment process is illustrated in Fig. 1(a). 13 participants were included in the experimental group and 15 in the control group, with a male-to-female ratio within each group controlled at approximately 5:2.

Fig. 1.

(a) CONSORT study flow diagram. (i) Reasons: excessive artifacts in EEG data (n = 1). (b) Overview of study procedure.

Materials

Pittsburgh sleep quality index (PSQI)

The PSQI is widely used to assess sleep quality, and the Chinese version (Liu et al., 1996), revised by Xianchen Liu, was used in this experiment. The PSQI consists of 4 open-ended questions and 19 self-assessment items (0–3 scale), including 7 dimensions: subjective sleep quality, sleep of latency, total sleep time, sleep efficiency, sleep disorders, sleep medication use, and daytime dysfunction. The PSQI can be used to assess the quality of sleep and factors interfering with sleep over a one-month period. A PSQI total score of <5 was considered to indicate good sleep and a score of ≥5 was considered to indicate poor sleep, with a higher PSQI total score indicating poorer sleep quality. The Chinese version of the PSQI has been demonstrated to have good reliability (Cronbach's α = 0.84) and validity (factor loadings for each component: >0.5).

Sleep diary (SD)

SD is the most widely used and cost-effective method of subjective assessment and is the "gold standard" for subjective sleep assessment. By completing a SD, the participant could record his/her sleep behavior patterns and sleep-affected daytime conditions.

Depression, anxiety, and stress scale (DASS-21)

Originally designed to assess depression and anxiety, the DASS was later expanded to include a stress component that reflects emotional states such as irritability, tension, and agitation (Lovibond et al., 1995). The DASS-21, however, is an abbreviated version that facilitates quick responses from participants. The Chinese version of the DASS-21 consists of three seven-item subscales that assess depression, anxiety, and stress. Items are rated on a four-point Likert scale (from 0 to 3), with higher scores indicating higher levels of depression, anxiety, or stress. A simplified Chinese version of the DASS-21 was used in this study (Chan et al., 2012). The reliability of the DASS-21 was tested in different groups and all were found to have Cronbach's alpha values above 0.80.

Anxiety sensitivity index (ASI-3)

The original ASI consists of 16 items to assess the different components of fear of anxiety-related feelings. Taylor et al. (2007) developed the ASI-3, improving on the original scale to make it a multidimensional measurement tool with a stable factor structure The ASI-3 consists of three subscales: physical, cognitive, and social problems. It has been shown that the Chinese version of the ASI-3 has good validity and reliability and can be used as a valid tool to assess anxiety sensitivity. The Cronbach's alpha coefficient for the Chinese version of the total scale was 0.926, and the Cronbach's coefficients for the three subscales were 0.828–0.841 (Cai et al., 2018).

Methods

The electrical stimulation device used was a constant-current electrical stimulator (NVX 36T, Russia), which was controlled and operated with the accompanying software NeoRec via a USB port. The stimulation device delivered current through two 5 × 7 cm2 sponge electrodes, which were kept wet (with saline) during stimulation, so that the contact resistance was always lower than 10 kΩ. According to the international 10/20 system of EEG electrode placement, the anodes were placed bilaterally in the dorsolateral prefrontal cortex (F3, F4) during the experiments, and the cathode was placed in the left upper arm. The sponge electrodes were fixed by EEG caps, and the stimulation intensity of the experimental group was 1.5 mA and with a 30 s current intensity rise (at the beginning) and fall (at the end). The stimulation time of a single session was 15 min, while the control group was also set at a 30 s current intensity rise and fall at the beginning and end, respectively. No stimulation current passed through the rest of the experiment. During the application of stimulation, the impedance was detected in real time, and the current was disconnected when the impedance was greater than 50 kΩ to prevent skin burns owing to poor contact. This did not occur in this study. During a one-week period, each participant underwent tDCS five times: a maximum of one time per day to be completed within one week. Participants were instructed to remained seated during the tDCS and were not allowed to talk, close their eyes for long periods of time, or move substantially.

For the control group, the experimenter asked the participants if they felt an electric current to ensure the accuracy of the pseudo-stimulation. The entire experimental procedure was conducted in accordance with safety considerations for the application of tDCS (Lefaucheur et al., 2016). Some studies have shown that the possible side effects of tDCS include skin irritation and itching under the electrodes, and rarely, headache, fatigue, and nausea (Nitsche et al., 2008). No adverse events were reported during or after tDCS. The study was approved by the Ethics Committee of the Third Affiliated Hospital of the Naval Medical University under the approval number EHBHKY2024-H021-P001.

Procedures

The study procedure is shown in (b) of Fig. 1. Prior to the experiment, participants were asked to maintain their original routine during the week of the official experiment, not to stay up late, and not to take substances that affect sleep, such as coffee or tea. Each day, the experiment was conducted between 8:00 am and 9:00 pm, and every participant was asked to receive tDCS at the same time each day. Participants were asked to wash their hair before each experiment to ensure that their heads were clean, and the experimenter confirmed this before putting an EEG cap on the participant. Before the first tDCS session, we tested all participants for sleep quality and mood (pre-test). Participants were asked to come to the lab for tDCS 5 days out of a week, preferably for 5 consecutive days. The tDCS protocol was the same for each day. We placed all electrodes in saline for one minute before placing them on the scalp. According to the 10/10 International System, we placed the anodes in the left and right dorsolateral prefrontal lobes (F3 and F4) and the cathode in the left upper arm. In the experimental group, the tDCS current intensity was 1.5 mA, and the duration of a single stimulation was 15 min, with a ramp current presentation at the beginning and end of the stimulation; in the control group, only 30 s of current rise and fall at the beginning and end of the stimulation were retained, with no current passing through the middle 14 min. During the stimulation, the participants were asked to refrain from talking, sleeping, head bobbing, blinking vigorously, drinking, eating, standing up, and other behaviors that would affect tDCS. The participants were told to inform the researcher immediately if they experienced any discomfort during the experiment and to discontinue the experiment. The researcher also confirmed the safety of tDCS in the participants after each tDCS. After the five tDCS sessions, each participant's sleep quality and mood were again tested (post-test). The temperature and light in the laboratory were kept constant daily.

The acquisition of EEG data was conducted twice. The EEG cap consisted of a total of 19 leads, evenly distributed throughout the brain regions. The pre-test was conducted before the participant came to the laboratory for the first time and received tDCS; the post-test was performed after the participant came to the laboratory for the last time and received tDCS. Each acquisition lasted no <5 min, and the participants were required to keep their eyes open for the whole time, and could blink normally but could not close their eyes for a long time or even sleep.

Statistical analysis

EEG data were preprocessed using the Analyzer2 software. Resting-state EEG data were analyzed and re-referenced using whole-brain averaging. Filtering from 0.1–40 Hz was subsequently performed. Independent component analysis (ICA) was used to remove artifacts caused by blinking and/or eye movements. Finally, the data were manually inspected for remaining eye and muscle artifacts. Finally, the data were cut off every two seconds, wavelet transformed, and the results were averaged within a single subject. The data will be analyzed using SPSS software (version 26.0) and GraphPad Prism software (version 9.0), with a conservative p-value of 0.05. Repeated measures ANOVA was performed on the sleep quality parameters and mood indicators measured before and after the experiment. The EEG data of one participant in the control group had too many artifacts and were therefore deleted, leaving 14 people with data.

Results

First, we describe the demographic information of the participants (Table 1). The overall age ranged from to 21–25 years, with a mean of 22.96 ± 0.87 years.

Table 1.

Demographic information.

Item	tDCS	Control	Z	P
age (x ± s)	22.77±1.05	23.13±0.62	−1.255	0.209
Male[n( %)]	9 (69.23)	11 (73.33)	−0.235	0.814
working years	4.15±0.86	4.07±0.77	−0.444	0.657
PSQI(baseline)	4.92±2.27	4.93±1.84	−0.422	0.673
DASS(baseline)	20.46±24.96	17.20±19.84	−0.326	0.744
ASI(baseline)	13.54±14.69	15.60±13.09	−0.232	0.817

Descriptive statistics for sleep quality and specific parameters, including means and standard deviations, anxiety, depression, stress, and anxiety sensitivity levels in the experimental and control groups, are shown in Table 2 and Fig. 2.

Table 2.

Pre- and post-test descriptive parameters of sleep quality, mood, anxiety sensitivity and theta band power in each group.

Variable		Pre-test		Post-test		p
	Group	Mean	Standard deviation	Mean	Standard deviation
PSQI
Total	Experimental Control	4.92	2.27	4.46	3.48	0.337
		4.93	1.84	4.33	1.74	0.095
Subjective sleep quality	Experimental Control	1.23	0.70	1.08	0.92	0.337
		1.07	0.44	0.87	0.34	0.082
Sleep of latency	Experimental Control	1.62	0.74	0.77	0.89	0.002^⁎⁎
		1.27	0.93	1.20	0.91	0.774
Total sleep time	Experimental Control	0.23	0.42	0.38	0.62	0.165
		0.73	0.68	0.60	0.61	0.334
Sleep efficiency	Experimental Control	0.08	0.27	0.23	0.58	0.436
		0.27	0.57	0.33	0.47	0.582
Sleep disorders	Experimental Control	1.15	0.66	1.00	0.68	0.337
		1.07	0.25	0.93	0.44	0.334
Sleep medication use	Experimental Control	0.00	0.00	0.08	0.27	0.337
		0.07	0.25	0.00	0.00	0.334
Daytime Dysfunction	Experimental Control	0.62	0.62	0.62	0.62	0.999
		0.47	0.50	0.40	0.49	0.582
DASS-21
Total	Experimental Control	20.46	24.96	16.92	23.15	0.221
		17.20	19.84	11.87	12.45	0.200
Depression	Experimental Control	6.46	7.85	5.85	7.82	0.673
		5.73	7.15	4.53	4.81	0.358
Anxiety	Experimental Control	7.85	9.69	6.62	9.93	0.264
		7.33	9.02	4.93	5.36	0.255
Stress	Experimental Control	6.15	8.82	4.46	5.77	0.203
		4.13	5.39	2.40	3.03	0.144
ASI-3
Total	Experimental Control	13.54	14.69	9.15	10.96	0.143
		15.60	13.09	12.47	12.39	0.047*
Somatic concerns	Experimental Control	4.31	4.63	2.46	3.61	0.086
		4.00	3.90	3.67	3.89	0.442
Social concerns	Experimental Control	5.54	5.51	4.00	4.51	0.227
		7.33	5.92	5.33	5.07	0.038*
Cognitive concerns	Experimental Control	3.69	5.21	2.69	3.97	0.261
		4.27	4.12	3.47	3.79	0.125
Theta Band Power	Experimental	−0.01	0.03	0.23	0.30	0.002^⁎⁎
	Control	0.00	0.01	0.00	0.02	0.331

*p < 0.05, ** p < 0.01, *** p < 0.001.

Fig. 2.

Pre- and post-test parameters of sleep quality, mood, and anxiety sensitivity in each group.

Normality and sphericity tests of the data were performed before applying repeated-measures ANOVA. Repeated measures ANOVA showed a significant main effect of testing time for Sleep of Latency in the PSQI (p = 0.009) and a significant interaction between group and time(p = 0.022) (Table 3), so a further simple effects analysis was performed, yielding: in the pre-test, the simple effect of group was not significant, F_(1,26)=1.100, p = 0.304, Δ η2=0.041; in the post-test, the simple effect of group was not significant, F_(1,26)=1.480, p = 0.235, Δ η2=0.054; in the experimental group condition, the simple effect of time was significant (F_(1,26) = 12.993, p = 0.001, Δη2 = 0.333; and in the control group condition, the simple effect of time was not significant, F_(1,26) = 0.093, p = 0.763, Δ η2 = 0.004).

Table 3.

Repeated measures ANOVA.

Variable		Group				Time				Group * time
	F	p	df	Δη²	F	p	df	Δη²	F	p	df	Δη²
PSQI
Total	0.004	0.948	1	0.000	3.590	0.069	1	0.121	0.061	0.807	26	0.002
Subjective sleep quality	0.673	0.419	1	0.025	3.721	0.065	1	0.125	0.063	0.803	26	0.002
Sleep of latency	0.018	0.894	1	0.001	8.100	0.009**	1	0.238	5.907	0.022**	26	0.185
Total sleep time	2.712	0.112	1	0.094	0.014	0.907	1	0.001	2.752	0.109	26	0.096
Sleep efficiency	0.842	0.367	1	0.031	1.021	0.322	1	0.038	0.160	0.693	26	0.006
Sleep disorders	0.181	0.674	1	0.007	2.011	0.168	1	0.072	0.010	0.920	26	0.000
Sleep medication use	0.010	0.920	1	0.000	0.010	0.920	1	0.000	2.011	0.168	26	0.072
Daytime Dysfunction	0.922	0.346	1	0.034	0.090	0.766	1	0.003	0.090	0.766	26	0.003
DASS-21
Total	0.298	0.590	1	0.011	3.188	0.086	1	0.109	0.130	0.721	26	0.005
Depression	0.158	0.694	1	0.006	0.919	0.347	1	0.034	0.095	0.760	26	0.004
Anxiety	0.120	0.732	1	0.005	2.318	0.140	1	0.082	0.240	0.628	26	0.009
Stress	0.856	0.364	1	0.032	4.169	0.051	1	0.138	0.001	0.981	26	0.000
ASI-3
Total	0.311	0.582	1	0.012	6.177	0.020**	1	0.192	0.171	0.683	26	0.007
Somatic concerns	0.090	0.766	1	0.003	4.550	0.043**	1	0.149	2.192	0.151	26	0.078
Social concerns	0.644	0.430	1	0.024	5.855	0.023**	1	0.184	0.100	0.755	26	0.004
Cognitive concerns	0.174	0.680	1	0.007	3.613	0.068	1	0.122	0.045	0.834	26	0.002
EEG
theta band	6.439	0.026**	1	0.349	9.366	0.010**	1	0.438	8.957	0.011**	25	0.427

*p < 0.05, ** p < 0.01, *** p < 0.001.

A comparison of resting-state EEG before and after the intervention between the experimental and control groups revealed that after one week of tDCS intervention, the participants in the experimental group showed significant differences in the frontal regions in the theta band power, whereas similar results were not obtained in the control group (Figs. 3 and 4).The resting-state theta band powerof the participants in the experimental group was significantly negatively correlated with PSQI scale dimension 2 (sleep of latency) (Pearson r = −0.58,p = 0.03). Linear regression analysis showed that there was a significant linear relationship between resting theta band power and PSQI scale dimension 2 (sleep of latency) for participants in the experimental group, with resting theta band power explaining 33.6 % of the variability in PSQI scale dimension 2 (sleep of latency). The regression equation was Y = −1–2.505 X, which was statistically significant (F = 6.072, p = 0.03).

Fig. 3.

Resting-state EEG differences in the experimental group (a) and the control group (b) before and after tDCS intervention.

Fig. 4.

Differences in the theta band power in the experimental group (a) and in the control group (b) before and after tDCS intervention.

Discussion

Based on the existing research, this study designed an randomized controlled trial(RCT) to investigate the sleep improvement effects of tDCS in a young adult population. Five repetitions of tDCS over a one-week period improved sleep in a young adult population, particularly in terms of shortening sleep of latency. This finding is consistent with those of previous studies. Previous studies have shown that single or multiple tDCS sessions can improve sleep quality, for example by prolonging the duration of slow-wave sleep. Mood, on the other hand, did not change significantly during the experimental period, which indicates that the subjects' emotional state was relatively stable during the experiment and did not have a significant impact on the results, and that the tDCS technique does not improve sleep by improving mood but rather has a direct positive effect on sleep.

Gordon et al. (2018) conducted tDCS experiments on 22 healthy participants. They underwent tDCS five times each at five different parameter settings in a double-blinded randomized crossover design, and resting-state EEG was recorded before and after tDCS. Rs-EEG power spectral analyses showed no difference between baseline and post-stimulation comparisons under any tDCS condition. This finding differs from the conclusions of our study. Although the heterogeneity of the conclusions may be due to different experimental conditions, subjects, and other factors, a meta-analysis of EEG spectral power changes after tDCS in the prefrontal cortex has shown little consistency in results across studies (Horvath et al., 2014). After five sessions of tDCS within a week, we found that participants in the experimental group had significantly higher delta and theta band power in the midline of the brain (especially in the central Cz location and prefrontal cortex), whereas the power of the higher frequency bands, such as alpha and beta, did not change significantly. This may be due to the fact that modulation of oscillatory activity in the low-frequency bandwidths is easier compared with the high-frequency bands (Donaldson et al., 2019). In addition, theta waves often appear when people feel sleepy, suggesting that tDCS may have a short-term sleep-promoting effect after stimulation. After multiple repetitions of stimulation, this short-term effect expands into a long-term effect, which also shortens the sleep of latency of participants' nocturnal sleep. However, more refined experiments are needed to test the above hypotheses. For example, at least resting-state EEG measurements should be conducted before and after each intervention.

Our results showed significant changes in the resting-state EEG of participants in the theta frequency band before and after stimulation, which is consistent with previous findings (Diaz et al., 2016). Our results also showed that power changes in the theta frequency band in the central zone (Cz) and prefrontal areas were significantly negatively correlated with the PSQI scale factor 2 data. That is, higher theta band power is strongly associated with shorter sleep of latency, which also represents greater ease with which individuals move from wakefulness to sleep. It has been demonstrated that the frontal lobe is the first region in the whole brain to ‘fall asleep’ (De Gennaro et al., 2001; Marzano et al., 2013; Siclari et al., 2014). This region appears to be one of the targeted brain regions that influences sleep of latency. Several studies have found that chronic insomnia results in increased beta band activity and decreased low-frequency band activity (delta, theta, alpha) during sleep or NREM phases, reflecting hyperarousal of the central nervous system in patients with insomnia (Merica et al., 2001, 2001). During the wakefulness phase, insomnia patients exhibit a significant decrease in theta and alpha band energies and a significant increase in beta band power relative to healthy sleepers (Freedman, 1986; Wolynczyk-Gmaj & Szelenberger, 2011). In conclusion, insomnia patients showed elevated high-frequency activity and reduced low-frequency activity during both the wakefulness and sleep phases, suggesting that insomnia patients are in a state of 24-hour hyperarousal. tDCS can effectively elevate the low-frequency rhythmic activity of the brain, thus improving sleep.

In studies related to sleep and cognitive function, theta band has been suggested to be related to a variety of cognitive activities, such as memory, learning, response inhibition, and cognitive decision-making (Buzzell et al., 2019; Cavanagh & Frank, 2014; Daume et al., 2017; Huycke et al., 2021). Rhythmic sinusoidal-like waves of medium-high amplitude with a frequency of 4–8 Hz that occur in the midline region of the anterior head are known as the Frontal Midline Theta Rhythm (fmTheta). The medial prefrontal cortex is also a source of frontal midline theta band generation. It is likely that fmTheta will be observed in the EEG when performing specific cognitive tasks (Maurer et al., 2014; Messel et al., 2021). Frontal midline theta bands may reflect synchronized activity between the prefrontal cortex and other brain regions, having a synchronizing effect on neuronal firing in cortical areas (Myers et al., 2021). According to previous conclusions, the lower the baseline level of resting state theta, the poorer the response efficacy during the task state and the higher the theta activity required for activation (Pscherer et al., 2019). In contrast, the significant increase in resting-state fmTheta power found in our experimental results may suggest that such an alteration makes it possible for subjects to have less theta activity to activate when they need to exercise inhibitory functions and to achieve inhibitory effects with greater ease, or even better inhibitory effects, that is, a reduction in sleep of latency. This also suggests the need to focus on the relationship between inhibitory function and sleep.

The novelty of this study lies in the discovery of biomarkers specific to resting-state EEG in individuals after tDCS. After 5 sessions of tDCS, we found a power rise in the frontal midline theta rhythm (fmTheta), which, to our knowledge, has not been reported in previous studies. fmTheta has special biological significance in terms of cognitive control, synchronized activity between brain regions, and so on. However, this study has some limitations. First, the long-term effects of tDCS have not yet been explored in depth. After five tDCS sessions, only the participants' sleep quality was measured at the end of the experiment. Subsequent studies should also focus on the long-term effects of tDCS on sleep quality and explore the maximum duration of the effects of tDCS. Second, in future experimental designs, the stimulation parameters and group settings can be delineated more finely. For example, current intensity, stimulation frequency, and stimulation site of tDCS. The exploration of specific parameters can help to bring tDCS into play faster and better in future clinical practice. Third, the sample size of this experiment was small, and more convincing conclusions may be drawn if a larger-scale experiment is conducted in the future. Fourth, although the present study focused on the frequency range of the theta band, activities in the alpha and gamma bands have been shown to play an important role in cognition (Fries, 2009; Köster et al., 2014; Roux & Uhlhaas, 2013), and further experiments in the future could be carried out for more bands.

Conclusion

tDCS can significantly reduce the sleep of latency of individuals and enhance theta band power to promote changes in the brain from wakefulness to drowsiness, thus enhancing sleep quality. This enhancement may be due to enhancement of inhibitory executive function. In the future, neuromodulation technology is expected to be applied to insomnia patients, especially those whose main symptom is difficulty falling asleep.

Informed consent statement

Informed consent was obtained from all the participants involved in the study.

Institutional review board statement

This study was conducted in accordance with the Declaration of Helsinki and was approved by the Review Board.

Funding sources

This project was funded by the Joint Project of Hubei Provincial Natural Science Foundation, 'Research on Neurobiological Markers of Exercise Intervention in Improving Mood Disorders'. Reference ID: JCZRLH202500697.

Acknowledgements

Author Contributions: This study was conceived by Hao Lian, with Wenpeng Cai contributing to the design. Preparation of the original manuscript draft was conducted by Hao Lian, with reviewing and editing by Wenpeng Cai and Mengyang He. All the authors have read and agreed to the published version of the manuscript.