Evaluation of Mindful Breathing Method Using Tablet Devices Through Chaos Analysis and Frequency Analysis: A Pilot Randomized Controlled Trial

Abstract

Introduction:

University students face various stresses, including academic and career anxieties and a lack of interpersonal relationships. These stresses can elevate psychological burdens, negatively affecting their studies and daily lives.

Objective:

This pilot study aims to quantitatively evaluate the effects of mindful breathing exercises using tablet devices on autonomic nervous system activity in university students by analysis of finger plethysmogram (pulse wave amplitude values) and chaos analysis (Lyapunov exponent and fractal dimension).

Methods:

In this parallel-group randomized controlled trial, 18 nursing students (Mindful Breathing Group [Mi group], n = 9; control group [nMi group], n = 9) were randomly assigned. On the first day, the Mi group performed mindful breathing, the nMi group performed cross fixation, and finger plethysmograms were measured. For the next 9 days, the Mi group performed mindful breathing at home before bedtime, while the nMi group performed cross gazing, and finger plethysmograms were measured on days 1 and 9. Data were analyzed using one-way analysis of variance and t-tests.

Results:

The Mi group showed a significant increase in pulse wave amplitude values over time (P = .001), whereas the nMi group showed a decrease (P = .001). Chaos analysis revealed no statistically significant differences between groups in the fractal dimension or Lyapunov exponent. Although descriptive differences were observed, these did not reach statistical significance. Both groups demonstrated positive Lyapunov exponents, suggesting nonlinear characteristics of the pulse wave signals.

Conclusions:

Mindful breathing using tablet devices may be associated with changes in pulse wave amplitude in university students, which could reflect alterations in peripheral autonomic activity under the present experimental conditions. However, no statistically significant differences were observed in chaos analysis indices. Further research with larger samples and additional physiological measures is required to clarify the relationship between mindful breathing and nonlinear autonomic dynamics.

Trial Registration:

UMIN Clinical Trials Registry (UMIN000056166; Registered November 15, 2024)

Keywords

mindful breathing method finger plethysmogram frequency analysis chaos analysis autonomic nervous system (ANS)mindfulness-based stress reduction (MBSR)

Introduction

University students face important turning points in their lives, such as academics, career choices, and building relationships, and are exposed to tremendous stress from these factors.¹ In recent years, new stress factors such as the weakening of interpersonal relationships due to the COVID-19 pandemic have been added, further increasing the psychological burden on students.² These stresses can damage students’ psychological health and have negative effects such as poor academic performance and difficulty adapting to social life.

Review of Literature

Theoretical Framework

Research has shown that many university students suffer from mental health problems such as anxiety and depression. A study on Pakistani university students reported that 75%, 88.4%, and 84.4% of students had depression, anxiety disorders, and stress, respectively.³ On the other hand, the lifetime prevalence of anxiety disorders in the United States is reported to be approximately 34%.⁴ In Singapore, the lifetime prevalence of major depressive disorder among the population aged 18 years and older was reported to be 6.3%.⁵ Compared to these international data, it is suggested that the university environment may increase the risk of developing mental disorders. Similarly, a study of health service use among England university students revealed that one-thirds of students using services did not seek help for their issues,⁶ suggesting that a significant proportion of students deal with mental health issues alone. Consequently, the importance of effective stress management and self-management strategies is increasing.

In recent years, the use of mobile health applications has been gaining attention in mental health support. In a study addressing the mental health issues of Korean university students, a psychological intervention using a mobile health application significantly improved depression, anxiety, and stress according to subjective evaluations, highlighting its effectiveness in supporting university students’ mental health.⁷ This indicates that mobile devices can be valuable tools for supporting students’ mental health.

Furthermore, mindfulness is gaining widespread attention as part of mental health support. Mindfulness is defined as “living in the moment, being aware of the present,” and acknowledging experiences as they are without evaluation or judgment.⁸ It aims to cultivate moment-to-moment awareness and to detach from strong attachments to beliefs, thoughts, and emotions.⁹ Mindfulness practice has been shown to effectively reduce stress and improve mental health.¹⁰ Additionally, studies suggest it may induce neuroplastic changes in the structure and function of brain regions associated with attention, emotion, and self-awareness control.¹¹ It is used in clinical settings as a third-generation cognitive behavioral therapy for stress relief and brain function improvement.^12,13 Mindful breathing involves intentional regulation of respiration to anchor attention and modulate physiological states, linking cognitive focus with autonomic regulation.¹⁴ However, further research is needed to better elucidate the relationship between neuroscientific findings and behavioral outcomes.

Studies incorporating internet and mobile-based mindfulness interventions have reported effectiveness in improving pre-treatment symptoms, treatment expectations, and self-efficacy.¹⁵ These findings underscore the potential of mobile devices in delivering psychological interventions like mindfulness to university students.

Conventional mental health intervention studies have often relied on subjective assessment scales (questionnaires, etc.), and evaluation using objective physiological indicators has been limited. Analysis of finger plethysmogram is an effective, noninvasive method for evaluating peripheral circulatory dynamics and autonomic nervous system function. The waveform pattern of the fingertip plethysmogram reflects cardiac output, vascular elasticity, autonomic balance; high pulse wave amplitude (PWA) reflects vasodilation and reduced sympathetic activity, whereas low PWA reflects vasoconstriction.^16,17 Recently, in addition to conventional frequency analysis, autonomic nervous system function can be evaluated in greater detail through nonlinear methods such as chaos analysis.¹⁸ Moreover, the efficacy of stress evaluation using finger plethysmogram has been demonstrated, clarifying the relationship between psychological stress and physiological responses.¹⁹

The novelty of this study lies in the evaluation of mindfulness intervention using a tablet application and a comprehensive assessment of autonomic nervous system function that integrates conventional frequency analysis and nonlinear analysis. While existing studies have relied on a single analytical method, this study combines multiple methods to enable the detection of subtle physiological changes that were previously difficult to detect.

Objective

We aimed to reduce stress, improve mental health in university students, and objectively evaluate mindfulness interventions using finger plethysmogram. The integration of nonlinear analysis with conventional methods is expected to more accurately capture changes in autonomic nervous system function and provide objective measures of intervention effects.

Method

Design

The design of the study used a randomized controlled trial.

Research Question

In this study, we hypothesized that practicing mindful breathing using a tablet application will influence the characteristics of pulse wave frequency and chaos analysis. We aim to verify whether significant differences exist in these indicators between the group practicing mindful breathing and the group that does not.

Sample

In studies using finger plethysmogram measurements, an adequate sample size is necessary to ensure reliable results and stable effect size estimation. Previous studies using conventional frequency analysis have reported meaningful findings with relatively small samples. For example, analysis of finger plethysmogram data from 10 participants demonstrated that heart rate variability can objectively assess autonomic nervous system function.²⁰ Similarly, pulse wave analysis in 12 participants indicated its utility in evaluating arteriosclerosis,²¹ although it is noted that small sample sizes in such studies remain a primary limitation regarding generalizability.

Based on prior studies and the feasibility of controlled laboratory measurements with repeated daily sessions, this exploratory study included 18 participants. Eighteen female nursing students (aged 18-22) from the first to fourth year at Hyogo University were randomly assigned to a mindfulness implementation group (Mi, n = 9) or a control group (nMi, n = 9) using the RAND function in Microsoft Excel (Microsoft Corp). The allocation sequence was concealed from the investigator until the participants were assigned to their respective groups.

To evaluate the statistical basis of the design, a post-hoc power analysis was conducted using G*Power. The analysis indicated that the statistical power (1 − β) was 0.27, below the conventional threshold of 0.80. However, the observed between-group difference corresponded to an effect size of d = 0.68 (medium to large). Despite this effect size, the low statistical power presents limitations in the statistical reliability of the findings and may lead to an overestimation of the observed effect size. Therefore, the results should be interpreted as exploratory. I used the CONSORT reporting guideline to draft this manuscript, and the CONSORT reporting checklist when editing, included in the Supplemental Files. The present study serves as a pilot investigation providing parameters for estimating the sample size required for future confirmatory studies. Based on prior intervention studies,^20,21 the study design was considered sufficiently reliable for exploratory purposes.

Regarding the experimental structure, the overall research project extended for more than 1 year due to sequential participant recruitment and data management procedures. However, the intervention and physiological measurements for each participant were conducted over nine consecutive days. The 9-day intensive practice was designed to examine short-term physiological responses to mindful breathing in a high-stress academic environment.

Inclusion/Exclusion Criteria

Participants were selected based on the criteria of being in good physical and mental health, understanding the purpose of this study, and voluntarily agreeing to participate. Participants with prior experience in meditation or mindfulness or a history of mental illness were excluded to avoid confounding effects on experimental outcomes.

Institutional Review Board Approval

All participants provided written informed consent after receiving a full explanation of the study. Participants received a gift card at 2000 yen as a token of appreciation for their participation. The study was approved by the Ethical Review Committee of Hyogo University and conducted between May 2022 and December 2023 (21007). The study protocol was retrospectively registered with the UMIN Clinical Trials Registry (Registration No. UMIN000056166; Registered on November 15, 2024; URL: https://center6.umin.ac.jp/cgi-open-bin/ctr/ctr_view.cgi?recptno=R000064176).

Experimental Structure

From May 2022 to November 2023, each student participated in the experiment.

The experiment consisted of 2 components: “fixed-point measurement” and “daily measurement” (Figure 1). During the fixed-point measurement on the first day, the Mi group practiced mindful breathing for 5 minutes using a tablet device in a university laboratory, while the nMi group gazed at a cross displayed on a tablet device for 5 minutes under identical conditions. Finger plethysmograms were recorded for both groups during the task.

Figure 1.

Experimental configuration. Fixed-point measurement: measurement in the university laboratory; daily measurement: measurement at home; Mi: mindfulness group; nMi: nonmindfulness group. Validation: Mi group used a tablet device to practice mindful breathing; nMi group focused on the cross on the tablet device.

For the daily measurements, following the fixed-point measurement, the Mi group practiced mindful breathing at home every night before bed using a tablet device, while the nMi group continued gazing at the cross on the tablet device. For both groups, finger plethysmograms were measured for 5 minutes on the first day and again on the final day, the ninth day, of the daily measurements.

Mindfulness

In this study, an original application was developed using Microsoft Access (Microsoft Corp) to provide audio guidance for mindful breathing based on Mindfulness-Based Stress Reduction (MBSR) developed by Kabat-Zinn. On the first day, participants were given a verbal explanation of mindfulness in the laboratory and practiced mindful breathing following the audio guidance. The study incorporated mindfulness techniques, including meditation, breathing, and body scanning.⁸

Finger Plethysmogram

The fluctuation in aortic pressure caused by heart contraction propagates through the body as a pulse wave. The volume pulse wave records changes in blood vessel volume in the peripheral arteries.¹⁶ Previous studies have reported that stress-induced systolic blood pressure increases are due to the early return of the reflected pulse wave, which reduces coronary artery perfusion.²²

This study focuses on the effects of psychological intervention before sleep on the autonomic nervous system and examines changes in the pulse waveform. A photoplethysmograph, the Backs Advance (TAOS Research Institute), was used to measure the pulse wave amplitude of the second finger on the left hand. While pulse wave amplitude varies individually, we analyzed between-group differences to evaluate relaxation-induced vasodilation, rather than specific voltage ranges. The fingertip contains a high density of blood vessels, allowing clear observation of the pulse wave amplitude and making it a suitable site for non-invasive pulse wave measurement.²³

Finger plethysmogram recording was performed at a sampling frequency of 200 Hz, which provides sufficient reliability for analyzing pulse wave variability and physiological dynamics.²⁴ An FIR band-pass filter of 0.8 to 12.0 Hz was applied to the measurement data to eliminate signal fluctuations caused by body movement, following established signal processing guidelines.¹⁶

As shown in Figure 2, the maximum plethysmogram amplitude was calculated as the difference between the highest point (point b) and the lowest point (point a or e) of the wave height associated with ventricular contraction.¹⁶

Figure 2.

Index of finger plethysmogram. Pulse wave amplitude is defined as the vertical distance between the systolic peak (point b) and the adjacent baseline trough (point a or e). This index reflects the change in blood volume in the fingertip during a single pulse cycle.

Chaos Analysis

Chaos analysis was conducted using the chaos analysis program developed by the TAOS Research Institute. The Lyapunov exponent was calculated from the measured fingertip plethysmogram data using the Sano–Sawada method,²⁵ while the fractal dimension was determined via the correlation dimension method (Grassberger–Procaccia method²⁶).

Lyapunov Exponent

In chaos analysis, Lyapunov exponents and fractal dimensions serve as key indicators for quantitatively evaluating trajectory instability. These measures are crucial for assessing the predictability and instability of chaotic systems.²⁷

In 3-dimensional chaotic dynamical systems, small spheres of radius in phase space deform over time, generally transitioning to an elliptical shape. The logarithm of the rate of expansion or contraction along each principal axis of this ellipse is referred to as the Lyapunov exponent. The collection of multiple Lyapunov exponents forms the Lyapunov spectrum, which characterizes the complexity of chaos analysis. The Lyapunov spectrum λi is expressed by the following formula,²⁵ where N denotes the total number of iterations used for the calculation and t represents the time index. The parameter τ denotes the evolution time (time delay) used to evaluate the divergence of neighboring trajectories. The term $e_{i}^{'} (t)$ represents the local stretching factor of the i-th orthogonal basis vector at time t, and $\log | e_{i}^{'} (t) |$ denotes the logarithmic divergence used to estimate the Lyapunov exponent.

\underset{n \to \infty}{λ_{i} = \lim} (1 / N τ) \sum_{t = 0}^{N - 1} \log | e_{i}^{'} (t) |

In this study, we used the maximum Lyapunov exponent λ1, which most accurately reflects the strength of chaos, in the analysis of the 4 Lyapunov exponents λ1 to λ4 corresponding to the number of embedded dimensions.

Fractal Dimension

The fractal dimension quantifies the complex geometric structure of an attractor. In this study, it was calculated using the correlation dimension method (Grassberger–Procaccia method: G-P).²⁶ The G-P method embeds the time series waveform into m-dimensional space following Takens’ embedding theorem.²⁸ In the embedded phase space, the hypersphere radius is defined as r, with an arbitrary point P_i at the center of the attractor. Let the total number of data points comprising the attractor be N. Gradually increasing the radius r, the Heaviside function H(||P_i − P_j||) is defined as the average of these values. The correlation integral C_r is expressed by the following formula:

C_{r} = \frac{1}{N^{2}} \sum_{i = l}^{N} \sum_{j = l}^{N} H (r - | p_{i} - p_{j} |)

The correlation integral C(r) is plotted against the logarithm of the radius r, forming what is known as a correlation diagram. In a correlation diagram, logC(r) is typically plotted against log(r), revealing a linear section referred to as the scaling region. Figure 3a provides an overview of a correlation diagram. A convergence diagram is then generated using the slope of the scaling region from the correlation diagram. In the convergence diagram, the horizontal axis represents dimension D, and the vertical axis represents the slope v; an example is shown in Figure 3b. Chaos and periodic fluctuations exhibit a characteristic where the slope v saturates when embedded in a sufficiently large dimension. In this study, the value of this saturated v was determined to calculate the correlation dimension.

Figure 3.

Overview of calculating the fractal dimension. (a) Overview of the correlation diagram showing the relationship between log r and log C(r). The dashed box indicates the scaling domain, where the linear slope represents the correlation dimension for each embedding dimension. (b) Overview of the convergence diagram showing the slope ν as a function of embedding dimension D. The region of interest (D = 5 − 9) indicates the saturation region where ν stabilizes to define the final correlation dimension.

Statistical Analysis

In this study, we performed statistical analyses of pulse wave amplitude values, Lyapunov exponents (λ1), and fractal dimensions in the Mi and nMi groups. The normality of pulse wave amplitude, λ1, and fractal dimensions was assessed using the Shapiro–Wilk test and Q–Q plots. Subsequently, the mean and standard deviation of pulse wave amplitude, λ1, and fractal dimensions were calculated for each group to enable quantitative evaluation. To further explore the effects of mindfulness, the rate of change for each variable during daily measurements was calculated relative to the fixed-point measurements. For group comparisons, t-tests were used to evaluate pulse wave amplitude, λ1, and fractal dimensions on each experimental day between the Mi and nMi groups. Within-group changes over time were analyzed using one-way analysis of variance (ANOVA). Multiple comparisons were conducted using the Games–Howell method for pulse wave amplitude, fractal dimensions, and λ1. To further control the family-wise Type I error rate across multiple statistical tests, the Bonferroni correction was applied.

The significance level for statistical analysis was set at 5%, and IBM SPSS^® Statistics version 26 was used for all statistical procedures.

To adjust for potential bias due to baseline differences at the start of measurement (at the university), an analysis of covariance (ANCOVA) was performed. The dependent variable was pulse wave amplitude on the ninth day of home measurement, the independent variables were each group, and baseline amplitude values were included as covariates. Statistical analysis was performed using R (version 4.5.2).

We used the CONSORT reporting guideline (1) to draft this manuscript, and the CONSORT reporting checklist (2) when editing, included as a Supplemental File.

Results

Sample Characteristics and Research Question Results

Frequency Analysis of Finger Plethysmogram

Table 1 presents the results of one-way ANOVA and Bonferroni correction for pulse wave amplitude values derived from the finger volume pulse wave in the Mi and nMi groups. Even with the Bonferroni correction significance level of α/3, significant differences were observed between the conditions, consistent with the ANOVA results. Covariance analysis (ANCOVA) was performed to adjust for baseline differences. The results revealed a significant main effect between groups (F(1, 6337) = 5.887, P = .015). The regression coefficient for the mindfulness group was 32.750 (SE = 13.498, t = 2.426, P = .015), indicating that the mindfulness intervention significantly enhanced pulse wave amplitude compared to the non-intervention group, even after adjusting for the initial baseline at the “University” measurement point.

Table 1.

Results of Multiple Comparisons of Pulse Wave Amplitude Values.

		Mi: P-value	Mi: Bonferroni-adjusted P-value	Mi: Ratio*	nMi: P-value	nMi: Bonferroni-adjusted P-value	N = 18nMi: Ratio
		Mi			nMi
Parameter		P-value	Bonferroni	*	P-value	Bonferroni	*
Fixed	Day 1	<.001	<0.001	1.12	.05	0.017	0.97
Fixed	Day 9	<.001	<0.001	1.27	<.001	<0.001	0.80
Day 1	Day 9	<.001	<0.001		<.001	<0.001

Abbreviations: Fixed, fixed-point measurement; Day 1, daily measurement Day 1; Day 9, daily measurement Day 9: Mi, mindfulness group; nMi, nonmindfulness group.

Exact P-values are rounded to 3 decimal places. Values less than .001 are reported as P < .001.

The ratio of daily measurement values to the fixed-point measurement values (calculated as Daily measurement/Fixed-point measurement).

Figure 4 illustrates the results of the inter-group t-test comparison and the one-way ANOVA conducted within each group for pulse wave amplitude values in the Mi and nMi groups. The findings reveal that, across experimental periods, the pulse wave amplitude values in the Mi group were significantly higher than the fixed-point measurement values on both the first and ninth days (P < .01). Additionally, the pulse wave amplitude values in the Mi group on the ninth day were significantly higher than those on the first day (P < .01). Conversely, in the nMi group, the pulse wave amplitude value recorded on the ninth day of daily measurements was significantly lower than the fixed-point measurement (P < .01). Furthermore, the pulse wave amplitude values on the ninth day were significantly lower than those on the first day of daily measurements (P < .01). A t-test comparing the groups revealed that the pulse wave amplitude values in the Mi group were significantly lower than those in the nMi group at the fixed-point measurement and on the first day (P < .01). Over time, the rate of change in the Mi group showed an increasing trend, whereas the rate of change in the nMi group demonstrated a decreasing trend.

Figure 4.

Results of multiple comparisons of pulse wave amplitude values. Mean pulse wave amplitude [mV] in the mindfulness implementation group (Mi) and control group (nMi) at the fixed-point measurement and daily measurements on Day 1 and Day 9. Error bars represent standard deviations (N = 18 per group). **P < .01 by one-way ANOVA with Games–Howell post-hoc tests and Bonferroni correction.

Chaos Analysis Using Lyapunov Exponents

Figure 5 presents an example of the Lyapunov index results calculated using the Sano–Sawada method. In Figure 5, the Lyapunov index is plotted on the vertical axis, while the number of calculations and the 4 Lyapunov indices corresponding to the number of embedded dimensions, λ1 to λ4, are plotted on the horizontal axis. The maximum Lyapunov index λ1, was calculated as 1.75. Under all measurement conditions, the λ1 values for both the Mi and nMi groups were positive.

Figure 5.

An example of the Lyapunov exponent. The plot shows the dynamic evolution of the Lyapunov spectrum (λ1, λ2, λ3, and λ4) as a function of the number of computation steps. While the final maximum Lyapunov exponent (λ1) is determined to be 1.75 (the average value at the end of the time series), the vertical axis shows the range of transient fluctuations during the initial stages of the calculation, with an observed maximum of 22.18 and a minimum of −55.05.

Chaos Analysis Using Fractal Dimension

Figure 6a presents an example of the fractal dimension of the Mi group calculated using the G-P method. For each embedding dimension, a correlation diagram is shown, plotting the relationship between the distance r on the horizontal axis and the logarithm of the correlation integral C(r) on the vertical axis. The scaling region is the interval where linearity is maintained over a wide range of r. Figure 6b displays an example of a convergence diagram, plotting the dimension D and slope v derived from the scaling region in Figure 6a. From this convergence diagram, it is evident that the dimension D converges at an embedding dimension of 6, and the fractal dimension is calculated as 3.1.

Figure 6.

Example of fractal dimension calculation results. (a) Correlation diagram showing the relationship between log r and log C(r) (correlation integral). The dashed box indicates the scaling domain, where the linear slope represents the correlation dimension. (b) Convergence diagram showing the slope ν as a function of embedding dimension D. The region of interest (D = 5 − 9) indicates the saturation region where ν stabilizes and defines the final correlation dimension.

Table 2 presents the changes over time in the mean and standard deviation of λ1 and fractal dimension values calculated from the finger volume pulse wave in the Mi and nMi groups, as well as the rate of change in daily measurement values relative to the fractal dimension values measured at fixed points. In the Mi group, λ 1 was highest during the fixed-point measurement and exhibited a decreasing trend over time during the daily measurements. Similarly, In the nMi group, λ1 was also high during the fixed-point measurement. The rate of change in λ1 during the daily measurements in the Mi group tended to be higher than in the nMi group.

Table 2.

Chaos Analysis Results.

		Mi Mean	Mi SD	Mi Ratio*	nMi Mean	nMi SD	nMi Ratio*	N = 18P-value
		Mi			nMi
Parameter		Mean	(S.D.)	*	Mean	(S.D.)	*	P-value
Fixed	λ1	5.21	(2.35)		4.46	(2.25)		.498
Daily	λ1 (Day 1)	3.38	(2.41)	0.65	3.93	(1.51)	0.88	.566
Daily	λ1 (Day 9)	3.88	(1.92)	0.74	3.89	(2.44)	0.87	.994
Fixed	Fractal	2.71	(0.25)		2.66	(0.19)		.678
Daily	Fractal (Day 1)	2.8	(0.27)	1.03	2.65	(0.25)	0.99	.244
Daily	Fractal (Day 9)	2.82	(0.27)	1.01	2.67	(0.41)	1	.365

Abbreviations: Daily, daily measurement; Fixed, fixed-point measurement; Fractal, fractal dimension; Mi, mindfulness group; nMi, nonmindfulness group; S.D., standard deviation.

P-value: Calculated using the independent samples t-test.

(Rate of change): The rate of change in daily measurement values was calculated relative to fixed-point measurement values (rate of change = daily measurements/fixed-point measurements).

A one-way ANOVA conducted on the changes over time in both groups revealed no significant differences between the groups. Furthermore, a t-test comparing the 2 groups at each measurement point showed no significant differences at any time.

Regarding fractal dimension, the Mi group showed an average of approximately 2.8, while the nMi group showed an average of approximately 2.6, confirming that both groups exhibit fractal structures.Through fixed-point measurements and daily measurements, a slight tendency toward higher fractal dimensions was observed in the Mi group compared to the nMi group. However, statistical analysis revealed no significant differences between the 2 groups in terms of fractal dimensions or their temporal changes. Additionally, the temporal change rates of fractal dimensions in both groups were stable.

Discussion

Principal Findings

This study demonstrated that short-term mindfulness-based breathing exercises over a 9-day period showed favorable changes in autonomic nervous system function, particularly evidenced by increased pulse wave amplitude. In particular, the finger plethysmogram characteristics suggest that there are different physiological responses between the Mi group and the nMi group.

Frequency Analysis of Finger Plethysmogram

The decision to implement a 9-day intervention is supported by evidence that brief, daily mindfulness and respiratory exercises can induce significant physiological benefits. Daily 5-minute structured breathing has been demonstrated to reduce physiological arousal, with effects emerging within the first week of practice.²⁹ A single-session mindfulness-based intervention has been reported to produce meaningful reductions in perceived stress, anxiety, and depressive symptoms within a short follow-up period.³⁰

For university students, brief and easily accessible interventions delivered via mobile applications are considered suitable for addressing their stress-management needs.⁷ Our 9-day observation window appeared to capture a transition from initial cognitive engagement to more stable autonomic regulation.

In the Mi group, the pulse wave amplitude values of daily measurements on days 1 and 9 were significantly higher than the fixed-point measurements, with values on day 9 being even higher than those on day 1. The pulse wave amplitude values showed a gradual increase throughout the intervention period. Previous research has demonstrated that sympathetic nervous system activation induces peripheral vasoconstriction, which typically reduces pulse wave amplitude, such as during stress or cold stimulation.^19,31,32 Thus, the increase in pulse wave amplitude values observed in this study suggests peripheral blood vessel dilation, likely reflecting suppressed sympathetic nervous system activity.

Mindfulness-based breathing exercises have been reported to regulate autonomic nervous system balance by activating the vagus nerve and suppressing sympathetic nervous system activity.³³ In this study, a short-term practice of mindfulness-based breathing exercises over 9 days showed a positive effect on autonomic nervous system function. These findings suggest that even relatively brief interventions may be sufficient to induce measurable autonomic modulation, supporting the feasibility of incorporating mindfulness-based breathing exercises into daily self-care routines.

In the nMi group, the pulse wave amplitude during daily measurements was significantly lower than the fixed-point measurement, with a further decrease was observed on the ninth day. Continuous exposure to monotonous visual stimuli or prolonged academic mental load has been reported to increase mental fatigue and psychological stress.^1,34 In this study, the mental load caused by monotonous visual stimuli in the nMi group likely contributed to increased sympathetic nervous system activity.

At the initial “University” measurement, the nMi group showed higher pulse wave amplitude than the Mi group. This difference may reflect task characteristics: the cross-fixation task was a passive, low-arousal visual activity that could induce quiet restfulness, whereas participants in the Mi group were introduced to a new mindfulness technique that may have increased cognitive load or performance anxiety, leading to slight sympathetic activation and lower amplitude during the first session.

To address potential bias from these baseline differences, ANCOVA was conducted using baseline amplitude as a covariate. The results showed a significant main effect of group (F(1, 6337) = 5.887, P = .015), indicating that after adjustment the Mi group exhibited significantly greater pulse wave amplitude on Day 9 than the nMi group. Although final amplitudes appeared similar, the physiological trajectories differed: the Mi group showed a progressive increase in amplitude, suggesting cumulative autonomic stabilization, whereas the nMi group showed a declining trend consistent with task-related fatigue.

Mental stress has been shown to influence peripheral vasodilation and blood flow,³⁵ with recent studies emphasizing the impact of acute stress on arterial stiffness and wave reflections.^19,22

Increased pulse wave amplitude is generally interpreted as reflecting improved peripheral circulation mediated by autonomic nervous system regulation. The adjusted mean difference (32.750) observed in the present analysis supports the interpretation that mindful breathing may facilitate autonomic regulation over time.

However, because ANCOVA statistically controls rather than eliminates baseline differences, the present findings should be interpreted as evidence of an intervention-associated effect after adjustment, rather than definitive proof of physiological causality. Further randomized and longitudinal studies are warranted to confirm these effects.

The initially low pulse wave amplitude values in the Mi group likely reflect changes in autonomic nervous system activity during the early stages of mindful breathing. Over 9 days of continuous training, sympathetic nervous system activity was progressively suppressed, resulting in peripheral blood vessel dilation, as evidenced by increased pulse wave amplitude. Thus, the time series pattern of pulse wave amplitude suggests that the effects of mindful breathing on autonomic nervous system regulation gradually manifest over time.

Chaos Analysis Using Lyapunov Exponents

Under all measurement conditions, the Lyapunov Exponent (λ1) values for both the Mi and nMi groups were positive, indicating that their time series data exhibited chaotic nonlinear dynamics.

In the Mi group, λ1 was highest during fixed-point measurements and demonstrated a decreasing trend during daily measurements. Similarly, in the nMi group, λ1 was also high during fixed-point measurements. The rate of change in λ1 during daily measurements was higher in the Mi group compared to the nMi group, though no significant differences were observed between the groups. The Lyapunov exponent is a measure of sensitivity to initial conditions, where higher values reflect increased chaos.^18,36 Practicing mindful breathing has been shown to influence the balance between the sympathetic and parasympathetic nervous systems,¹¹ which govern autonomic nervous functions.⁸ Recent nonlinear analyses of PPG signals suggest that human physiological signals possess complex chaotic characteristics that change in response to mental states.¹⁸ The findings of this study suggest that mindful breathing may increase the complexity of pulse wave variability by activating autonomic nervous system activity. However, with continued practice, this activity stabilizes, leading to an optimized balance between the sympathetic and parasympathetic nerves, thereby improving predictability and exerting a relaxing effect.

An increase in chaos, indicated by higher λ1, represents greater instability in pulse wave variability.¹⁸ Thus, in the early stages of mindful breathing, autonomic nervous system activity becomes heightened, increasing the complexity of heart rate variability. Over time, continued practice appears to optimize autonomic balance, stabilizing heart rate variability and promoting relaxation.

Chaos Analysis Using Fractal Dimension

The fractal dimension quantifies the complexity of self-similar patterns and the extent to which they fill space.^26,37,38 While the concept was established in early chaos theory,^26,38 modern applications in EEG and PPG analysis have further validated its use in identifying physiological shifts.³⁷ In this study, the Mi group had a fractal dimension of approximately 2.8, while the nMi group had a value of around 2.6, confirming the presence of fractal structures in both groups. Statistical analysis revealed no significant difference in fractal dimension or its change over time between the 2 groups. Moreover, the rate of change in fractal dimension over time remained stable in both groups.

Statistical analysis showed no significant differences in fractal dimension between groups or over time. These findings suggest that the short-term mindfulness intervention did not produce detectable changes in the nonlinear dynamics of the autonomic nervous system, and that fractal dimension may be less sensitive to subtle short-term physiological changes than frequency-domain indices.

Strengths and Limitations

One limitation of this study was the control condition, as the cross-gaze task was static and passive. Future studies should include a broader range of control tasks, such as paced-breathing without mindfulness instructions or relaxation-based conditions, to clarify whether observed physiological changes reflect respiratory regulation or specific mindfulness effects.^10,39 Moreover, increases in fractal dimension do not necessarily indicate greater autonomic complexity,³⁷ and because heart rate variability (HRV) was not directly measured, future research would be beneficial.

Implications for Practice

The results of this study hold potential for improving mental health across a wide range of age groups in modern society, beyond just university students who face daily stressors. The scalability and adaptability of such applications could make them valuable tools for addressing the mental health challenges of diverse populations.

In this study, the efficacy of tablet applications in supporting the mental health of university students was demonstrated through pulse wave frequency analysis and chaos analysis. These findings could contribute to the development of more effective psychological support programs using mobile applications, such as smartphones. By leveraging the unique characteristics of mobile applications—being accessible anytime and anywhere—they can serve as a new option for mental health care tailored to university students.

Conclusions

This study examined the effects of mindful breathing on autonomic nervous system activity using frequency-domain and chaos analyses of fingertip plethysmogram signals. Pulse wave amplitude showed significant changes over the 9-day intervention in the mindfulness group, suggesting alterations in peripheral vascular responses under the present conditions. In contrast, chaos indices, including the Lyapunov exponent and fractal dimension, showed no significant differences between groups. Although descriptive tendencies were observed, these results should be interpreted cautiously. Overall, short-term mindful breathing may influence certain peripheral physiological parameters; however, further studies are required to clarify its effects on nonlinear autonomic regulation.

Supplemental Material

sj-docx-1-bec-10.1177_11795972261452228 – Supplemental material for Evaluation of Mindful Breathing Method Using Tablet Devices Through Chaos Analysis and Frequency Analysis: A Pilot Randomized Controlled Trial

Supplemental material, sj-docx-1-bec-10.1177_11795972261452228 for Evaluation of Mindful Breathing Method Using Tablet Devices Through Chaos Analysis and Frequency Analysis: A Pilot Randomized Controlled Trial by Eiichi Togo in Biomedical Engineering and Computational Biology

Footnotes

ORCID iD

Eiichi Togo

Author Contributions

Eiichi Togo: Conceptualization; Methodology; Investigation; Data curation; Software; Formal analysis; Visualization; Writing – original draft; Writing – review & editing; Project administration.

Funding

The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by JSPS Grant-in-Aid for Scientific Research 21K10566.

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Supplemental Material

Supplemental material for this article is available online.

References

Córdova Olivera

Gasser Gordillo

Naranjo Mejía

La Fuente Taborga

Grajeda Chacón

Sanjinés Unzueta

. Academic stress as a predictor of mental health in university students. Cogent Educ. 2023;10(2):2232686.

Murayama

Okubo

Tabuchi

Increase in social isolation during the COVID-19 pandemic and its association with mental health: findings from the JACSIS 2020 study. Int J Environ Res Public Health. 2021;18(16):8238.

Asif

Mudassar

Shahzad

Raouf

Pervaiz

Frequency of depression, anxiety and stress among university students. Pak J Med Sci. 2020;36(5):971-976.

Szuhany

Simon

NM.

Anxiety disorders: a review. JAMA. 2022;328(24):2431-2445.

Subramaniam

Abdin

Vaingankar

, et al. Tracking the mental health of a nation: prevalence and correlates of mental disorders in the second Singapore mental health study. Epidemiol Psychiatr Sci. 2019;29:e29.

Macaskill

The mental health of university students in the United Kingdom. Br J Guid Couns. 2013;41(4):426-441.

Lee

Kim

Development and evaluation: a behavioral activation mobile application for self-management of stress for college students. Healthcare. 2022;10(10):1880.

Kabat-Zinn

Full Catastrophe Living: Using the Wisdom of Your Body and Mind to Face Stress, Pain, and Illness. Delta Books; 1990.

Bishop

Lau

Shapiro

, et al. Mindfulness: a proposed operational definition. Clin Psychol Sci Pract. 2004;11(3): 230-241.

10.

Khoury

Sharma

Rush

Fournier

Mindfulness-based stress reduction for healthy individuals: a meta-analysis. J Psychosom Res. 2015;78(6):519-528.

11.

Tang

Hölzel

Posner

MI.

The neuroscience of mindfulness meditation. Nat Rev Neurosci. 2015;16:213-225.

12.

Ameli

Sinaii

West

, et al. Effect of a brief mindfulness-based program on stress in health care professionals at a US biomedical research hospital: a randomized clinical trial. JAMA Netw Open. 2020;3(8):e2013424.

13.

Oberoi

Yang

Woodgate

, et al. Association of mindfulness-based interventions with anxiety severity in adults with cancer: a systematic review and meta-analysis. JAMA Netw Open. 2020;3(8):e2012598.

14.

Hooi

Chen

Tan

de Vries

Wong

HK.

Effects of mindfulness breathing meditation on stress and cognitive functions: a heart rate variability and eye-tracking study. Sci Rep. 2025;15(1):37185.

15.

Küchler

Schultchen

Pollatos

Moshagen

Ebert

Baumeister

StudiCare mindfulness-study protocol of a randomized controlled trial evaluating an internet- and mobile-based intervention for college students with no and “on demand” guidance. Ment Health. 2020;21(1):975.

16.

Allen

Photoplethysmography and its application in clinical physiological measurement. Physiol Meas. 2007;28(3):R1-39.

17.

Hashimoto

Chonan

Aoki

, et al. Pulse wave velocity and the second derivative of the finger photoplethysmogram in treated hypertensive patients: their relationship and associating factors. J Hypertens. 2002;20(12):2415-2422.

18.

Sviridova

Sakai

Human photoplethysmogram: new insight into chaotic characteristics. Chaos Solitons Fractals. 2015;77:53-63.

19.

Celka

Charlton

Farukh

Chowienczyk

Alastruey

Influence of mental stress on the pulse wave features of photoplethysmograms. Healthc Technol Lett. 2020;7(1):7-12.

20.

Selvaraj

Jaryal

Santhosh

Deepak

Anand

Assessment of heart rate variability derived from finger-tip photoplethysmography as compared to electrocardiography. J Med Eng Technol. 2008;32(6):479-484.

21.

Djeldjli

Bousefsaf

Maaoui

Bereksi-Reguig

Pruski

Remote estimation of pulse wave features related to arterial stiffness and blood pressure using a camera. Biomed Signal Process Control. 2021;64:102242.

22.

Vlachopoulos

Kosmopoulou

Alexopoulos

Ioakeimidis

Siasos

Stefanadis

Acute mental stress has a prolonged unfavorable effect on arterial stiffness and wave reflections. Psychosom Med. 2006;68:231-237.

23.

Bharati

Gidveer

Waveform analysis of pulse wave detected in the fingertip with PPG. Int J Adv Eng Technol. 2012;3(1):92.

24.

Choi

Shin

Photoplethysmography sampling frequency: pilot assessment of how low can we go to analyze pulse rate variability with reliability?

Physiol Meas. 2017;38(3):586-600.

25.

Sano

Sawada

Measurement of the Lyapunov spectrum from a chaotic time series. Phys Rev Lett. 1985;55(10):1082-1085.

26.

Grassberger

Procaccia

Characterization of strange attractors. Phys Rev Lett. 1983;50(5):346-349.

27.

Wolf

Quantifying chaos with Lyapunov exponents. In: Holden

, ed. Chaos. Princeton University Press; 1986:273-290.

28.

Takens

Detecting strange attractors in turbulence. In: Rand

Young

eds. Dynamical Systems and Turbulence, Warwick 1980. Lecture Notes in Mathematics. Vol. 898. Springer; 1981:366-381.

29.

Balban

Neri

Kogon

, et al. Brief structured respiration practices enhance mood and reduce physiological arousal. Cell Rep Med. 2023;4(1):100895.

30.

Rubin

Fischer

Telch

MJ.

Efficacy of a single session mindfulness based intervention: a randomized clinical trial. PLoS One. 2024;19(3):e0299300.

31.

Elgendi

On the analysis of fingertip photoplethysmogram signals. Curr Cardiol Rev. 2012;8(1):14-25.

32.

Korhonen

Yli-Hankala

Photoplethysmography and nociception. Acta Anaesthesiol Scand. 2009;53:975-985.

33.

Nijjar

Puppala

Dickinson

, et al. Modulation of the autonomic nervous system assessed through heart rate variability by a mindfulness based stress reduction program. Int J Cardiol. 2014;177(2):557-559.

34.

Thackray

RI.

The stress of boredom and monotony: a consideration of the evidence. Psychosom Med. 1981;43(2):165-176.

35.

Freyschuss

Hjemdahl

Juhlin-Dannfelt

Linde

Cardiovascular and sympathoadrenal responses to mental stress: influence of beta-blockade. Am J Physiol Circ Physiol. 1988;255(6):H1443-H1451.

36.

Wolf

Swift

Swinney

Vastano

JA.

Determining Lyapunov exponents from a time series. Phys D. 1985;16(3): 285-317.

37.

Mohd Radzi

Asirvadam

Yusoff

MZ.

Fractal dimension and power spectrum of electroencephalography signals of sleep inertia state. IEEE Access. 2019;7:185879-185892.

38.

JH.

The fractal dimension of EEG as a physical measure of conscious human brain activities. Bull Math Biol. 1988;50:559-565.

39.

Ponsiglione

Russo

Petrellese

, et al. Comparison of techniques for respiratory rate extraction from electrocardiogram and photoplethysmogram. Sensors. 2025;25(16): 5136.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.12 MB