The experiment took place in April 2024 at Panda Valley National Forest Park, Foping County, Shaanxi (33.658°N, 107.807°E, elevation 1,080–2094 meters; see Figure 1). The park’s warm, humid climate and high forest coverage, with 25,000 negative ions/cm³, earned it the name “Forest Oxygen Bar.”

We divided participants into four groups: Group A performed combined dynamic-static forest bathing (60 min total: 30 min dynamic, 30 min static, with a 15 min break), Group B did dynamic forest bathing alone (30 min), Group C did static forest bathing alone (30 min), and Group D was a control group (30 min). The 60 min duration for the combined intervention was chosen to reflect common forest therapy practices, which typically last 45–60 min or longer and often integrate both dynamic and static phases (Park et al., 2010; Vermeesch et al., 2024; Zhang et al., 2023).

The intervention groups were tested in Panda Valley National Forest Park, Foping County, Shaanxi, due to its uniform open forest structure, minimizing coverage and spatial structure effects. The control group experiment was conducted in the Northwest A&F University College of Landscape Architecture and Arts laboratory. An environmental control system adjusted the lab’s temperature and humidity to match Panda Valley, ensuring experimental consistency and validity. This approach maintained comparable climatic conditions between the control and forest groups, minimizing confounding effects of temperature and humidity, while providing a non-natural baseline against which the effects of forest bathing could be highlighted.

To ensure scientific rigor, Groups A, B, and C used the same forest area with environmental controls. Experiments were scheduled at 9 a.m. to minimize light and temperature changes. Walking routes and sitting locations were consistent. Ecologists assessed plant species, air quality, noise, terrain safety, and wildlife absence. In the static group, participants engaged in guided forest meditation while wearing standardized headphones with a professionally developed meditation script. The script was carefully evaluated and validated by landscape architecture experts to ensure consistency and effectiveness (validation report available in Forest Meditation Evaluation Report and Text Version). This setup ensured that all participants received identical auditory input, minimizing variability and external environmental sound exposure.

Group A used Route 1, combining a walk from Panda Forest Hotel to Golden Monkey Grand Canyon and static activities at Jingxin Pavilion. Jingxin Pavilion, near the hotel, served as the starting point for all routes. Group B used only the walking route of Route 1, and Group C used only Jingxin Pavilion.

Group D’s experiment was conducted in the 12 m² landscape lab at Northwest A&F University, isolated by PVC whiteboards to eliminate visual distractions. The experiment occurred from 9 to 11 a.m. in April, maintaining an indoor temperature of 21.5 °C and 60–80% humidity, with no external noise (Figure 1) (Wen et al., 2023; Mao et al., 2017). As Group D served as the blank control group to provide a baseline for comparison, no elements simulating the forest environment were introduced.

We recruited middle-aged tourist volunteers through online and offline channels in Foping County and Yangling District, Shaanxi Province, from early March to the end of April 2024. The trial was conducted in April from 9:00 to 11:00 a.m., with all participants completing the assigned tasks as planned. A total of 72 participants were enrolled (mean age 62.5 ± 7.22 years; range 60–75), with a gender ratio of 1:1. To ensure homogeneity and reliability, individuals who were smokers, drinkers, menstruating, experiencing physical discomfort, or with olfactory/hearing impairments or severe health conditions (e.g., cancer, heart disease, mental illness) were excluded (Yau and Loke, 2020).

All participants provided informed consent and were free to withdraw at any time, though no withdrawals or exclusions occurred during the study. Each volunteer was assigned a unique ID number and received a small gift after participation. The study protocol adhered to the ethical standards of the Psychology Center at Northwest A&F University and the Declaration of Helsinki. It was approved by the Psychology Development and Education Center of Northwest A&F University (Approval No. 202403121185) and registered under NWAFU202407643.

To minimize the Hawthorne effect, participants were not informed of the study objectives. Participants were randomly assigned to one of four groups (A–D) through a computer-generated sequence by an independent researcher, with allocation concealed in sequentially numbered opaque envelopes; both participants and investigators remained blinded until the intervention began. Each group consisted of 18 participants (9 males and 9 females), totaling 72. All participants completed their assigned activities and were included in the primary outcome analysis (Figure 2). This sample size is methodologically robust and consistent with previous forest bathing studies, which typically enrolled 12–20 participants per group. However, as each gender subgroup contained only nine participants, the statistical power for subgroup analyses was limited.

The Perceived Restorativeness Scale (PRS) assesses psychological restoration in various environments. Participants rate descriptive items on a five-point scale from 1 (strongly disagree) to 5 (strongly agree). Higher scores indicate higher perceived restoration. PRS is based on Attention Restoration Theory (ART), which suggests that exposure to natural environments helps reduce cognitive fatigue and enhances well-being (Liu et al., 2024). Since forest bathing is designed to promote relaxation and mental recovery, PRS is an appropriate tool for measuring its psychological benefits (Takayama et al., 2022). This study uses PRS to measure perceived restoration under different activities (Wen et al., 2023; Mao et al., 2017).

The Brief Profile of Mood States (BPOMS) assesses mood, covering dimensions like pleasantness, tension, and fatigue. Participants rate each item from 1 to 5. This study uses BPOMS to evaluate mood changes under different activities. The questionnaire is in Chinese, with explanations provided. Data are used for research and education, ensuring privacy (Maji, 2018; Mao et al., 2017). BPOMS was chosen for its conciseness and efficiency, especially for our older adult participants, who may find lengthy questionnaires challenging. This simplified version has been widely validated in both recent and past studies, demonstrating high reliability and strong correlation with the full POMS, making it a justified and effective tool for our research (Cella et al., 1987; Shao et al., 2024; Shacham, 1983; Yan et al., 2024; Yoo et al., 2005; Zhao et al., 2024).

Electroencephalography (EEG) measures brain responses to stimuli and is a key physiological indicator. We used the portable SAGA 64 + EEG from TMSI, adhering to the 10–20 system for electrode placement. This device provides high-quality data, filtering out noise and muscle interference, and is suitable for field and lab research. We focused on frontal electrodes (Fp1, Fp2, F3, F4, F7, F8) because frontal EEG activity is strongly associated with emotional and cognitive states. Specifically, Left frontal electrodes (Fp1, F7, F3) correlate with positive emotions (Zhao et al., 2018), while right frontal electrodes (Fp2, F8, F4) correlate with negative emotions). We used TMSI’s algorithms to analyze EEG data from these electrodes, obtaining alpha wave data per second in the frontal area. Increased alpha wave activity and decreased beta wave activity have been linked to reduced stress and improved mental clarity (Grassini et al., 2022).

Galvanic Skin Response (GSR) measures skin conductivity, reflecting emotional responses, stress, and autonomic nervous system activity. GSR is particularly relevant for assessing emotional arousal in response to environmental stimuli, such as the calming effects of natural settings. Increased skin conductance indicates emotional tension and sympathetic activation, while decreased conductance signals calmness (Kim et al., 2019). GSR has been widely applied in stress and environment-related studies.

Heart Rate (HR) indicates cardiac activity and cardiovascular function. Variability in HR reflects health, emotional state, and response to stimuli. Increased HR can result from excitement, activity, or stress, while decreased HR occurs with relaxation. HR is particularly useful in forest therapy research as it reflects both acute physiological relaxation and long-term cardiovascular benefits. Previous studies have confirmed its sensitivity to autonomic regulation and stress reduction.

Blood Pressure (BP) is the force of blood on vessel walls, comprising Systolic (SBP) and Diastolic Pressure (DBP). BP is affected by cardiovascular status, vessel elasticity, blood viscosity, and autonomic regulation. Stress, tension, or activity raises BP, while relaxation lowers it. Since forest bathing has been linked to reductions in stress-related hypertension and improved cardiovascular health, BP serves as a critical measure of the physiological benefits of exposure to natural environments (Yau and Loke, 2020; Ideno et al., 2017; Peterfalvi et al., 2021).

This study designed a combined dynamic-static forest bathing group (Group A) to explore its effects on physiological and psychological health and compare different methods. Volunteers were randomly assigned and transported by shuttle bus at 8 a.m., given 30 min to rest and acclimate.

Participants began with dynamic forest hiking. Electronic devices and alcoholic or caffeinated beverages were prohibited. Hikes started at 5 min intervals, followed by a 15 min rest. Next, they moved to the static phase, receiving cushions, eye masks, and headphones for 30 min of forest meditation. Physiological and psychological indicators were measured pre- and post-experiment. Participants received small gifts afterward (Figure 4).

The dynamic forest bathing group (Dynamic Group) was formed by random assignment. Volunteers were given unique IDs, and their health, drinking, and smoking habits were confirmed. The procedure was described objectively to prevent psychological suggestion, such as creating expectations about relaxation, mood improvement, or health benefits.

The schedule was as follows: Volunteers arrived at 8 a.m., rested for 30 min, and had baseline data measured. The experiment started at 9 a.m. The dynamic group hiked a 3-km route at 5 min intervals, with a target duration of 30 min (acceptable range: 28–33 min) along a trail ranging from 1,400 to 1,570 meters in altitude. To help maintain a consistent walking pace (~6 km/h), research staff provided a brief pacing demonstration before departure, and visual time markers were placed along the route for self-regulation. However, volunteers were instructed to walk at a comfortable, natural speed, and only those who completed the walk within the 28–33 min range were included in the final analysis. This hiking route was located within Foping Panda Valley National Forest Park, Shaanxi Province, where the overall elevation spans from 1,080 to 2094 meters. Electronic devices, alcohol, and smoking were prohibited. After the hike, volunteers rested for 15 min before their physiological and psychological indicators were remeasured. Gifts were distributed only after all measurements had been completed to ensure they did not influence volunteers’ responses.

The static forest bathing group (Static Group) was formed by randomly selecting volunteers, each with a unique ID. The activity began at 8 a.m., with volunteers shuttled to the site. After 30 min of acclimation, baseline measurements were taken for EEG, heart rate, blood pressure, and psychological stress.

Participants were provided with cushions, disposable eye masks, and JBL TUNE 500BT headphones. The forest meditation instructions, validated by experts in landscape architecture, were played at a baseline volume of 50 decibels. Before the experiment, participants were asked to confirm whether they could clearly hear the audio and were allowed to adjust the volume slightly within a comfortable range (typically 50–60 dB) to suit individual preferences (Moore, 2012). The 30 min session began at 9 a.m. Mobile phones and removal of eye masks or headphones were prohibited. Participants could signal to end the experiment if uncomfortable. Post-meditation, physiological and psychological indicators were measured to assess the static forest’s health impacts. Participants received small gifts. This study explored the benefits of static forest environments on health.

The blank control group (Group D) provided a baseline for assessing environmental impacts on participants’ psychological and physiological responses. The experiment was conducted in the College of Landscape Architecture’s laboratory at Northwest A&F University, with standardized conditions to minimize interference. White PVC boards divided the lab into individual units to limit participant interaction.

The control group session lasted 30 min. Participants were prohibited from talking, smoking, or using devices. Heart rate, blood pressure, mood, and perceived restorativeness were measured pre- and post-experiment to evaluate environmental impact and establish a baseline. Gifts were given at the end to encourage participation.

After the three forest bathing activities, EEG effect sizes showed Group A (combined dynamic-static) showed statistically significant but moderate improvements in positive emotions (Fp1: d = 0.645, 95% CI [0.088, 0.959]; F7: d = 0.529, 95% CI [0.038, 0.911]; F3: d = 0.488, 95% CI [−0.008, 0.971]) and modest reductions in negative emotions (F4: d = 0.509, 95% CI [0.010, 0.994]; Fp2: d = 0.351, 95% CI [−0.131, 0.822]). Group B (dynamic) and Group C (static) had minimal effects. Group D (control) showed small changes. In summary, Group A tended to show the largest statistical effects among groups, though effect sizes were moderate (Figure 5; Supplementary Table 2 for full results).

Post-forest bathing, GSR effect sizes showed Group A (combined dynamic-static) had the highest (d = 0.642, 95% CI [0.085, 0.954]), followed by Group C (static) with d = 0.477, 95% CI [−0.021, 0.974]. Group B (dynamic) had the lowest (d = 0.089, 95% CI [−0.409, 0.587]). Overall, Group A demonstrated statistically significant changes, although the clinical relevance may be limited due to variability. Yet the pattern suggests that combined activities may elicit stronger autonomic engagement.

Heart rate changes revealed Group B (dynamic) showed statistically significant but small improvements (d = 0.136, 95% CI [−0.330, 0.599]), followed by Group C (static) with d = 0.128, 95% CI [−0.338, 0.590]. Group A (combined dynamic-static) showed minimal change (d = 0.047, 95% CI [−0.416, 0.509]). Although effect sizes were small, the consistency across groups suggests reliable physiological engagement, supporting the robustness of these findings.

Forest bathing activities produced statistically significant changes in blood pressure, though effect sizes were small to moderate. Group A (combined dynamic-static) had effect sizes of 0.366 (95% CI [−0.117, 0.838]) for systolic and 0.200 (95%CI [−0.270, 0.664]) for diastolic pressure, with modest regulatory effects particularly on systolic pressure. Group B (dynamic) showed a low effect on systolic (d = 0.073, 95% CI [−0.391, 0.535]) and a medium effect on diastolic pressure (d = 0.292, 95% CI [−0.184, 0.760]). Group C (static) had effect sizes of 0.351 (95% CI [−0.130, 0.823]) for systolic and 0.211 (95% CI [−0.260, 0.675]) for diastolic pressure, with modest effects particularly on systolic pressure. Groups A and C showed moderate regulatory effects, while Group B had the most impact on diastolic pressure. Even modest blood pressure reductions are consistent with prior evidence and may contribute to meaningful long-term health benefits.

In the BPOMS mood questionnaire, Group A (combined dynamic-static) showed statistically significant reductions in negative emotions and TMD, with effect sizes of 1.669 (95% CI [0.935, 2.381]) for tension-anxiety and 1.646 (95% CI [1.006, 2.272]) for TMD, and moderate increases in vitality. Group B (dynamic) reduced negative emotions and increased vitality with moderate-to-large effect sizes (0.758, 95% CI [0.223, 1.276]) for tension-anxiety and 1.120,95% CI [0.516, 1.704] for TMD). Group C (static) showed the largest reductions in negative emotions, increased vitality (d = 2.227, 95%CI [−3.091, −1.343]), and reduced fatigue (d = 2.411, 95% CI [1.475, 3.328]), and large effect sizes for tension-anxiety (d = 2.012, 95% CI [1.188, 2.817]). Group D (control) showed only small changes (d = 0.517, 95% CI [0.018, 1.004] for tension-anxiety and d = 0.495, 95% CI [−0.002, 0.979] for TMD) (Supplementary Table 3).

The study revealed that the dynamic group (Group B, d = 2.590, 95% CI [−3.559, −1.602]) had the highest effect sizes and perceived recovery compared to the combined dynamic-static (Group A, d = 0.662, 95% CI [0.142, 1.167]), static (Group C, d = 0.948, 95% CI [−1.499, −0.379]), and control (Group D, d = 0.5, 95% CI [0.002, 0.984]) groups, indicating statistically significant but varied impacts of different activity types on psychological states. However, PRS captures subjective perceptions of environmental restorativeness, which is a distinct construct from emotional states and physiological indicators; therefore, its peak in the dynamic group does not contradict the broader improvements observed in the combined group. These results suggest that activity type meaningfully shapes perceived psychological restoration, even when effect sizes differ in magnitude.

EEG effect sizes revealed gender differences in emotional responses to forest activities. Combined dynamic-static activities (Group A) had statistically significant effects for both genders, with males showing larger positive emotion effects (Fp1: d = 1.319, 95% CI [0.388, 2.210]) and greater reductions in negative emotions (Fp2: d = 0.754, 95% CI [0.006, 1.457]). Single dynamic activities (Group B) showed the most impact on males (Fp1: d = 0.731, 95% CI [−0.028, 1.456]). Static activities (Group C) were more effective for females, increasing positive emotions (Fp1: d = 0.297, 95% CI [−0.381, 0.956]) and reducing negative emotions (Fp2: d = 0.375, 95% CI [−0.300, 1.028]). Group D (control) showed minimal changes (Figure 6; Supplementary Table 4). Given subgroup sizes (n = 9), these gender-specific findings should be considered exploratory.

GSR effect sizes showed males had higher responses across all groups, with static activities (Group C) showing the highest for males (d = 1.107, 95% CI [0.336, 1.828]) and combined dynamic-static activities (Group A) the highest for females (d = 0.495, 95% CI [−0.269, 1.199]). All activity groups outperformed the control group. However, subgroup results are based on small samples and require cautious interpretation.

Heart rate effect sizes suggested gender-specific patterns, although the magnitudes were small. Males tended to show greater recovery in Group C (static, d = 0.208, 95% CI [−0.949, 0.590]), while females showed relatively higher responses in Groups A (combined dynamic-static, d = 0.150, 95% CI [−0.613, 0.874]) and B (dynamic, d = 0.148, 95% CI [−0.615, 0.872]). While all confidence intervals crossed zero, the consistent directional trends indicate potential gender-related differences in autonomic regulation. These preliminary findings warrant further investigation with larger samples to clarify the robustness of gender effects in HR responses.

Forest bathing activities suggested potential gender differences in blood pressure responses. Females exhibited relatively higher effects, particularly in static activities (systolic: d = 0.658, 95% CI [−0.118, 1.375]; diastolic: d = 0.999, 95% CI [0.235, 1.758]). Males showed modest diastolic pressure changes in combined dynamic-static (d = 0.376, 95% CI [−1.128, 0.402]) and static activities (d = 0.127, 95% CI [−0.889, 0.645]). Although these subgroup findings are based on small samples and should be interpreted with caution, the consistent patterns suggest that females may benefit more from static activities, while males show relatively greater tendencies toward diastolic regulation.

Forest bathing activities showed statistically significant effects on psychological indicators, with gender differences observed. Combined dynamic-static and dynamic activities were associated with improvements in females’ psychological health, with effect sizes up to 2.874 (95% CI [1.86, 3.86]) for anger-hostility and 2.805 (95% CI [1.82, 3.79]) for fatigue. Static activities showed marked reductions in males’ negative emotions, with effect sizes of 3.766 (95% CI [2.36, 5.13]) for tension-anxiety and 3.689 (95% CI [2.30, 5.06]) for anger-hostility. Control group results supported these benefits (Figure 7; Supplementary Table 5).

We evaluated gender differences in perceived restorativeness (PRS) for different forest bathing activities, showing significant differences. Females consistently reported higher PRS across all groups, with the static group showing the largest effect size (d = 6.723, 95% CI [5.10, 8.35] for females vs. d = 0.048, 95% CI [−0.72, 0.82] for males), indicating substantial perceived psychological benefits for females. These findings, while based on small subgroups, reveal a consistent trend that merits confirmation in larger samples.

The combined dynamic-static group showed significant improvements in emotional outcomes and blood pressure regulation, with moderate effects on GSR. However, heart rate changes were minimal, and perceived restorativeness (PRS) was lower than in the static group. This likely reflects construct differences, as PRS emphasizes subjective environmental perceptions, whereas the combined group produced broader and more balanced benefits across physiological and psychological indicators. These outcomes may also be influenced by the shorter exposure to each activity compared with single-mode groups, transitional rest periods that rebalanced physiological responses (Song et al., 2017), and the additional cognitive load of switching activities for older adults. Taken together, the combined mode does not simply yield additive effects but instead provides a more integrated outcome, reflecting the complex nature of real-world forest therapy practices.

The study found that moderate dynamic activities like forest walking reduce stress hormones, promote relaxation, and balance the autonomic nervous system. They increase high-frequency heart rate variability, lower the low-frequency to high-frequency ratio, and reduce systolic and diastolic blood pressure (Wen et al., 2023).

In our study, dynamic activities notably improved heart rate and diastolic blood pressure, while EEG and GSR effects were minimal, possibly due to short exposure and lower sweating rates in older adults. The BPOMS results indicated reduced negative emotions and increased vitality, though overall benefits were greater in the combined group. PRS scores were highest for the dynamic group, reflecting that walking more strongly evokes immediate perceptions of restorativeness, even if comprehensive improvements were more evident in the combined mode.

Research shows that static activities like forest meditation can lower blood pressure, heart rate, and cortisol levels, reduce stress, anxiety, depression, and negative emotions, and increase calmness and happiness. These activities enhance immune function, boost natural killer cell activity, and improve attention and cognitive flexibility (Wen et al., 2019; Park et al., 2022; Stier-Jarmer et al., 2021; Twohig-Bennett and Jones, 2018).

Static activities offer varied benefits, with effects differing by individual, particularly in middle-aged and older females. Our study on single static activities found low EEG effect sizes, moderate GSR effects, and insignificant heart rate reductions, likely due to participants’ age. Static activities significantly regulated systolic blood pressure and reduced negative emotions and fatigue, improving vitality. Perceived restorativeness was high but lower than dynamic activities, possibly due to older adults’ preference for shorter-duration static activities.

Long-term or intense forest bathing may reveal trend-level psychological and physiological differences between genders. Males tended to show greater positive emotional response to combined dynamic-static activities, while females appeared to experience a more significant reduction in negative emotions, indicating better acceptance.

GSR results revealed exploratory evidence of higher effects in males due to greater sweat gland density and sweating rates, leading to more pronounced changes. Differences in thermoregulation and skin properties also may affect GSR sensitivity (Erfanian et al., 2019; Bari, 2020; Horvers et al., 2021; Posada-Quintero et al., 2018).

Females had slightly higher heart rates than males, with minimal difference. Males’ heart rates appeared to recover faster post-exercise, likely due to greater cardiac efficiency, while females’ recovery was slower, possibly due to lower cardiac output and smaller heart volume.

In combined dynamic-static activities, females had higher systolic but slightly lower diastolic blood pressure than males, with slower increases during exercise. These findings suggest a trend that forest bathing activities should be tailored to gender-specific characteristics.

The BPOMS questionnaire indicated that combined dynamic-static activities significantly improved positive emotions and reduced negative emotions in females, with greater total mood disturbance (TMD) improvement in females compared to males. Females may be more responsive to emotional regulation, using emotion-focused strategies like seeking support and expressing feelings. The sensory stimuli and social interaction from combined activities seem to help females alleviate negative emotions more effectively (Deng et al., 2016). Males often manage stress alone or via exercise, resulting in less emotional recovery than females. The perceived restorativeness questionnaire also showed that females tended to report significantly higher recovery effects than males (Deng et al., 2016).

Moderate dynamic activities like forest walking release stress hormones, trigger relaxation, reduce sympathetic activity, and enhance parasympathetic activity, lowering heart rate and blood pressure while boosting self-regulation. Dynamic forest bathing increased high-frequency heart rate variability, reduced the LF/HF ratio, and lowered systolic and diastolic blood pressure, and pulse rate (Wen et al., 2023).

In our study, dynamic activities notably impacted heart rate, but systolic blood pressure results differed from previous studies, with diastolic blood pressure showing the greatest improvement. Factors such as the national forest park setting, experiment duration, uncontrolled walking speed, and the older age of participants may explain these differences. EEG data revealed that single dynamic activities had the smallest effect on emotions, likely due to the limited emotional change they induce in older adults. GSR results also indicated the lowest effect size for single dynamic activities, possibly due to lower sweating rates and less sensitive skin responses in the older adult. The BPOMS questionnaire showed that single dynamic activities reduced negative emotions and increased vitality, but less effectively than combined activities. However, single dynamic activities had the highest restorative effects in the perceived restorativeness questionnaire, likely due to their nature.

In single static activities (e.g., forest meditation), males and females show trend-level differences in physiological and psychological responses. EEG data revealed that static activities significantly improved positive emotions in females but had a smaller effect on males. This may be because females use static activities for emotional regulation, while males prefer dynamic activities for stress relief (Kivikangas et al., 2014). GSR responses showed that males had a higher effect size during static activities, indicating they sweated more, which could be related to their higher basal metabolic rate and sympathetic nervous system response (Bari, 2020; Posada-Quintero et al., 2018).

For heart rate recovery, males showed a higher effect size than females, suggesting static activities might have been more effective for them. This may be due to better cardiac function in males, enhancing their recovery from static activities (Bari, 2020; Posada-Quintero et al., 2018). Static activities significantly regulated blood pressure in females, particularly diastolic pressure, likely due to their ability to relax and lower blood pressure more effectively (Posada-Quintero et al., 2018).

The BPOMS questionnaire showed that while males had a higher effect size for reducing negative emotions, they had lower vitality compared to females. Single static activities improved overall emotional health more for females, likely due to their greater sensitivity to emotional regulation and use of emotion-focused coping strategies. The perceived restorativeness questionnaire revealed that females tended to report significantly higher restorative effect than males, indicating better emotional recovery and regulation during static activities.

These differences are consistent with the idea that fundamental physiological and psychological variations between genders shape forest therapy responses. Males seem to exhibit stronger physiological responses in static activities, while females may excel in emotional regulation and recovery (Deng et al., 2016; Kivikangas et al., 2014). Therefore, forest bathing activities should consider these gender differences as potential factors to maximize health benefits.

Although several indicators showed statistically significant changes, some effect sizes (e.g., heart rate and EEG) were modest. This distinction highlights the difference between statistical and clinical significance. However, even small but consistent changes in physiological and psychological indicators may still be meaningful for middle-aged and older adult populations, where incremental improvements accumulate to long-term health benefits. Importantly, the overall trends across multiple measures were coherent and aligned with previous findings, supporting the exploratory robustness our results. Future studies with larger sample sizes and longer follow-up periods are needed to further establish the clinical relevance and generalizability of these findings.