Introduction

It has been widely reported that exposure of the five senses to nature—such as walking in the forest, viewing plants and flowers, or listening to natural sounds—can improve mood, reduce stress, and alleviate symptoms of depression and anxiety1,2,3,4. Natural stimuli have been shown to affect heart rate variability (HRV), an indicator of autonomic nervous system activity5. In healthy individuals, increases in high-frequency (HF) in HRV, suggesting enhanced parasympathetic nervous activity, and decreases in low-frequency/high-frequency ratio (LF/HF) in HRV, indicating reduced sympathetic nervous activity, have been reported6. However, other studies have found no significant changes in HRV6, and some even reported increases in LF/HF7, highlighting inconsistent results. Additionally, in our previous study with patients suffering from depression and anxiety disorders8, no significant changes in HRV were observed.

To address these inconsistencies, the “physiological adjustment effect” theory has been proposed3. While traditional studies have generally assessed the effects of natural stimuli using average values, recent research has highlighted the importance of considering individual differences, where physiological responses vary among individuals exposed to the same stimuli. Recognizing this variability, Miyazaki introduced the “Physiological Adjustment Effect,” a theory that aims to explain these divergent physiological changes3,9,10.

Grounded in Wilder’s “law of initial value”11, the Physiological Adjustment Effect theory posits that physiological values change toward an optimal level for the individual, depending on the baseline values before intervention. For instance, forest therapy has been reported to lower blood pressure in hypertensive subjects and raise it in those with low blood pressure9. A study involving 214 healthy individuals confirmed that visual stimulation with roses induces the physiological adjustment effect on the LF/HF index of HRV, an indicator of sympathetic nervous activity10. It is believed that this physiological adjustment effect could explain the individual variability in responses to natural stimuli.

In depression, decreases in LF and HF and increases in LF/HF have been reported12,13, while in anxiety disorders, decreased HF has been observed14. No studies have yet confirmed the physiological adjustment effect of natural stimuli on these patients. The primary aim of this study was to examine whether visual stimulation with natural images induces a physiological adjustment effect on HRV in patients with depression and anxiety disorders.

Furthermore, no studies have explored the relationship between the physiological adjustment effect of natural stimuli and mood changes. Therefore, the secondary aim of this study was to investigate the association between the physiological adjustment effect and mood changes.

Methods

Participants

This study is a secondary analysis using data from our previous research. In a randomized crossover trial, we reported the mood-enhancing effects and physiological changes induced by visual stimulation with natural images in 60 patients that had been diagnosed with depressive or anxiety disorders and were visiting an outpatient psychiatric clinic at the time of the enrollment8. Patient details are provided in our previous publication8. In the current study, we used the dataset to examine whether there was a physiological adjustment effect on HRV. One patient with depression and one with anxiety disorder were excluded due to missing HRV data, resulting in a total of 58 patients included in the analysis.

The average age of the patients was 45.3 ± 10.8 years, with 20 males and 38 females. Among them, 29 were primarily diagnosed with depressive disorder and 29 with anxiety disorder, with 9 patients having comorbid depressive and anxiety disorders. The mean score for depressive symptoms, measured by the Beck Depression Inventory (BDI-II), was 16.34 ± 11.38. The mean scores for state and trait anxiety, measured by the state-trait anxiety inventory (STAI), were 43.05 ± 9.46 and 50.81 ± 11.90, respectively. The study was approved by the Institutional Review Board of Yamaguchi University Hospital and conducted in accordance with the latest Declaration of Helsinki. Informed consent was obtained in writing from all participants.

Heart rate variability and physiological adjustment effect

The experimental intervention used in the study dataset was a 3 min visual stimulation with green natural images, while the control was a 3 min visual stimulation with urban images15. The stimuli consisted of 12 natural images and 12 control images. The natural images featured green trees in natural landscapes of forests and meadows, while the control images showed urban scenes with predominantly buildings. The images were selected as follows: three research staff members first collected copyright-free images and categorized them into four types: distant views, close-up views, views with paths, and upward views. From each category, the three most typical natural or artificial images were selected. Using these images, we conducted a subsequent study15 with healthy individuals and confirmed mood-enhancing effects and changes in prefrontal cortex oxygenated hemoglobin levels comparable to previously reported effect sizes16,17. Previous studies using natural visual stimuli have employed stimulus durations of 60 seconds18, 90 seconds17,19,20,21, 3 minutes16,22, and 6 minutes23,24. While there is generally a dose-dependent effect such that longer stimuli are associated with more enhanced effects, in the current study, we used visual stimuli via images and excessively long stimulus times could lead to decreased concentration. Therefore, a 3 min stimulus time was chosen. In fact, the effect size of the 3 min visual stimulus on mood15 was larger than that of the 6 min stimulus24. Although studies with 60 and 90 s stimuli did not report effect sizes for mood, which makes direct comparisons with the 3-min stimulus impossible, the effect size of the 90 s visual stimulus on prefrontal cortex oxygenated hemoglobin levels17 was comparable to that of the 3 min visual stimulus that we used15. Based on these findings, the 3 min stimulus duration was considered appropriate. Additionally, previous studies examining the effects of natural stimuli on HRV have reported washout periods of 30 s following 90 s visual stimuli21, and 1 min following 60 second18 or 90 s visual stimuli17,20,21. Compared to these studies, our study used a 3-min washout period following a 3 min visual stimulus, resulting in a similar or higher ratio of washout time to stimulus duration. Furthermore, our study assessed not only HRV but also frontal brain blood flow using fNIRS. In fNIRS research, it is generally considered appropriate for the washout period to be 1–1.5 times the duration of the stimulus task to allow brain blood flow to return to baseline25,26. Based on this standard, we adopted a 3 min washout period after the 3 min stimulus.

HRV was measured during the visual stimulation. HRV, which reflects the variability in the time intervals between consecutive R-waves (RR intervals) on an electrocardiogram, was analyzed using the ArteC-C system manufactured by Yumetica Co., Ltd.27,28. Figure 1 shows the study protocol, the images used for the stimuli, and photos of the ArteC-C system being worn. The power levels of the high-frequency component (HF; 0.15–0.40 Hz) and the low-frequency component (LF; 0.04–0.15 Hz) of HRV were calculated using the maximum entropy method. HF and LF/HF data were transformed into natural logarithms to normalize them29.

Fig. 1
figure 1

Study protocol, images used for stimulation, and photos of the measurement device in use. (A) Overall flowchart of the randomized crossover trial. (B) Example of green natural images. (C) Example of urban images. (D) Illustration of ArteC-C worn for HRV evaluation. HRV: Heart rate variability. Adapted from Mizumoto et al.8.

Full size image

The physiological adjustment effect theory posits that physiological values change towards an optimal value for the individual after intervention, depending on pre-intervention values3. For example, if the pre-intervention value is high and decreases after intervention, a physiological adjustment effect is considered present. A significant correlation between baseline values and changes induced by the intervention indicates a physiological adjustment effect10. Baseline values were defined as HRV during the control urban image stimulation, and change values were calculated by subtracting HRV during control image stimulation from HRV during natural image stimulation.

To examine the relationship between the physiological adjustment effect and mood, participants were divided into two groups: those with a physiological adjustment effect (n = 41), defined as having a low initial value (Ln(LF/HF) < 0) that increased after natural image stimulation, or a high initial value (Ln(LF/HF) > 0) that decreased after natural image stimulation, and those without a physiological adjustment effect (n = 17). We compared mood changes in Comfort, Relaxation, and Vigor between the two groups, as measured by the Chen-Hagiwara mood test (CHAMT) immediately after image viewing30. According to the two-dimensional valence-arousal theory of affect, Comfort reflects valence or pleasure, Relaxation represents a combination of positive valence and low arousal, and Vigor is associated with positive valence and high arousal. The CHAMT is useful because it allows for a more detailed evaluation of mood compared to existing measures and is less susceptible to comparative bias30. Firstly, traditional evaluation scales assess valence and arousal separately, which overlook subtle emotional changes arising from the interaction between valence and arousal, such as the difference between experiencing arousal with positive valence and experiencing it with negative valence. CHAMT can evaluate the synergistic effects of valence and arousal with the items of Relaxation and Vigor, making it particularly useful when investigating mood-enhancing effects. Secondly, unlike Likert scales, CHAMT uses a visual analog scale, where participants mark an "X" on a blank line. The distance (mm) from the left end to the mark is quantified and used as a mood scale from 0 to 100. This scale is less susceptible to comparative bias from previous responses, making it a suitable evaluation scale for crossover comparison studies. In contrast, Likert scales are easily remembered, which potentially causes respondents to unconsciously relate their current response to previous ones, leading to social desirability bias31.

Additionally, the effect of natural image stimulation on mood was examined separately in both groups with and without the physiological adjustment effect.

Statistical analysis

Matlab R2018b and IBM SPSS Statistics 28.0 were used for the analyses in this study. Normality was tested using the Shapiro–Wilk test. For paired samples, the paired t-test was used when data followed a normal distribution, and the Wilcoxon signed-rank test was used when data did not follow a normal distribution. For independent samples, the independent t-test was applied if the data were normally distributed, and the Mann–Whitney U test was used otherwise. Correlation analysis employed Pearson’s correlation coefficient for normally distributed data and Spearman’s correlation coefficient for non-normally distributed data. For contingency table analyses, the Chi-square test or Fisher’s exact test was applied depending on the expected frequencies. Statistical significance was set at p < 0.05, two-sided.

Results

Physiological adjustment effect in all participants

To examine the physiological adjustment effect of visual stimulation with natural images, we used the HRV measured during the control urban image as the baseline value and investigated the relationship between this baseline and the change in HRV induced by the intervention with natural images. A statistically significant negative correlation was found between the baseline LF/HF ratio, an indicator of sympathetic nervous activity, and the change in LF/HF following natural image stimulation (Rho = -0.450, p < 0.001, Fig. 2). This indicates that participants with a high initial LF/HF ratio showed a decrease after natural stimulation, while those with a low initial LF/HF ratio showed an increase, demonstrating a physiological adjustment effect. No significant correlation was found for HF, an indicator of parasympathetic nervous activity.

Fig. 2
figure 2

Relationship between baseline values and HRV changes induced by visual stimulation with natural images. The left panel shows the results for LF/HF, an indicator of sympathetic nervous activity, and the right panel shows the results for HF, an indicator of parasympathetic nervous activity. Ln(HF) represents the natural logarithm of the 3 min average HF value, and Ln(LF/HF) represents the natural logarithm of the 3 min average ratio of LF to HF. The horizontal axis indicates the baseline values measured during the control urban image stimulation, while the vertical axis shows the changes in HRV induced by natural image stimulation compared to the urban images, calculated as the value from the natural images minus the value from the urban images. Statistically significant correlations (p < 0.05) identified by Pearson or Spearman correlation analysis are noted. A statistically significant negative correlation was found between the baseline LF/HF and the change induced by natural image stimulation (Rho = −0.450, p < 0.001). Even after excluding outliers in Ln(LF/HF), defined as values exceeding the third quartile + 1.5 × IQR or below the first quartile—1.5 × IQR (n = 52), a significant negative correlation remained (r = −0.295, p = 0.034).

Full size image

Relationship between physiological adjustment effect and mood improvement

Among the participants, 41 (71%) showed a physiological adjustment effect, defined as having a low initial value (Ln(LF/HF) during city < 0) that increased after natural image stimulation (data points in the upper left or second quadrant of Fig. 2, left panel), or a high initial value (Ln(LF/HF) during city > 0) that decreased after natural image stimulation (data points in the lower right or fourth quadrant of Fig. 2, left panel). Conversely, the remaining 17 participants (29%) did not exhibit this adjustment effect.

To explore how the physiological adjustment effect relates to mood improvement, we compared mood changes in response to viewing natural versus urban images in groups with and without the physiological adjustment effect. In the group with a physiological adjustment effect (n = 41), viewing natural images significantly enhanced feelings of being Comfortable (p < 0.001, d = 0.888), Relaxed (p < 0.001, d = 0.948), and Vigorous (p < 0.001, d = 0.808) compared to urban images (Fig. 3, left panels). Conversely, in the group without the physiological adjustment effect (n = 17), significant mood improvements were seen for Comfortable (p = 0.004, d = 0.740) and Relaxed (p = 0.003, d = 0.833), but no significant change was observed in Vigorous (p = 0.217) (Fig. 3, right panels).

Fig. 3
figure 3

Effects of visual stimulation with natural images on mood groups in groups with and without physiological adjustment effect. The gray bars represent mood values after urban image stimulation, and the green bars represent mood values after natural image stimulation. The left graph shows the group with a physiological adjustment effect, while the right graph shows the group without it. Both groups showed significant improvements in Comfortable and Relaxed moods with natural image stimulation. However, a significant improvement in Vigorous was observed only in the group with a physiological adjustment effect. **p < 0.01, ***p < 0.001, paired t-test or Wilcoxon signed-rank test.

Full size image

We compared the differences in mood after viewing natural images versus urban images between the two groups: those with a physiological adjustment effect (n = 41) and those without (n = 17). The group with a physiological adjustment effect showed a greater increase in Vigorous (natural images minus urban images), with a trend toward statistical significance (p = 0.058, d = 0.558). However, no significant differences were found between the groups for Comfortable (p = 0.348) and Relaxed (p = 0.209) (Fig. 4).

Fig. 4
figure 4

Comparison of mood changes after natural image stimulation between groups with and without physiological adjustment effect. The gray bars represent the group without the adjustment effect (n = 17), and the pink bars represent the group with the adjustment effect (n = 41). In Vigorous, the group with the physiological adjustment effect showed a significantly greater mood enhancement trend (natural images minus urban images) compared to the group without the effect (p = 0.058, d = 0.558, independent t-test).

Full size image

Characteristics of participants with and without a physiological adjustment effect

We explored the characteristics of participants with and without a physiological adjustment effect. Statistical comparisons revealed that participants without the physiological adjustment effect were significantly more likely to report alcohol use compared to those with the effect (52.9 vs. 24.4%, p = 0.035, Chi-square test, Fig. 5). Additionally, there was a trend towards a higher prevalence of hypertension among participants without the adjustment effect compared to those with it (17.6 vs. 2.4%, p = 0.071, Fisher’s exact test). No other significant differences were found between the groups, as shown in (Fig. 5).

Fig. 5
figure 5

Comparison of characteristics between participants with and without a physiological adjustment effect. The gray bars represent the group without the adjustment effect, and the pink bars represent the group with the adjustment effect. The first panel displays data on sex, education, smoking, and alcohol consumption. The second panel presents age and symptoms of depression, state anxiety, and trait anxiety, as measured by BDI-II and STAI. The third panel shows data on primary diagnoses and comorbidities. Here, autonomic disease encompasses hypertension, diabetes, and thyroid conditions. The fourth panel illustrates medication use, including antidepressants, antianxiety drugs, antipsychotics, and mood stabilizers. *p < 0.05.

Full size image

Discussion

In patients with depressive or anxiety disorders, visual stimulation with natural images demonstrated a physiological adjustment effect in LF/HF of HRV. This result is consistent with previous findings in healthy individuals exposed to visual stimulation with fresh roses10. On the other hand, no significant physiological adjustment effect was observed in the HF component of HRV.

The LF/HF ratio, which demonstrated a physiological adjustment effect in this study, is considered an indicator of sympathetic nervous activity5. It has been reported to increase under high-stress conditions32,33 and decrease during drowsiness34, suggesting its association with stress response and arousal levels. Stimulation from nature has been shown to decrease LF/HF during relaxation6,21,35, which is interpreted as a reduction in sympathetic nervous activity.

In environmental psychology, theories explaining nature-based recovery include the Stress Reduction Theory36 and the Attention Restoration Theory37. Stress Reduction Theory posits that natural environments can block negative thoughts and emotions, thereby reducing stress responses. Attention Restoration Theory suggests that natural elements, such as vegetation, have the power to capture attention effortlessly, which enables the recovery of neurocognitive mechanisms that support directed attention based on effort38. Previous studies have shown that viewing or visiting natural environments is associated with a negative correlation where blood pressure and sympathetic nervous activity decrease more significantly when initial levels are higher9,10.The observed physiological adjustment effect in LF/HF suggests that visual stimulation with natural images may help regulate autonomic nervous system balance, which brings the body to an optimal arousal state. This optimal state, characterized by activation, energy, and alertness, are key elements of vigor. Vigor represents a state where positive valence (pleasant emotions) and positive arousal (heightened activation) coexist30, aligning closely with increased sympathetic activity. This study was conducted and analyzed to validate the physiological adjustment effect theory, which posits that, depending on the baseline values before intervention, post-intervention values shift in a healthy direction for each subject, i.e., towards an optimal value. Therefore, from the perspective of the adjustment effect theory, both an increase in low values and a decrease in high values are considered similar in that they both approach an optimal value.

It has been reported that LF/HF increases in high-stress situations32,33 and depression12,13. Therefore, excessively high LF/HF may lead to unpleasant sensations and be perceived as tension rather than vigor. Consequently, it is possible that the adjustment of excessively high LF/HF to an optimal value after natural stimulus intervention resulted in a shift from tension to positive emotions such as vigor. Importantly, feelings of vigor have been considered a vital affective resource39. Vigor plays a crucial role in driving engagement in daily activities, including work and goal pursuit. It is essential for effectively tackling challenges and managing stress, which supports overall mental health40. Moreover, vigor is associated with improved sleep quality41, enhanced resilience in the face of adversity42, and greater creativity43. In the workplace, feelings of vigor boost work engagement and productivity39. These findings not only highlight the importance of vigor in promoting psychological well-being and optimal functioning across various aspects of life but also underscore the significance of the physiological adjustment effect, as it appears essential for the occurrence of vigor after viewing natural images.

In contrast, no significant association was found between physiological adjustment effects and improvements in comfort or relaxation, suggesting that these mood changes may be mediated by mechanisms independent of the autonomic nervous system. Comfort and relaxation are characterized by low arousal and increased parasympathetic activity rather than sympathetic activation. These states are typically associated with reduced stress and enhanced calmness, often reflected by increases in HF components of HRV rather than changes in the LF/HF ratio, with HF being a widely recognized indicator of parasympathetic nervous activity5. The absence of significant changes in HF in this study indicates that visual stimulation with natural images may have limited effects on the parasympathetic nervous system, particularly in patients with depressive and anxiety disorders.

The results of this study offer valuable insights into the potential benefits of nature therapy for individuals with depressive and anxiety disorders. Similar to findings in healthy individuals, a large majority of patients with these mental health conditions also exhibited physiological adjustment effects, which were associated with improvements in vigor—a state characterized by energy, alertness, and positive mood. However, a notable finding was that patients without physiological adjustment effects had a higher prevalence of alcohol consumption and showed a tendency towards a higher prevalence of comorbid hypertension.

Alcohol consumption is well-documented to activate the sympathetic nervous system and lead to increased heart rate and blood pressure44,45. Chronic alcohol use further impairs autonomic regulation and disrupts the balance between sympathetic and parasympathetic activity46. Similarly, hypertension is closely linked to elevated sympathetic nerve activity and contributes to ongoing autonomic dysfunction47,48. These factors likely contribute to the observed differences in the presence of physiological adjustment effects among patients, highlighting how lifestyle behaviors and physical health conditions can influence responses to nature therapy.

Although these dietary behaviors and comorbid physical diseases are not directly related to the core psychopathology of depressive and anxiety disorders, they significantly affect how patients respond to nature exposure and their subsequent mood improvements in terms of vigor. This suggests that patients who do not exhibit physiological adjustment effects may have underlying autonomic nervous system abnormalities, particularly within the sympathetic nervous system, that interfere with the therapeutic benefits of nature exposure.

Given the role of autonomic dysfunction in modulating the body’s response to environmental stimuli, targeted therapeutic approaches that address these autonomic imbalances may enhance treatment outcomes for patients with depressive and anxiety disorders. Further research is warranted to explore the broader implications of physiological adjustment effects in the context of depressive and anxiety disorders. Understanding how autonomic nervous system regulation impacts the therapeutic response to nature could lead to more personalized and effective treatment strategies. By addressing both mental and physical health factors, a more holistic approach to managing depressive and anxiety disorders can be developed, which may help enhance overall patient well-being.

This study has several limitations. Firstly, in this paper, LF/HF was used as an index of sympathetic nervous system activity, but there are also criticisms that LF/HF is not appropriate as an index of sympathetic nervous system activity due to the non-linear relationship between sympathetic and parasympathetic nervous system activity49. However, as mentioned above, LF/HF has been widely used in recent studies as an index of sympathetic nervous system activity6,21,32,33,34,35. Future studies are required to develop more advanced indices and further validate our findings. Secondly, following many previous studies, we used LF/HF and HF as indices of HRV. However, in recent years, indices such as the Standard Deviation of NN intervals (SDNN) have also been adopted. Future research should validate our findings using SDNN and similar alternative indices. Thirdly, it is generally considered that vigor increases with heightened sympathetic nervous system activity. In the current study, the observed increase in vigor among the “physiological adjustment effect group” (n = 41) may primarily reflect the effects seen in participants with initially low values, who comprised 29 individuals—accounting for 71% of the group. For a more rigorous evaluation, it would be necessary to first divide participants into two groups, an “initial low-value group” and an “initial high-value group,” and then compare participants with increased values to those without in the “initial low-value group,” and subjects with decreased values to those without in the “initial high-value group.” However, our dataset included only 58 subjects, with 41 in the “initial low-value group” and just 17 in the “initial high-value group”. The small sample size of the “initial high-value group” (n = 17) lacked sufficient statistical power to detect any meaningful relationships. Therefore, additional analyses based on this classification were not performed.

Fourthly, we did not measure respiratory rate or examine the influence of respiratory state in this study. The relaxing effect of natural environmental stimuli naturally reduces respiratory rate50. Therefore, changes in respiratory rate are part of the relaxation effect, and controlling for its influence would mitigate or eliminate the relaxation effect itself. Consequently, respiratory rate was not considered in this study. However, future research should explore the role of respiratory rate in the physiological adjustment effect more thoroughly.

Conclusion

This study examined the physiological adjustment effects on HRV response to visual stimulation with natural images in patients with depressive and anxiety disorders. Patients with initially high LF/HF ratios showed a decrease in LF/HF following natural image stimulation, while those with initially low values showed an increase, demonstrating a physiological adjustment effect. This adjustment was linked to improved vigor, suggesting that natural interventions may boost mood through this mechanism. However, the absence of adjustment effects was associated with behaviors and conditions related to autonomic dysfunction, which highlights the need for targeted therapies to address these imbalances in depressive and anxiety disorders.