Psychophysiological Effects of a Natural Forest Environment on Chinese University Students Aged 19-25 Years

ABSTRACT

Recall of natural environments evokes human psychophysiological response, there is a lack of systematic, comprehensive analysis in existing research. Moreover, the limited experimental conditions lead to various laboratory research, and more field studies are needed. Employing the field research design and relative environmental psychology methods, this research investigates physical and psychological indexes changes of 60 Chinese university students aged 19–25, which include positive affect (PA), negative affect (NA), heart rate (HR), heart-rate variability (HRV), concentration (Pa), and relaxation (Pm), and regards the Zhangjiajie National Forest Park in China as the experimental site. This study evaluates the psychophysiological effects of the natural forest environment on participates and explores the overall psychophysiological changes, and verifies the positive effect of the natural forest environment on university students. The main findings indicated that the natural forest environment was beneficial for increasing positive affect, decreasing negative affect, affecting HR, and increasing HRV, Pa and Pm. This study adds some basic evidence to a growing theoretical literature emphasizing the role of natural forest environment in psychophysiological restoration.

KEYWORDS:

• Natural forest environment
• environmental psychology
• environmental restoration
• psychophysiology

Introduction

With the pace of people’s lives speeded up in modern society, people especially who work in urban areas are facing great pressure from all sides. People are emotionally exhausted from work and life, which in turn leads to excessive consumption and the increased occupation of psychological resources. Such mental states can produce a tendency toward laxity and disorder, leading to the phenomenon of “spiritual entropy”(Csikszentmihalyi, 2008). Those who live in big cities also face negative factors such as haze, traffic congestion, and noise. Given such phenomena, the incidence of psychological burn out, depression, and exhaustion syndrome has been increasing. Psychology researchers have therefore begun to pay attention to ways to recover human psychological functioning. Faced with mounting challenges to human health and the high costs associated with traditional medical methods, public health institutions are seeking preventive health solutions, with an increasing focus on using natural resources as a means to promote health (Kathleen, 2017). In this regard, it is widely believed that exposure to natural forest environment can have certain health benefits (Park et al., 2015).

Although some scholars have used field research methods to interpret the psychophysiological restorative effects of the natural environment, there is still limited evidence shown in current studies. Besides, the effect of the natural environment on psychophysiology is a complex process, and more valuable indicators need to be selected to fully reveal this process. To achieve the overarching research purposes, this research adopted the method of field research, and selected 60 Chinese university students as the research objects, and used PANAS to judge the emotional changes of the experimental participates, the ECG data HR and HRV and the EEG data α, β, Y, θ and δ waves to analyze the physiological changes. Meanwhile, this study analyzed the effects of the natural environment on human psychology and physiology relatively and comprehensively. This research exploratorily studied the effects of emotion on the changes of physiological indicators, and explained to what extent the emotional indicator changes affect the physiological indicator changes. The experiment was carried out in a subtropical evergreen and deciduous broad-leaved mixed forest, and the results verify the effect of this forest vegetation type on the psychological and physiological restoration of university students. This finding is expected to contribute to the literature by adding evidence to the effects of different forest vegetation types on human psychology and physiology. Meanwhile, this research contributes to improve the public’s awareness of forest conservation by guiding people walk into forests and enjoy the picturesque view here, which also contributes to the sustainable management and utilization of the forest.

Literature review

The natural environment not only provides material resources for survival but can also supply the psychological resources needed to cope with modern life. Rich natural environment can moderately cause indirect attention, thus allowing directional attention to be supplemented(Berto, 2005). The recovery of attention is the result of interactions between humans and a certain environment such as walking in the forest(White et al., 2013), backpacking in the wild(Hartig et al., 1997), visiting a garden or zoo(Pals et al., 2009). Such activities can promote attention recovery because of the corresponding decrease in attention fatigue or stress(Berto, 2005; Korpela et al., 2001). In this case, the environment can be regarded as a restorative environment, which is one characterized by conditions that can support individual recovery. As for the effects of natural environments on individual recovery have been often explored by focusing on demonstrating the value of restorative environments(Elisabet et al., 2014; Melissa et al., 2014) .

Contacting with nature can promote psychological recovery. Experiencing nature can reduce stress(Shin, 2007), improve subjective well-being(Stieger et al., 2020; White et al., 2017), strengthen cognitive function(Jones et al., 2021; Marc et al., 2008), and strengthen personal ties to others(Richardson et al., 2016; Weinstein et al., 2015). Emerging evidence showed that natural environments can promote recovery, reduce the negative effects of stress, and enhance positive affect(Kabisch et al., 2021; S. Park et al., 2022; Takayama et al., 2014). Pals et al. (2009)developed the Perceived Restorative Characteristics Questionnaire (PRCQ) to investigate the restorative effects of scenic spots with animals as the main attraction. Martens et al. (2011) found significant differences in the restorative effects of virgin, secondary, and artificial forests on mental health. Gatersleben and Andrews (2013) found that natural environments with a clear field of vision and low concealment had greater restorative effects and better regulation of negative affect than those with more concealment. Chiang et al. (2017) found that internal forest conditions produced significant stress recovery while forest edges were related to better attention recovery; further, high vegetation density produced better attention recovery, though there was a higher preference for medium vegetation density. The range of environment suitable for human survival is limited. When certain environmental factors are above or below a certain threshold, they will have direct or indirect effects on human physiology and pathology. Therefore, previous studies have found that forest environments can reduce blood pressure(Song, Harumi, Yoshifumi et al., 2015b), heart rate(Park et al., 2015), skin conductance(Parsons et al., 1998), sympathetic nerve activity(B-J Lee et al., 2019; Park et al., 2010), and cortisol(Tsunetsugu et al., 2013). In addition, the stimulation of a forest environment has been found to reduce the blood flow of the prefrontal cortex. A natural forest environment can increase the activity of natural killer (NK) cells and enhance immunity; such effects can last for at least seven days. Bratman et al. (2015)suggested that transient natural experience can reduce rumination and neural activity in the prefrontal cortex.

Human responses to forest environments are the result of multiple sensors such as vision, hearing, touch, and taste(Yuko et al., 2007). It is necessary, therefore, to conduct empirical field research to verify the effects of natural forest environments on human psychophysiological health. Field research can better present the actual effects of natural forest environment(Groenewegen et al., 2006). Despite the strong demand for field research, given the difficulty of sampling and analyzing physiological indexes in the field, previous studies often used a limited number of sense stimulations in their experiments. Based on above discussion, the following hypotheses are proposed:

Hypothesis 1. The natural forest environment leads to affect changes. Specifically, it has a boost (weakened) effect on people’s positive (negative) affect.

Hypothesis 2. The natural forest environment causes physiological responses. Specifically, it has a negative (positive) influence on people’s HR (HRV, Pa, Pm).

Hypothesis 3. The correlation exists between physiological responses and affect changes. Specifically, it may differ between positive and negative affect, and for each physiological response indicator, the correlation varies as well.

Methodology

Experimental objective

By comparing changes in the psychophysiological indexes of participants who normally lived and studied in the city with those who entered a forest environment, this study identified the regulating effect of a natural forest environment on human psychophysiology.

Experimental site

The experiment was conducted at the front of the scenic Jinbianxi spot in Zhangjiajie National Forest Park. The park was established in 1982 and is the first national forest park in China, with a vegetation coverage rate of 98% (Yang et al., 2006), and the vegetation is dominated by subtropical evergreen and deciduous broad-leaved mixed forest(Zhang et al., 2017). The site is relatively flat and open, where the light is suitable. Therefore, it is suitable for walking, and the slope of the road is comfortable (no higher than 3 degrees). There are several rest places, and the environment is relatively quiet.

Selection of participants

Sixty participants were selected through volunteer recruitment. For data validity, participants were required to have lived in a city for a long time, be healthy, have no smoking habit, have no major emotional fluctuations, have no history of mental disorders or allergic diseases, and have not used drugs. The participants were university students (48 undergraduates, 12 postgraduates) with an average age of 22 years; there were 30 males and 30 females.

Experiment time

The experiment was carried out in the summer. The weather was hot, the forest microclimate phenomenon was obvious, and the forest showed more color.

Psychological measurement

PANAS was used to measure self-reported affect. Ten words in the scale represent positive affect (PA), and 10 represent negative affect (NA). All are positively scored (1 = almost none; 2 = less; 3 = medium; 4 = more; 5 = very much). Questions 1, 3, 5, 9, 10, 12, 14, 16, 17, and 19 measure PA; the higher the number is, the more positive the affect will be. Questions 2, 4, 6, 7, 8, 11, 13, 15, 18, and 20 measure negative affect likewise. A high score for PA indicates an energetic, engrossed, and happy mood; a low score indicates indifference. A high NA score indicates a subjective feeling of confusion and pain while a low score indicates calmness.

Physiological measurement

Studies have confirmed that physiological signals such as electroencephalogram (EEG), heart rate (HR), and heart rate variability (HRV) contain rich emotional information, and the natural environment has a strong correlation with changes in these indicators(Hernando et al., 2018). The physiological signals used in this study included EEG, HR, and HRV. By collecting information, we could judge the participant’s psychological stress state and then judge the influence of the natural environment on health and physiology. In this study, the collected EEG data were processed to obtain the eSense index, which was used to describe the degree to which participants were entering a state of concentration and relaxation. When people are paying attention, the y wave and Y wave account for a large proportion of the energy of the EEG signals. When people are in a state of fatigue, the theta wave and δ wave account for a large proportion of EEG energy. In this way, we can obtain the formula to calculate concentration and relaxation. The formula for the “eSense concentration index” is as follows:

(1) $\mathrm{P}\mathrm{a}=\left(\mathrm{m}\mathrm{Y}+\mathrm{n}\beta +\mathrm{t}\alpha \right)\times 100$(1)

where Pa is concentration; y, β, and α are, respectively, the percentages of Y wave, y wave, and α wave in EEG energy; and m, n, and t are the weight coefficients of Y wave, β wave, and α wave, respectively, which can be obtained through analytic hierarchy process (AHP).

The eSense concentration index indicates the intensity of participants’ mental “concentration” level or “attention” level. The index values range from 0 to 100. The higher the score is, the more focused the attention will be. The formula for the “eSense relaxation index” is as follows:

(2) $\mathrm{P}\mathrm{m}=\left(\mathrm{x}\theta +\mathrm{y}\delta +\mathrm{z}\alpha \right)\times 100$(2)

where Pm represents relaxation; θ, δ, and α represent, respectively, the percentage of θ wave, δ wave, and α wave in the energy of the EEG signal; and x, y, and z represent the weight coefficients of θ wave, δ wave, and α wave, respectively, which can be obtained by AHP. The eSense relaxation index indicates participants’ level of mental “calm” or “relaxation.” The index values range from 0 to 100; the higher the score, the more relaxed.

Experimental process and control

The experiment was conducted in three phases. Phase 1: Before starting, the participants entered the laboratory and was explained the experimental rules, processes, and precautions. After 10-minute rest, the questionnaire survey of PANAS was used to obtain participants’ affect data. The measurement of physiological indicators included Pa, Pm, HR, and HRV. The acquisition times for these indicators were divided into three Phases: 24 hours before going, 24 hours in the park, and 24 hours after coming back. After obtaining the initial physiological index data, the experimenters and participants went to Zhangjiajie National Forest and the experiment entered Phase 2. The experimental site had relatively few tourists, but with a wide field of vision, a quiet environment, mountain and water views, and resting facilities. Participants were required to carry out activities within a designated scope, ensuring that the activities were gentle and effortless (e.g., walking, sitting, enjoying scenery). To avoid the effects of “unexpected” events on affect, participants were asked not to use mobile devices or chat with each other during the activity, which lasted about 15 minutes. Then, participants were asked to fill out the PANAS questionnaire, and EEG data were obtained during this time. Other indicators included 24-hour data. The participants’ physiological data were also monitored while in the park. After the experiments were completed, the participants left Zhangjiajie National Forest Park and the experiment entered Phase 3. Participants were asked to begin the 24-hour monitoring at 20:00 in the evening and return to the laboratory at 15:00 in the afternoon the next day for the experiment. After a 10-minute rest, the questionnaire survey and physiological index measurement were conducted.

Results

Affect changes

In psychology experiments, 0.70 is generally considered as an acceptable scale reliability value(Nunnally, 1978). The results showed that the sample reliability values of Phases 1, 2, and 3 of the experiment were 0.768, 0.778, and 0.855, respectively, indicating high reliability. When analyzing the difference of indicators between any two Phases, if the results in all Phases meet the normal distribution, the Paired sample t-test is used, while if any one of them does not meet the normal distribution, the Wilcoxon test is used.

Positive affect

The Wilcoxon test was used to examine the differences in positive affect between Phase 1 and Phase 2, and between Phase 2 and Phase 3. Paired-sample t-tests were used to verify the differences in positive affect between Phase 1 and Phase 3. As can be seen from Figure 1, there were significant differences in positive affect between Phase 1 and Phase 2 (p = .000 < 0.01), Phase 1 and Phase 3 (p = .001 < 0.01), Phase 2 and Phase 3(p = .011 < 0.05).

Figure 1. Changes in participants’ positive affect (n = 60).

There was 75% of participants in Phase 2 whose positive affect increased than Phase 1, and 58% of participants in Phase 3 whose positive affect gained compared with Phase 1. Meanwhile, there was 31% of participants in Phase 3 whose positive affect was more than that in Phase 2. Generally, the natural forest environment had a regulatory effect on the participants’ affect, and the range of positive and negative affect changes varied with individual differences, which supported hypothesis 1.

Cohen’s d was used to indicate the magnitude of effect (difference amplitude); the larger the value is, the greater the difference will be. When Cohen’s d was used to indicate the effect magnitude in paired-sample t-tests, the critical points for distinguishing small, medium, and large effect were 0.20, 0.50, and 0.80. The result shows that Cohen’s d between Phase 1 and Phase 2 was 0.817; Phase 1 and Phase 3 was 0.458; Phase 2 and Phase 3 was 0.356, which shows that the natural forest environment had a significant regulatory effect on positive affect.

Negative affect

The Wilcoxon test was used to test the differences in negative affect among the three phases. As can be seen from Figure 2, there were significant differences in negative affect between Phase 1 and Phase 2 (p = .000 < 0.01), Phase 1 and Phase 3 (p = .000 < 0.01), but there was no significant difference in negative affect between Phase 2 and Phase 3 (p = .452 > 0.05).

Figure 2. Changes in participants’ negative affect (n = 60).

There was 81% of participants in Phase 2 whose negative affect was less than that in Phase 1, and 71% participants in Phase 3 whose negative affect was inferior to that in Phase 1. Besides, there was 33% of participants in Phase 3 whose negative affect decreased than that in Phase 2. The results revealed that the natural forest environment has a dented influence on people’s negative affect, which supported hypothesis 1. The Cohen’s d between Phase 1 and Phase 2 was 0.556; Phase 1 and Phase 3 was 0.582, which shows that the natural forest environment had a significant regulatory effect on the participants’ negative affect.

Physiological response

HR

The Wilcoxon test was used to test the differences in HR between Phase 1 and Phase 2 and that between Phase 1 and Phase 3, and Paired-sample t-tests were used to test the differences in HR between Phase 2 and Phase 3. As can be seen from Figure 3, there were significant differences in HR between Phase 1 and Phase 2 (p = .004 < 0.01), Phase 1 and Phase 3 (p = .000 < 0.01), Phase 2 and Phase 3(p = .000 < 0.01).

Figure 3. Changes in participants’ HR (n = 60).

Compared to Phase 1, there was 65% of participants in Phase 2 whose HR increased, and 78% of participants in Phase 3 whose HR decreased. Compared to Phase 2, there was 93% of participants whose HR decreased in Phase 3. In general, the natural forest environment was related to a decrease in participants’ HR, but the range of HR decrease varied among them, which partially supported hypothesis 2. The Cohen’s d between Phase 1 and Phase 2 was 0.420; Phase 1 and Phase 3 was 0.812; Phase 2 and Phase 3 was 1.403, which shows that the forest environment had a significant regulatory effect on HR, and the reduction of HR was mainly reflected in Phase 3.

HRV

The Wilcoxon test was used to verify the differences in HRV between Phase 1 and Phase 2 and that between Phase 1 and Phase 3, and Paired-sample t-tests were used to inspect the differences in HRV between Phase 2 and Phase 3. As can be seen from Figure 4, there were significant differences in HRV between Phase 1 and Phase 2 (p = .000 < 0.01), Phase 1 and Phase 3 (p = .000 < 0.01), but there was no significant difference in HRV between Phase 2 and Phase 3 (p = .663 > 0.05).

Figure 4. Change in participants’ HRV (n = 60).

Compared to Phase 1, there was 68% of participants in Phase 2 whose HRV increased significantly, and 75% in Phase 3 whose HRV added as well. In addition, there was 53% of participants in Phase 3 whose HRV gained compared with that in Phase 2. These results supported hypothesis 2. The Cohen’s d between Phase 1 and Phase 2 was 0.428; Phase 1 and Phase 3 was 0.530. Thus, the results showed that the natural forest environment had a significant effect on HRV.

Pa

Paired-sample t-tests were used to test the differences in Pa among the three phases. As can be seen from Figure 5, there were significant differences in Pa between Phase 1 and Phase 2 (p = .000 < 0.01), Phase 1 and Phase 3 (p = .000 < 0.01), but there was no significant difference in Pa between Phase 2 and Phase 3 (p = .0502 > 0.05).

Figure 5. Changes in participants’ Pa (n = 60).

Compared to Phase 1, there was 73% of participants in Phase 2 and 86% in Phase 3 whose Pa increased. Besides, there was another 35% increased from Phase 2 to Phase 3. The results supported hypothesis 2. The Cohen’s d between Phase 1 and Phase 3 was 0.867; Phase 1 and Phase 2 was 1.337, which shows that the natural forest environment could significantly improve Pa.

Pm

Paired-sample t-tests were used to test the differences in Pm between the three. As can be seen from Figure 6, there were significant differences in Pm between Phase 1 and Phase 2 (p = .000 < 0.01), Phase 1 and Phase 3 (p = .000 < 0.01), Phase 2 and Phase 3(p = .000 < 0.01).

Figure 6. Changes in participants’ Pm (n = 60).

Compared to Phase 1, there was 95% of participants in Phase 2 and 48% in Phase 3 whose Pm increased. While compared to Phase 2, there was 83% of participants in Phase 3 whose Pm decreased, which partially supported hypothesis 2. The Cohen’s d between Phase 1 and Phase 2 was 1.399; Phase 1 and Phase 3 was 0.644; Phase 2 and Phase 2 was 0.824, which shows that the natural forest environment could significantly improve Pm.

Interactive analysis of psychophysiology

Correlation analysis between PANAS and physiological indexes

Table 1 shows the correlations between HR, HRV, Pa, and Pm and positive and negative affect. Spearman’s correlation coefficient was used to express the strength of the correlations. HR and positive affect did not show a significant correlation (p = .881 > 0.05). HR and negative affect showed a significant correlation (p = .026 < 0.05). HRV and positive affect showed a significant correlation (p = .044 < 0.05). HRV and negative affect showed a significant correlation (p = .033 < 0.05). Pa and positive affect showed a significant correlation (p = .047 < 0.05). Pa and negative affect did not show a significant correlation (p = .554 > 0.05). Pm and positive affect showed a significant correlation (p = .036 < 0.05). Pm and negative showed a significant correlation (p = .021 < 0.05). Above all, which partially supported hypothesis 3.

Table 1. Correlations among affect indexes and physiological indexes.

PANAS		HR	HRV	Pa	Pm
Positive affect	Correlation coefficient	0.021	0.416*	0.401*	0.405*
	p	0.881	0.044	0.047	0.036
Negative affect	Correlation coefficient	0.418*	−0.505*	−0.218	−0.329*
	p	0.026	0.033	0.554	0.021

*p < 0.05.

Ridge regression analysis of PANAS and physiological indexes

Taking negative affect as the independent variable and HR as the dependent variable for ridge regression analysis, the K value was 0.990. Table 2 shows that the R squared value of the model was 0.071, which means that negative affect could explain the 7.1% change in HR. The model passed the F-test (F = 2.286, p = .033 < 0.05); thus, negative affect influenced HR. The model formula was HR = 72.115–0.062 * negative affect.

Table 2. Ridge regression analysis of negative affect and HR.

Name	Nonstandard coefficient		Standard coefficient	t	p	R ^2	Adjusted R ^2	F
	B	Standard error	Beta
Constant	72.115	1.854	-	38.901	0.000**	0.071	0.035	F (1,58) = 2.286, p = .033*
Negative affect	−0.062	0.100	0.041	0.616	0.026*

The dependent variable is HR. *p < 0.05, **p < 0.01.

Negative and positive affect were taken as the independent variables and HRV was takeTTn as the dependent variable for ridge regression analysis. The K value was 0.990. Table 3 shows that the R squared value of the model was 0.114, which means that negative and positive affect could explain 11.4% of the change in HRV. The model passed the F-test (F = 1.404, p = .022 < 0.05), indicating that negative and positive affect influenced HRV. The regression coefficient of negative affect was −0.106 (t = −0.145, p = .032 < 0.05), which means that negative affect influenced HRV. The regression coefficient of positive affect was 0.116 (t = 1.038, p = .028 < 0.05), which means that positive affect influenced HRV. The model formula was HRV = 42.562–0.106 * negative affect + 0.116 * positive affect.

Table 3. Ridge regression analysis of affect and HRV.

Name	Nonstandard coefficient		Standard coefficient	t	p	R^2	Adjusted R^2	F
	B	Standard error	Beta
Constant	42.562	4.681	-	9.093	0.000**	0.114	0.038	F (2,57) = 1.404, p = .022*
Negative affect	−0.106	0.141	−0.003	−0.145	0.032*
Positive affect	0.116	0.112	0.068	1.038	0.028*

The dependent variable is HRV. *p < 0.05, **p < 0.01.

Taking positive affect as the independent variable and Pa as the dependent variable, the K value was 0.990. Table 4 shows that the R squared value of the model was 0.045, which means that positive affect could explain the 4.5% change in Pa. The model passed the F-test (F = 1.863, p = .035 < 0.05); thus, positive affect influenced Pa. The model formula is Pa = 53.136 + 0.208 * positive affect.

Table 4. Ridge regression analysis of positive affect and Pa.

Name	Nonstandard coefficient		Standard coefficient	t	p	R^2	Adjusted R^2	F
	B	Standard error	Beta
Constant	53.136	3.601	-	14.755	0.000**	0.045	0.013	F (1,58) = 1.863, p = .035*
Positive affect	0.208	0.100	0.070	1.247	0.028*

The dependent variable is Pa. *p < 0.05, **p < 0.01.

Taking positive and negative affect as the independent variables and Pm degree as the dependent variable for ridge regression analysis, the K value was 0.990. Table 5 shows that the R squared value of the model was 0.074, which means that positive and negative affect could explain the 7.4% change in Pm. The model passed the F-test (F = 1.394, p = .024 < 0.05), which means that positive and negative affect influenced Pm. The regression coefficient of negative affect was −0.127 (t = – 1.024, p = .028 < 0.05), which means that negative affect influenced Pm. The regression coefficient of positive affect was 0.601 (t = 2.057, p = .031 < 0.05), which means that positive affect influenced Pm. The model formula is Pm = 52.265 + 0.601 * positive affect – 0.127 * negative affect.

Table 5. Ridge regression analysis of affect and Pm index.

Name	Nonstandard coefficient		Standard coefficient	t	p	R^2	Adjusted R^2	F
	B	Standard error	Beta
Constant	52.265	4.128	-	12.662	0.000**	0.074	0.011	F (2,57) = 1.394, p = .024*
Positive affect	0.601	0.098	0.104	2.057	0.031*
Negative affect	−0.127	0.124	−0.167	−1.024	0.028*

The dependent variable is Pm. *p < 0.05, **p < 0.01.

Discussion

This study evaluated the benefits of the natural forest environment to Chinese university students. Previous research on the effects of the natural environment on psychophysiology has tended to lack systematic and comprehensive analysis. To bridge this research gaps, this study used PANAS to analyze participants’ psychological changes, and these physiological indicator changes under psychological changes were measured by means of intended to investigate the effect of the natural forest environment on individuals.

The natural forest environment had a significant regulatory effect on some physiological indexes of the participants, such as by decreasing HR. Some scholars believe that being in the natural forest environment makes people feel relaxed, which reduced their HR(Hiroko et al., 2015; Roger et al., 1991). However, the natural forest environment may cause excitement and affect HR. In this study, most of the participants were visiting Zhangjiajie National Forest Park for the first time and they were energetic young people who were very interested in everything around them. This state of excitement sustained until the end of the trip in Phase 3. Maybe these explains why HR in Phase 2 was higher, but in Phase 3 was lower. Therefore, this study found that the role of the natural forest environment in reducing HR is not just being in the natural forest environment, but only after leaving the natural forest environment for a certain period of time, which is related to the participants’ visiting times to the experimental site and personality characteristics of participates.

Short-term HRV characteristics can reflect participants’ real-time mental stress. The results show that natural forest environment also significantly increased the HRV, which is consistent with the research results of scholars (Gokay et al., 2015; Peng et al., 2020). Therefore, it can be judged that the natural forest environment is conducive to reducing the pressure of university students and relaxing them.

Through the EEG data analysis results, the Pm of university students is significantly improved, which further shows that the natural forest environment does improve the relaxation of university students. The results also show that the natural forest environment could significantly improve Pa. According to the attention recovery theory (ART), directed attention is the key factor for human to improve work efficiency. Directed attention fatigue will produce severe psychological problems. Therefore, an effective way is needed to recover. The natural environment can restore directional attention by increasing human non directional attention, so as to reduce stress and improve learning or working efficiency(Berto, 2005; Kaplan, 1995). Because the participates’ non directional attention is supplemented in the natural environment, the directional attention is improved.

A major interest of scholars is the effect of natural environment on people from a single perspective of human psychology or physiology (Berto, 2005; Korpela et al., 2010; Liisa et al., 2014; Park et al., 2011; Takayama et al., 2014). Based on the relevant theories of environmental psychology, the effect of natural environment on people is all-round, not only reflected in one aspect of psychology or physiology, so it should be studied from the overall perspective of psychology and physiology. Moreover, the existing studies do not deeply explore the relationship between affect changes and physiological index changes, and lack a specific explanation of affect changes on physiological index changes. To connect them, this study found that there was a certain correlation between affect and physiological indexes. Therefore, it can be judged that psychological changes affect physiological changes to a certain extent.

For the research on the influence of natural forest environment on people, most scholars often focus on the influence of natural forest environment on people’s psychology and physiology(Kathleen, 2017; Mayer et al., 2009; Park et al., 2011; Ryan et al., 2010), and the research results only prove that natural forest environment is beneficial to people. In addition, some scholars compared the restorative effects of urban environment and natural forest environment on people(Hiroko et al., 2015; Mcmahan & Estes, 2015; Song et al., 2015a). But most of them do not consider forest vegetation types. This study verifies the effect of the subtropical evergreen and deciduous broad-leaved mixed forest on the psychological and physiological restoration of university students. This finding contributes to the in-depth study of the effects of different forest vegetation types on human psychology and physiology, and it is beneficial for the public to enjoy the benefits brought by forests, and for forestry managers and practitioners, it is also conducive to the sustainable management of forests. Additionally, most of the existing relevant studies are conducted from a certain perspective of human psychology or physiology. The relevant theories of environmental psychology and psychophysiology believe that the effect of the natural environment on human beings is all-round, not only reflected in one aspect of psychology or physiology. This research examines the effect of natural forest environment on human beings from the perspective of psychological and physiological integration, therefore to contribute to the subsequent relevant research.

Conclusion

This study employed environmental psychology methods to investigate the restorative effects of natural forest environment by means of an experiment and provided evidence for the psychological and physiological benefits of natural forest environment to university students. The natural forest environment helped to increase positive affect and reduce negative emotions. The results show that the natural forest environment has a significant regulatory effect on university students’ psychology and physiology. Although this study enriches the theory of environmental restoration in some degree, there nevertheless exist some limitations. Regarding experimental controls, although restrictions were implemented as far as possible (e.g., regulation of participants’ behavior, time selection considering temperature and humidity, wavelet de-noising of data), the results could have been affected by the participants’ subjectivity and by changes in field conditions. Future research can more strictly control the experimental conditions and increase the number of participants to obtain more accurate data. Measuring changes in physiological indicators is complex. Although this study attempted to verify the relationship between natural environments and human psychophysiology, whether the relationship is causal requires further study, along with the specific mechanisms of the relationship. This study examined the restorative effects of a natural forest environment but did not consider more specific landscape features, such as color, sound, or water flow, or specific microenvironmental features, such as air anion, plant essence, or atmospheric quality. Such characteristics can be considered in future work.

Acknowledgment

We thank LetPub (www.letpub.com) for its linguistic assistance during the preparation of this manuscript.

Disclosure statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability statement

Data will be made available upon request from the corresponding author.

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

This work was supported by the Forestry Science and Technology Popularization Project of National Forestry and Grassland Administration, Government of China [Grant No. 2018-R22]; Outstanding Youth Project of Education Department of Hunan Province, Government of China [Grant No. 19B582]; Ministry of Science and Technology of The People’s Republic of China [Grant No. 2019YFD1100404].