The Pathways and Efficacy of Horticultural Therapy in Promoting College Students' Mental Health: An Explanatory Sequential Mixed-methods Study

Summary

This article analyzes the use of horticultural therapy as a natural intervention to enhance the mental health of college students. The strategy, which uses a mixed-methods methodology and is based on an environment-cognition-emotion-behavior pathway model, improves life satisfaction, affect balance, and sense of life purpose.

Abstract

With increasing societal competition, mental health issues among college students have become increasingly prominent, while traditional mental health education often suffers from limited formats and effectiveness. This study investigates the mechanisms and effectiveness of horticultural therapy (HT), a nature-based intervention, in improving college students' mental health using a sequential mixed-methods explanatory design. In the quantitative phase, 112 college students were randomly assigned to an experimental group receiving HT intervention or a control group, with the experimental group participating in eight weekly HT sessions (90 min each) involving activities such as flower arrangement and herbal tea tasting. Intervention effects were assessed using standardized psychological scales and facial expression video analysis. In the qualitative phase, textual data reflecting participants' authentic experiences were collected and analyzed using grounded theory with NVivo 12. Results showed that, compared with the control group, the experimental group demonstrated significantly higher post-intervention scores in meaning in life (p = 0.017), overall affect index (p = 0.044), and life satisfaction (p = 0.046). Within the experimental group, post-intervention levels of meaning in life (p = 0.02), affect balance (p = 0.007), and positive affect (p = 0.019) were significantly higher than pre-intervention values. Facial expression analysis revealed notable differences in the distribution of seven expressions between groups, with positive expressions increasing over time, particularly during flower arranging and tea tasting activities. Qualitative analysis further identified a four-layer pathway model (Environment → Cognition → Emotion → Behavior), explaining 36%, 32%, 16%, and 15% of the variance, respectively. Overall, horticultural therapy effectively enhances college students' life satisfaction, sense of life meaning, and affect balance, thereby improving mental health through a multi-layered psychological pathway.

Introduction

China has experienced success in reforming its health sector. The population's health continues to progress toward higher levels. However, all of this continual process of changing with technology and industrialization, combined with the changing condition of the environment, and the current lifestyle of Chinese society, has created different issues related to subjective well-being and mental health (MH) of individuals. One of the most significant challenges being faced is that college students are facing enormous amounts of psychological stress, interpersonal conflict, and emotional strain due to the rapid speed of change in Chinese society¹. The speed at which these social changes are occurring, combined with the use of new digital technologies, online communication tools, and artificial intelligence (AI), has caused a high degree of emotional distress and overload for many young adults and as a result, many of them may not have the emotional support to cope with their challenges, leading to higher rates of stress and feeling of social isolation².

As stated in China's National MH Development Report, the issues that college students have to balance in their lives include the following factors: Overwhelming amounts of technology that college students use daily, the high academic expectations for the workload placed on college students today, and the loss of connection to people through social media³. As a result of these unique and complex factors, horticultural therapy offers an appropriate solution to these issues. Horticultural therapy provides a working environment for individuals to be engaged in meaningful, restorative, nature-based activities that create an opportunity for participants to develop psychological balance and a foundation for developing a holistic approach to their mental well-being. This study is designed to develop and assess a structured horticultural therapy intervention programme designed specifically for college students in China. This study also aims to promote the improvement of the individual's feeling of life meaning, level of emotional balance, and degree of subjective well-being⁴.

Following the COVID-19 pandemic, college students experienced heightened levels of anxiety, depression, and stress, which were further exacerbated by lockdowns and quarantine measures that reduced access to peer interaction and social support⁵. The disruption of social and environmental contexts caused by the COVID-19 pandemic has created an urgent demand for campus-based mental health intervention programs that provide experiential, supportive, and non-pharmaceutical support to students experiencing anxiety, depression, and stress associated with their mental health issues⁶. More broadly, as digital technology proliferates into all aspects of our lives, we have seen an increase in information overload (e.g., from e-mail, website links), cognitive fragmentation (e.g., through social media platforms), and virtual interactions (e.g., through Facebook). These changes are eroding our ability to develop a clear sense of self and the ability to derive meaning from our experiences. This is leading to increased disconnection between self-perception and reality, as well as an inability to create a cohesive sense of identity⁷. During periods of heavy reliance on artificial intelligence (AI), our reliance on AI has also contributed to weakened emotional bonds and decreased face-to-face interactions, both of which contribute to confusion about self-identity. In addition, the development of "virtual" identities through digital self-presentation has resulted in increased instances of identity confusion, identity crises, and nihilism⁸.

Monotony present in the physical world, lack of interaction with truly meaningful stimuli, and minimal attention to a variety of sensory inputs result in chronic boredom, and subsequently lead to a sense of lack of purpose in one's life⁹. Chronic boredom is a significant negative predictor of one's sense of purpose in life. Additionally, chronic boredom is a predictor of decreases in one's subjective well-being. Conversely, improved subjective well-being facilitates the creation of life's purpose, and increased subjective well-being can assist in developing intrinsic motivation to learn, which reduces disengagement and boredom at school¹⁰.

Based on these conditions, there are significant mental health issues with contemporary Chinese college students, specifically boredom, emotional emptiness, and being unable to find any meaning in life. When individuals are in these states of existence, they experience a decrease in their sense of life's meaning and a decrease in well-being, resulting in increased vulnerability to becoming mentally ill¹¹. Therefore, effective interventions designed to overcome these issues should include increasing sensitivity to one's environment, creating strong connections to the natural world, increasing one's connection to and interaction with their world, and enhancing their experiences through the addition of novel sensory experiences into their daily lives. All of these interventions will help create a greater sense of life's purpose and provide long-lasting mental health stability for students¹². A theoretical framework proposed in a previous study has outlined experiential mechanisms through which horticultural therapy operates, such as multi-sensory stimulation and enhanced emotional self-regulation; however, this framework has not been empirically validated using randomized controlled trials or longitudinal data, nor does it sufficiently address key psychological challenges faced by college students in China, including boredom proneness, diminished sense of meaning, weakened connection to nature, and social alienation. In this context, horticultural therapy (HT) is increasingly being recognized as a promising nature-based mental health intervention, warranting systematic empirical examination of its mechanisms and effectiveness within the Chinese college student population¹³. HT is defined by the American Horticultural Therapy Association (AHTA) as using plants and gardening to positively influence an individual's physical, mental, social, and cognitive functioning. Several studies have shown that spending time in nature can relieve attention fatigue, alleviate stress, and improve one's ability to manage emotions and behaviors. It has been theorized that being exposed to the less threatening aspects of nature results in increased physiological recovery from stress and improved mood due to Ulrich's stress recovery theory¹⁴. A meta-analysis conducted in 2024 of forty different horticultural therapy studies indicated statistically significant improvements in overall well-being and satisfaction with life, as well as that more frequent participation in gardening-related activities was associated with increased subjective well-being. However, there are limitations with the current body of literature. Most of the published research on horticultural therapy combines various forms of nature-based interventions (i.e., education in nature, walking in nature, etc.), making it difficult to determine the specific effects of horticultural therapy¹⁵. Additionally, many studies do not employ rigorous designs that provide evidence of causality. Finally, many studies do not build a generalizable body of evidence for horticultural therapy among Chinese college students because of cultural and contextual variations, including academic pressure, digital overload, and changes following the COVID-19 pandemic¹⁶. While the introduction of HT to university campuses in China is relatively new, ongoing exploratory research into HT has demonstrated that the currently available HT applications have been developed in a minimally standardized manner, making it extremely difficult to create generalizable models for the intervention protocols to be used to implement these applications in practice¹⁷. Based on this research, the purpose of this study will be to develop and validate a culturally relevant HT intervention model specifically designed for Chinese college students. The primary focus of the program will be on addressing: (a) existential meaninglessness, (b) emotional emptiness, and (c) chronic boredom through the provision of structured natural multisensory experiences. Along with aiding in symptom reduction, this intervention will help activate cognitive functions, support emotional regulation, and facilitate social connectedness by providing a culturally relevant, scalable, and accessible method for increasing psychological resilience and enhancing the overall mental health of young adults.

By employing a four-level Environment-Cognition-expression-Behavior pathway model to explain psychological mechanisms, integrating face expression recognition to objectively evaluate emotional reactions, and concentrating on Chinese college students, this study improves horticulture therapy (HT) research. It fills in holes in previous research by offering culturally appropriate insights into HT's impacts on life meaning, emotional balance, and general mental well-being through the combination of quantitative and qualitative methodologies.

Although horticultural therapy has been shown to provide overall benefits in reducing stress and improving emotional well-being, existing studies have rarely examined the specific psychological challenges faced by Chinese college students, such as chronic boredom, emotional emptiness, and a diminished sense of meaning in life, nor have they sufficiently explored which intervention approaches may be effective in addressing these issues¹⁰. Horticultural therapy frameworks include multisensory stimulation and emotional regulation, but there is limited evidence supporting these theories due to a lack of RCT and longitudinal data¹⁴. Studies on horticultural therapy primarily combine various types of nature-based interventions and do not differentiate the unique contributions of horticultural therapy, and the methodologies used in most of these studies are quite weak and do not establish the cause-and-effect relationship between the use of horticultural therapy and improved psychological functioning¹⁷. Cultural and contextual factors (e.g., the high levels of academic stress in China, over-dependence on technology, and the impact of COVID-19 on how college students see themselves and their futures) are often overlooked in these studies, and therefore, it is difficult to develop horticultural therapy interventions that effectively address the unique needs of Chinese college students. To design horticultural therapy protocols in China that can be replicated, standardized, and generalized, it is essential to conduct more rigorous research to determine the efficacy of horticultural therapy with respect to the unique characteristics of Chinese college students. To address the theoretical, empirical, and methodological voids identified, this research proposes the creation and validation of a structured, culturally appropriate horticultural therapy intervention model designed specifically for Chinese college students. The intervention model targets the three key indicators of absence of purposefulness, emotional desolation, and ongoing boredom. Through the implementation of various types of multisensory, expressive, and nature-based activities designed systematically, a Randomized Controlled Trial and Qualitative Thematic Analysis will be used to produce a solidly supported HT model to provide support to university mental health services and direct future studies in this area.

Protocol

This study involving human participants was reviewed and approved by the Ethics Committee of the University of Electronic Science and Technology of China. All procedures were conducted in accordance with the Declaration of Helsinki and relevant national guidelines. Written informed consent was obtained from all participants prior to their enrollment in the study. Participants were informed of the study objectives, procedures, potential risks, and their right to withdraw at any time without penalty. Personal information and video data were anonymized and used solely for research purposes.

NOTE: This study employed a sequential explanatory mixed-methods design, in which quantitative data were collected and analyzed first to assess intervention outcomes. Qualitative data were subsequently collected and analyzed to provide contextual interpretation and deeper understanding of the quantitative results.

1. Quantitative phase (data collection)

1. Recruit participants using convenience sampling from community centres. A quasi-experimental pretest-posttest design with partial random assignment was adopted to evaluate the impact of horticultural therapy. Ensure the participants are adults between the ages of 18-65 with no history of horticultural therapy.
2. Exclude individuals with severe physical limitations or those currently receiving psychological treatment. Randomly assign the eligible participants to either the intervention group or the waitlist control. Ensure that the participants in the control group are allowed to do free activities throughout the eight weeks, while the participants in the intervention group meet weekly for 90 min sessions for eight weeks. Use the demographic and outcome data at baseline to verify group equivalency. Evaluate the within-group changes using paired sample t-tests.
3. Before conducting the paired-sample t-tests, examine the normality of the pre-post difference scores and confirm that they are approximately normal, supporting the validity of the parametric analyses.
4. Evaluate the treatment group's pre-post changes using a paired t-test. Identify the key experiences and psychological processes that underlie participants' reactions to the HT intervention through the transcription, coding, and thematic analysis of qualitative discussions¹⁸.
5. Recruit the participants using convenience sampling, screen them for eligibility criteria, and randomly assign them to treatment groups.
   NOTE: Since baseline data are collected before eligibility screening, no psychological measurements are provided as the participants have not been assigned to treatment groups yet, and therefore will not have a validated measurement of their mental health or psychological status. Normality of the data is checked using difference score histograms and paired t-tests to compare pre-post treatment for statistical significance.

2. Horticultural therapy intervention protocol

1. Initial intervention: Before commencing hands-on horticultural activities, ensure that all participants receive an introduction session explaining the expectations for participation, the structure of the activities, safety standards, and therapeutic objectives.
2. Seat participants in a controlled waiting area for 45 min and provide neutral reading materials to equalize pre-stress conditions. This helps stabilize emotional states and improves between-session comparability.
3. Administer the State-Trait Anxiety Inventory and related scales to establish baseline emotional state¹⁹.
   NOTE: The State-Trait Anxiety Inventory and related scales are used in accordance with established procedures to reliably obtain a measure of participants' initial emotional states²⁰.
4. Expose participants to horticultural therapy activities, including flower arranging, hydroponic plant care, and plant-based self-portrait creation. Structured horticultural activities with a step-by-step guide from a trained facilitator will be used²¹.
5. Provide guided instructions and demonstrations, and direct participants to explore plant colors, textures, shapes, and scents. Ensure that the facilitators assist the participants in focusing on the color, texture, and scent of plants, and encourage them to engage in multisensory exploration for emotional engagement and aware behavior.

3. Adjustment/post-reintubation phase

1. Adjust subsequent sessions according to participants' engagement and performance. Provide paired work, brief tasks, or additional instruction to participants requiring support. Facilitators can alter tasks based on the length of a specific session. In instances where there is difficulty displayed from a participant, facilitators may team up participants with another participant or provide simplified instructions or other types of support.
2. The activities were developed in complexity that prompted participants to reflect on the symbolic meaning of plants to express themselves, and the meaning of their lives, e.g., the future goals represented by fruit and the emotional states represented by colors of leaves. Participants take part in symbolic meaning-related activities by selecting colours, textures, or fruits from plants that will symbolize their aspirations, values, and hopes for the future.
3. Continue horticultural therapy exercises, including herbal tea preparation and plant-based meditation, to strengthen emotional regulation and mindfulness. Facilitators ensure that participants engage in ongoing mindfulness activities to help individuals manage their emotions more effectively across many sessions. Activities include how to prepare herbal tea and engage in plant-based meditation.

4. Qualitative phase (follow-up exploration)

1. Collect written reflections from participants after the intervention to document their experiences and interpretations. Analyze the textual materials qualitatively to identify pathways through which horticultural therapy influences MH.
2. The qualitative findings were then used to elucidate the underlying mechanisms of the intervention effects.
3. The participants record their experiences and provide a written reflection that discusses their experiences. The written reflections are coded using content analysis to examine how HT has impacted mental health.

5. Diagnosis and plan

1. Participants were recruited from a university campus to serve as the study population. Eligibility criteria required that members be in good physical condition, free of soil or pollen allergies, and cognitively normal. Before the start of the trial, all contributors provided informed consent.
2. Exclude participants with significant physical illness, psychiatric disorders, use of psychiatric medication, or exposure to major negative life events in the previous three months. There were 112 participants in all, 58 in the group doing the experiment and 54 in the control group, to account for any absence from the eight horticultural therapy sessions. Table 1 provides a summary of each group's demographic attributes.
3. Screen the participants for eligibility, ensure that they sign the informed consent, and complete the baseline assessments. Place them into groups and check the demographic data to confirm group similarities.
4. Determine the sample size suitability using G*Power software. Use independent t-tests to compare mean differences between groups, with implication set at 0.05, an effect size of 0.5, and a desired statistical power of 0.8.
   NOTE: Controls indicated that a minimum of 34 participants was required, supporting the validity and reliability of the study design.
5. Participants reported diverse personal, family, and social backgrounds, with the majority living on campus and spending limited time with their families.
   NOTE: Some participants reported prior experiences of mild stress or sleep disturbances, but no major medical conditions were present. Psychological symptoms such as boredom, emotional emptiness, low life satisfaction, and a perceived lack of meaning in life were noted. Participants also reported limited effectiveness of previous coping approaches, including short-term counseling, self-guided activities, and meditation. Baseline physical examinations indicated normal vital signs and no abnormalities, confirming that the primary focus of the intervention was on mental health-related concerns.

6. Diagnosis, assessment, and plan

1. Administer standardized psychological scales to assess life satisfaction, affect balance, and sense of meaning in life. Perform facial expression analysis and conduct qualitative interviews to corroborate symptoms such as emotional emptiness, boredom, and low positive affect.
2. A formal clinical diagnosis is not required; instead, consider potential stress-related mood disturbances through discrepancy evaluation. Expose participants in the intervention group to an 8-week horticultural therapy (HT) program consisting of multisensory, nature-based activities designed to enhance emotional regulation, restore attention, and reinforce meaning in life. Monitor adverse effects, noting that any side effects are typically minimal and may include mild fatigue or occasional frustration related to physically demanding tasks.
3. Use Validated psychological instruments to measure emotional states. Analyze facial expression data and qualitative interview materials to confirm emotional symptoms. Deliver the HT intervention continuously over an eight-week period to promote participant well-being
4. After enrolling participants into experimental and control groups, conduct statistical analyses on psychological scale data collected before and after the intervention. Analyze video recordings obtained during on-site HT activities. Upon completion of the intervention, collect written textual materials reflecting participants' subjective experiences and analyze them qualitatively.
5. The psychological measurement instruments included the following: Video recording was done consistently; individual frames were systematically retrieved; facial recognition was performed using Haar Cascades; emotions were recognized based on the output from the VGG19 algorithm trained on the FER2013 dataset²².
6. The Chinese-Meaning in Life Questionnaire (C-MLQ), originally developed by Steger and adapted into Chinese by Liu Sisi and Gan Yiqun²³, comprises nine items divided into two subscales: "Experience of Meaning" and "Search for Meaning." Rate the responses on a seven-point Likert scale (1 = strongly disagree, 7 = strongly agree), with higher scores indicating a stronger sense of life purpose.
7. The Index of Well-Being (IWB), developed by Campbell in 1976, evaluates subjective happiness²⁴. The scale includes nine items, with the first eight items forming an emotional index and the last item assessing overall life satisfaction. Higher total scores indicate greater subjective well-being.
8. Affect Scales, which assessed Positive Affect, Negative Affect, and Affect Balance in the general population²⁵. Analyze the positive and negative effects separately, and calculate the Affect Balance by combining the two scores.
9. Satisfaction with Life Scale (SWLS), which consists of five items scored on a seven-point Likert scale (1 being strongly disagree and 7 being strongly agree)²⁶. Higher scores indicate a higher level of overall life satisfaction.
10. Record video data using a fixed tripod and mobile device at a consistent angle.
    NOTE: Due to equipment limitations, video recording began throughout the fourth week of the program.
11. Record and analyze the participants' facial languages to quantify the frequency of optimistic and negative expressions. Begin each recording 5 min after the official start of the activity and extract 6 min of footage for analysis. Extract surrounds at a rate of 1 frame per 10 frames, with an average of over 1,000 frames processed per video.
12. Perform facial emotion analysis in two steps: detect faces using the Haar Cascade classifier and classify expressions using the VGG19 CNN model. The pre-trained Haar Cascade Frontal Face Default classifier in OpenCV was used to recognize faces. The facial areas were then clipped for examination. Facial expression recognition employed a convolutional neural network (CNN) model, VGG19, which utilized multiple small convolutional kernels to extract localized features from input images. The model was trained on the FER2013 dataset, containing over 30,000 RGB images labeled with seven primary emotion categories: 0 = anger, 1 = disgust, 2 = fear, 3 = happiness, 4 = sadness, 5 = surprise, and 6 = neutral.
13. The "disgust" category contains the fewest images (approximately 600), while other categories contain roughly 5,000 each. The trained model analyzes the cropped facial images, counts occurrences of each emotion, and determines the frequency of each facial expression.
14. After the 8 weeks of the HT intervention program, ask the members of the group to reflect on their experiences and summarize their impressions at the end of the intervention in writing. Photos feature the authentic emotions and experiences of the participants throughout the horticultural events, thoughts, and recollections of the events, and any perceived effects or changes that the intervention brought about. Participants have an opportunity to write about their experiences, feelings, and perceiving changes; these reflections are called "unrestricted reflections". Once completed, collect these together, clean, and prepare for coding purposes.
    NOTE: The style and length were not limited, and all that was needed was creativity and no plagiarism. The number of reflection documents gathered by all the participants in the intervention group was 58, which produced 64,895 words following the elimination of irrelevant content.
15. Analyze the textual data using procedural grounded theory in NVivo 12. Import the unrestricted reflections into NVivo 12 for coding, and then systematically organize the codes into a hierarchical structure using grounded theory approaches

7. Conducting open, axial, and selective coding to develop a pathway-based theoretical model

1. Open Coding: Examine each reflection document line by line. Ensure that the researchers maintain a neutral stance, avoiding preconceptions, and conceptualize the text using psychological terminology. Identify, categorize, and relabel the reference points iteratively until no new primary codes emerge.
   NOTE: Axial Coding, Selective Coding. Open coding is the first step in coding data; this process involves reviewing the data line-by-line and identifying concepts associated with each line.
2. Axial Coding: Synthesize the primary codes and compare them to identify patterns and relationships. Codes were grouped into broader categories to form core conceptual linkages. Primary codes are grouped, compared, and linked to identify relationships forming broader thematic categories.
3. Selective Coding Further refine the major categories derived from axial coding. Examine the underlying logical relationships to develop a pathway-based theoretical framework. Integrate and refine the core categories to construct a theoretical pathway explaining HT's psychological mechanisms.
4. This structured approach ensured rigorous and systematic analysis of participants' subjective experiences, contributing to a deeper, thoughtful understanding of the psychological effects of horticultural therapy.

8. Horticultural therapy-based intervention program design

1. Design the horticultural therapy program using horticultural activities as the primary medium and integrating expressive art therapy principles. Ideas of expressive art therapy were taken into consideration, and the horticultural activities were combined with the possibility of being self-expressive, creative, and personally incorporated. The concept of combining horticultural tasks and expressive art principles provides participants with an opportunity to experience planned emotional outcomes through creative expressions in plant materials.
2. Frame activities thematically to support exploration of life meaning, self-discovery, and appreciation of life. The HT activity leader was experienced. All the sessions were held indoors and at a specific time, and both the experimental and control groups were split into seven smaller groups with about eight participants per group. Table 2 shows the detailed intervention plan of the HT program. Sessions follow life-meaning themes; participants work in small groups under a trained facilitator following a standardized session plan.

Results

Statistical results of psychological measurement scales

After the participants were split up into groups, the experimental and control groups' pre-test results were compared using an independent t-test. Table 3 summarizes the analysis, which showed no discernible differences among the 10 groups. The control group and experimental group did not differ significantly in psychological measures at the baseline (Table 3). Group equivalence was confi...

Discussion

The analysis highlights the advantages of nature-based therapies for university students who are at risk, but it is constrained by significant variation in intervention designs, small sample sizes, and inconsistent measurement methods among research. Conclusions on long-term effects on well-being are limited by the absence of longitudinal data, which also restricts generalizability to larger young populations¹⁶. The evaluation of the e-Nature model is constrained by its use of self-reported result...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Hydroponic Plant Kit	GreenJoy Co., Ltd.	HJ-2023	Used for plant care and observation in horticultural therapy sessions
NVivo 12.0 Software	QSR International Ltd.	NV12-STD	Used for qualitative data analysis (grounded theory coding)
OpenCV + VGG19 Model	Open Source	–	Used for facial-expression recognition and video analysis