Short-term responses across two sequential plant-based activity sessions: an exploratory fixed-order pilot study

Recent investigations into community wellness programs have increasingly highlighted the psychological and physiological benefits of structured, nature-based interventions. A pilot study conducted by researchers at National Taiwan University examined the short-term impacts of hands-on plant-based activities on community-dwelling adults. The research team, led by corresponding author Yu-Sen Chang alongside colleagues L-l Chang, Y-H Ku, and C-Y Chuang, sought to characterize human responses during two sequential workshop formats. While the findings point toward notable short-term reductions in stress and anxiety, the study’s authors emphasize that the exploratory nature of the trial—compounded by a fixed session order and the absence of a control group—requires cautious interpretation before broader causal claims can be established.

Background and Theoretical Context

The integration of accessible, plant-based group activities into community, educational, and healthcare settings has grown significantly over the last decade. Existing literature surrounding horticulture therapy frequently points to measurable improvements in depressive symptoms, chronic stress, and overall well-being. These positive outcomes are theoretically anchored in foundational psychological frameworks such as Stress Recovery Theory and Attention Restoration Theory. These frameworks posit that direct engagement with natural elements, combined with purposeful manual labor and social interaction, facilitates restorative psychological responses by gently engaging human attention and mitigating cognitive fatigue.

Despite the proliferation of garden-based programs, horticultural interventions are far from uniform. Programs typically vary widely in duration, setting, structural complexity, and specific craft components—ranging from outdoor container gardening to indoor flower arrangements and leaf-crafting. While meta-analyses and previous comparative trials have occasionally contrasted active plant care with passive exposure or evaluated varying session lengths, direct comparisons between fully specified, multicomponent, hands-on workshop packages in community adults remain scarce. Furthermore, isolating specific variables such as aroma, manual dexterity exercises, and facilitator contact continues to challenge researchers in the field.

Chronology and Study Design

The single-group, fixed-order pilot study was structured across two distinct weekend sessions held one week apart in October 2024 at community and adult-education facilities in Taiwan.

The recruitment phase successfully enrolled twenty-five adult participants, consisting of 6 men and 19 women with a mean age of 46.9 years. Notably, roughly 68 percent of the enrolled cohort reported a prior background in agriculture or horticulture. Eligibility criteria required participants to be at least 20 years old, capable of communicating in Chinese, in stable health, non-pregnant or lactating, and free of known adverse reactions to tea, essential oils, or horticultural materials.

The intervention timeline followed a rigid, non-counterbalanced sequence:

• Session 1 (October 20, 2024): Focused primarily on horticultural crafting. The 75-minute session included a 10-minute introduction to plant materials, a 10-minute instructional demonstration, 40 minutes of hands-on potted-arrangement work using foliage plants, and a 15-minute group sharing period. Baseline and post-session psychological questionnaires alongside five-minute seated physiological recordings were administered.
• Session 2 (October 27, 2024): Followed a week later with a 75-minute multi-phase session. This second meeting combined a 35-minute balm-crafting workshop with a subsequent 40-minute potted-plant arrangement exercise. The balm formulation incorporated distilled water, a blend of carrier oils (olive, grapeseed, shea-derived, and rosehip), dissolved menthol, and specific essential oils (wintergreen, lavender, tea tree, and eucalyptus) at an approximate 1.6 percent total concentration. Mid-session assessments followed balm preparation, while final post-assessments succeeded the secondary horticultural activity.

Out of the original 25 participants, 23 attended both sessions, though effective sample sizes for specific analyses varied due to occasional questionnaire incompleteness or physiological signal quality limitations.

Quantitative Findings and Psychometric Data

The primary outcomes of the investigation focused on standardized psychological inventories: the 21-item Depression Anxiety Stress Scales (DASS-21) and the 20-item State-Trait Anxiety Inventory-State form (STAI-S).

Both sessions yielded statistically significant short-term psychological improvements. For Session 1, mean DASS-21 scores dropped by 8.57 points [95% CI (5.41, 11.72), $n=23$], while STAI-S state anxiety scores decreased by 7.30 points [95% CI (5.23, 9.37), $n=25$]. In Session 2, participants experienced a mean DASS-21 reduction of 5.53 points [95% CI (2.65, 8.41), $n=17$] and a STAI-S reduction of 6.86 points [95% CI (4.07, 9.66), $n=22$]. All four primary Holm-adjusted $p$-values reached statistical significance at levels below 0.001.

Exploratory evaluations using a seven-item horticultural wellbeing questionnaire similarly tracked upward trajectories, reflecting enhanced perceptions of personal accomplishment, relaxation, and horticultural cultivation interest across both gatherings. Furthermore, exploratory heart-rate variability (HRV) metrics—specifically SDNN and RMSSD parameters—exhibited increases across the sessions, pointing toward potential autonomic shifts during the workshop periods.

However, complete-case direct comparisons between Session 1 and Session 2 improvement scores revealed no statistically detectable differences for DASS-21 change (mean difference: 3.12 points, 95% CI [-1.51, 7.75]) or STAI-S change (mean difference: 0.34 points, 95% CI [-3.15, 3.83]). To probe potential equivalence between the two formats, two-one-sided tests (TOST) were applied. Using an illustrative standardized bound of $d_z = pm 0.50$, the equivalence criterion was met exclusively by the STAI-S outcome, and this result failed to hold under narrower margins such as $pm 0.30$ or $pm 0.20$.

Methodological Limitations and Author Caveats

In their comprehensive reporting, the research team was explicit regarding the structural limitations of the pilot design. Because every participant experienced the exact same fixed sequence—attending the horticultural crafting session first and the combined balm-crafting and horticultural session second—activity packages were entirely confounded with visit order, familiarity, expectancy effects, and natural day-to-day variations.

Furthermore, the absence of an untreated, wait-list, or active control group prevents the attribution of observed psychological improvements strictly to the plant-based activities. Social interaction, supportive guidance from facilitators, and the routine passage of time remain alternative explanations for the psychological gains. Physiological data collection also faced technical constraints, including single-channel forehead electroencephalography (EEG) with incomplete processing documentation and short HRV recordings that lacked full environmental and respiratory controls.

Phase-specific analyses within Session 2 revealed that numerical improvements in psychological scores were more pronounced following the initial balm-crafting segment than the subsequent plant arrangement phase. The authors caution against misinterpreting this pattern as evidence of superiority for balm crafting, noting that participants had already undergone initial stress reductions before entering the second phase, potentially leaving less room for further measurable changes.

Implications for Future Research

The investigation functions primarily as a descriptive pilot designed to generate hypotheses and supply precise variance estimates for future, rigorously controlled randomized trials. The authors stress that future research in environmental psychology and horticulture therapy must incorporate counterbalanced session sequences, prospective sample-size calculations based on predefined minimal clinically important differences, and appropriate control conditions to isolate active therapeutic ingredients from general social and contextual confounders.

The study protocol received formal ethical approval from the National Taiwan University Research Ethics Committee for Behavioral and Social Sciences (NTU-REC No. 202304HM014), with all participants providing written informed consent prior to enrollment. The complete dataset, data dictionaries, and analytical code remain available from the corresponding author upon reasonable request under institutional guidelines.