Psychophysiological responses to tactile contact with hairy versus hairless plant leaves

Abstract

Tactile contact with ornamental plants promotes human health and healing. However, the impact of touching plant leaves on emotional and physiological responses requires further investigation. This study examined how leaf surface texture, hairy versus hairless, influences human psychophysiological responses in 30 Chinese female participants, using these two leaf types as stimuli. Subjective perception was measured using the Semantic Differential (SD) method, and physiological responses were assessed through electroencephalogram (EEG), heart rate variability (HRV), and electrodermal activity (EDA). Results showed that hairless leaves significantly increased α wave values in the C4 channel, indicating more pleasant and relaxed emotional responses, whereas hairy leaves significantly increased γ wave values, suggesting enhanced attention and exploration desire. Hairy leaves significantly decreased HRV indicators (pNN50, pNN20) compared with hairless leaves, indicating reduced parasympathetic activity and heightened emotional arousal; conversely, touching hairless leaves was associated with stronger parasympathetic activity and more relaxed emotional states. Additionally, EDA was significantly elevated when touching hairy leaves, indicating stronger emotional arousal. Therefore, hairless leaves demonstrate better performance in tactile comfort and emotional regulation. These findings provide theoretical and practical guidance for tactile design of plant landscapes and plant selection in sensory gardens for the blind.

Introduction

The current built environment in daily life may impose various forms of mental stress on human health. Green plants play an important role in the health of the human living environment, exerting positive effects on human physiology, psychology, and spirit through carriers such as their form, color, fragrance, sound, and texture^1,2. As an indispensable and effective method in complementary and alternative medicine (CAM), horticultural therapy is often used in areas such as convalescent gardens, sensory gardens, and gardens for the blind³. The healing pathways of plants can be achieved through the five senses: vision, hearing, smell, taste, and touch, and their impacts on human physiological and psychological health have been widely documented^3–8. Although the sense of touch plays a crucial role in transmitting important information about the surrounding environment, its significance has rarely received attention⁹.

Women generally bear higher psychological stress and are more prone to mental health issues such as anxiety¹⁰, and the prevalence of anxiety disorders in women is significantly higher than that in men¹¹. This difference occurs not only in adults but also in children and adolescents. From an evolutionary perspective, women have historically assumed primary caregiving roles within families, which may contribute to heightened sensitivity to environmental stimuli and emotional responses¹². Such gender differences in stress vulnerability appear to be rooted in both biological factors and social role expectations^13,14. National epidemiological survey data show that the prevalence of common mental disorders in Chinese women is slightly higher than that in men¹⁵. These phenomena highlight the obvious gender differences in mental health challenges and emphasize the importance of incorporating a gender perspective in interventions and research.

The skin serves as a crucial sensory organ for perceiving the physical properties of the external world through direct contact¹⁶, and is sometimes referred to as the “third brain“¹⁷, playing a crucial role in human emotions. There is a close relationship between tactile sensation and emotions¹⁸, and skin contact with different surfaces can evoke sensory and emotional responses^19,20, including positive and negative emotions, influencing touch preference, health, and satisfaction²¹. Touching wood gives people a sense of relaxation, while touching metal elicits a stress response^22,23. Japanese cypress and artificial boards have different effects on blood pressure²⁴. These studies illustrate the tactile sensations and emotional changes associated with different materials.

Relevant research on plant tactile perception remains relatively scarce. Touching grasses of different heights induces more positive physiological responses compared to artificial materials²⁵. The tactile experience of Japanese cedar trunks with varying thicknesses evokes feelings of comfort and relaxation compared to artificial materials²⁶. The psychological and physiological impacts of human interaction with plant leaves, assessed using the semantic differential method and brain activity metrics, indicate that natural stimuli elicit a sense of physiological calmness²⁷. Tactile interaction with natural horticultural elements, including geraniums, solid hardwood, and moss, promotes physiological stability and reduces stress by calming prefrontal cortex activity²⁸. The elderly experience feelings of naturalness, happiness, relaxation, and comfort when engaging with plant materials compared to artificial materials^29,30. Tactile interaction with natural and artificial grass among Chinese women shows that touching real grass elicits increased alpha and gamma brain wave activities, as well as decreased systolic and diastolic blood pressure, whereas artificial grass does not produce similar beneficial responses³¹. These findings collectively suggest that tactile contact with natural plant materials can improve psychological and physiological states.

Electroencephalography (EEG), a precise and non-invasive method, is extensively employed to investigate the effects of sensory stimuli on brain activity and diagnose stress, anxiety, and other mental states³². The occurrence of alpha waves (8–13 Hz) is closely associated with a relaxed, awake state, and has been linked to the positive regulation of emotions and emotional stability; increased alpha wave activity coincides with pleasant and calm states, facilitating relaxation^33,34. Gamma waves (30–100 Hz) are typically associated with higher cognitive functions, attention concentration, and emotional arousal; elevated gamma activity is observed during states of focused attention and in response to stimulating situations^35–38. Heart rate variability (HRV), the variation in successive R-R intervals of sinus heartbeats, reflects the dynamic interplay between the sympathetic and parasympathetic branches of the autonomic nervous system at the sinoatrial node, serving as a non-invasive marker of autonomic regulation³⁹. Emotional responses such as discomfort, anxiety, or alertness activate the sympathetic nervous system, reducing HRV, whereas relaxed emotions enhance parasympathetic activity, increasing HRV^40,41. Electrodermal activity (EDA) reflects changes in the electrical conductance of the skin arising from eccrine sweat gland activity, which is exclusively regulated by the sympathetic nervous system; heightened EDA signifies tension or stress, whereas diminished EDA suggests a relaxed state^42–46.

Previous studies have primarily relied on subjective evaluation methods such as questionnaires or used single physiological parameters, lacking comprehensive evaluation of objective physiological indicators such as EEG and HRV. It remains difficult to comprehensively quantify the effects of different plant tactile sensations on emotions. To address this limitation, the present study examined human responses to the leaves of two types of ornamental plants with distinct surface characteristics—hairless and hairy textures.

From an evolutionary perspective, plant trichomes (leaf hairs) serve as protective structures against herbivores, functioning as the first line of defense by creating physical barriers and sometimes secreting chemical deterrents^47,48. Given this defensive function, humans may perceive hairy leaves differently from hairless leaves, potentially triggering heightened alertness responses to the unfamiliar or protective surface texture. Based on this theoretical framework, the following hypotheses were proposed:

H1a: Touching hairy leaves induces increased gamma wave activity in the somatosensory cortex compared to hairless leaves.

H1b: Touching hairy leaves induces elevated EDA compared to hairless leaves.

H1c: Touching hairy leaves induces reduced HRV indices (pNN50, pNN20) compared to hairless leaves.

H2: Touching hairless leaves induces enhanced alpha wave activity in the somatosensory cortex.

H3: Subjective evaluations reveal that hairless leaves are perceived as more calming, stable, and friendly, whereas hairy leaves are perceived as more stimulating and special.

By integrating physiological measures (EEG, EDA, HRV) and subjective reports, the investigation systematically analyzed the effects of plant leaf texture on human tactile experience. The research results will provide a scientific basis for plant selection and application in the fields of landscape architecture design, multisensory gardens, gardens for the blind, and horticultural therapy.

Methods

Stimuli and subjects

This study examines the leaves of two ornamental plants with distinct tactile features: the hairless Peperomia tetraphylla and the hairy Pelargonium hortorum. Whole potted plants were purchased from Zhengdong Flower Market, Zhengdong New District, Zhengzhou, China, and the species were identified by Xudong Wang from the School of Human Settlements, North China University of Water Resources and Electric Power. The plants were maintained under laboratory conditions (temperature 28 ± 1 °C, relative humidity 50 ± 5%) until leaf collection. Peperomia tetraphylla is a succulent perennial herb characterized by thick, fleshy leaves⁴⁹(approximately 1.3–1.9 mm in thickness) with a prominent waxy cuticle layer, offering a smooth, glossy, and glabrous tactile experience. The leaves of Pelargonium hortorum are pubescent, bearing both glandular and non-glandular trichomes on adaxial and abaxial surfaces⁵⁰. The non-glandular trichomes are typically 340–500 μm in length, creating a soft, fuzzy texture upon contact. These two species were selected to represent two commonly encountered tactile categories in ornamental plants: smooth/glabrous versus pubescent/hairy leaf surfaces.

Mature, healthy leaves were collected from the middle portion of each plant fifteen minutes before each trial. To ensure experimental accuracy and minimize interference from factors such as leaf edge, thickness, hardness, and texture on tactile perception, plant leaves were pressed into a 6 cm × 12 cm rectangular cardboard frame with a 3 cm × 4 cm central window after harvesting, creating a planar sample (Fig. 1)²⁶. This ensured that subjects only interacted with the designated area. Between trials, samples were inspected for quality and any damaged or significantly dried leaves were promptly replaced with fresh ones to maintain consistent tactile stimuli.

Fig. 1.

Flow chart of the experiment.

A priori power analysis was conducted using G*Power 3.1.9.7, setting the significance level at α = 0.05, the statistical power (1 - β) at 0.80, and the effect size d at 0.50, with a two-tailed test employed. The calculations indicated that the minimum sample size requirement before the experiment was 27 subjects to ensure the required power at the expected effect level. The study involved 30 female subjects (mean age: 41.73 ± 14.27 years, age range: 20–65 years). Prior to the experiment, subjects underwent health screenings. They abstained from alcohol, smoking, and neuroactive drugs for 24 h before the experiment, refrained from strenuous exercise within 2 h prior, and were in good mental health. The experiments were conducted in a controlled laboratory on the university campus, maintaining a temperature of 28 ± 1 °C and relative humidity of 50 ± 5%. Light intensity was set between 50 and 100 lx to minimize visual interference, and background noise was kept at ≤ 40 dB to limit external stimuli. All experimental operations followed ethical requirements. This study has been approved by the scientific research ethics review of the affiliated institution (Approval No.: NCWU-SA-2023-002, Approval Date: November 15, 2023). All subjects were voluntary.

Methods and procedures

Physiological feedback

EEG data closely correlate with physiological and psychological conditions, such as human emotions, cognitive conformity, and stress responses. The brain regions related to tactile perception mainly include the Postcentral Gyrus (PoCG), Inferior Parietal Lobule (IPL), Supramarginal Gyrus (SMG), Precentral Gyrus (PrCG), and Superior Temporal Gyrus (STG). The functional areas are located in the Primary Somatosensory Cortex (SI), Secondary Somatosensory Cortex (SII), Primary Motor Cortex (MI) and Secondary Motor Cortex (MII). Previous findings indicate that under tactile stimulation, the maximum activation intensity was concentrated in the contralateral primary somatosensory cortex (PoCG and SI)^51–53. Two electrodes, C3 and C4, located in the primary somatosensory cortex and directly related to tactile perception, were selected for the experiment⁵⁴. EEG data were recorded using a 16-channel portable wet-electrode device with saline solution as the conductive medium to ensure stable electrode-skin contact, at a single-channel sampling rate of 1024 Hz. Electrode impedances were verified to be below 10 kΩ before each recording session. EEG signals were bandpass filtered (0.5–100 Hz) to remove baseline drift and high-frequency noise, and a 50 Hz notch filter was applied to eliminate power line interference. Ocular artifacts were removed using independent component analysis (ICA), and ECG artifacts were rejected, followed by average re-referencing within the ErgoLAB 3.0 platform. Epochs containing artifacts exceeding ± 100 µV were identified and rejected prior to analysis, resulting in exclusion of approximately 8% of epochs across all participants. Signal quality was further inspected visually to minimize residual EMG contamination in the gamma band. Alpha (8–13 Hz) and gamma (30–100 Hz) power spectral densities were extracted using Fast Fourier Transform (FFT). Alpha activity was selected as an indicator of relaxed wakefulness^33,34, while gamma activity was included to assess attentional engagement and cognitive processing^36,37. Beta-band activity (13–30 Hz) was not analyzed in this study, as beta oscillations overlap with the sensorimotor mu rhythm during active tactile exploration, making it difficult to dissociate perceptual from motor contributions with the current electrode montage.

HRV provides a non-invasive reflection of activity and balance of the sympathetic and parasympathetic nervous systems³⁹. Sex differences in HRV reactivity have been documented, with women displaying heightened autonomic sensitivity to emotional and tactile stimuli^55–57. Women exhibit heightened sensitivity to the emotional responses elicited by tactile stimuli, a phenomenon linked to estrogen-enhanced limbic-autonomic connections. Such stimuli can induce mild stress or emotional arousal, with HRV effectively capturing these subtle autonomic responses. HRV provides a precise reflection of the autonomic nervous system’s regulatory state when interacting with various plants, particularly in balancing relaxation and tension, as well as stress and calmness⁴⁰. In this study, the R-R interval was recorded using the ErgoLAB wearable chest-strap sensor. R-R intervals were extracted and artifacts were corrected through interpolation for ectopic beats. The short-term HRV indicators, pNN50 (percentage of successive R-R intervals differing by more than 50 ms) and pNN20 (percentage differing by more than 20 ms)^58–61, were employed to assess subjects’ emotional and stress responses to tactile stimulation by different plant leaves. Both indices were included because pNN50 reflects vagal modulation of cardiac activity, while pNN20 provides greater sensitivity to subtle parasympathetic changes in brief recording periods⁶². Ultra-short-term HRV analysis has been validated for time-domain indices including pNN50 in stress monitoring contexts⁶³.

EDA was assessed using the ErgoLAB EDA wireless sensor with wet Ag/AgCl electrodes and saline gel as the conductive medium to track fluctuations in electrodermal signals, indicative of sympathetic nerve activation and emotional responses. Disposable electrodes were replaced for each participant to ensure hygiene and avoid cross-subject interference. Following electrode placement, a 5-minute stabilization period was observed before recording commenced, allowing the electrode-skin interface to reach a stable impedance state. This method captures swift physiological changes over brief intervals. As a non-invasive, real-time monitoring method, EDA offers immediate feedback without disrupting subjects, making it ideal for analyzing physiological responses in short-term experiments. In this experiment, Phasic Skin Conductance Response (SCR) was chosen as the analysis parameter for EDA. SCR effectively captures short-term electrodermal fluctuations and the sympathetic nervous system activation triggered by stimuli, enabling researchers to rapidly assess the impact of various stimuli on subjects⁶⁴. The phasic component was separated from the tonic component using continuous decomposition analysis.

The ErgoLAB 3.0 human-machine-environment synchronization platform (Kingfar International Inc.) was utilized to collect multi-modal physiological data in this study. This platform has been extensively employed in human factors research on environmental perception, particularly in the fields of landscape architecture^65,66. EEG, EDA, and HRV modules were selected to monitor physiological indicators of subjects. Data segments with poor signal quality were identified through visual inspection and automatic artifact detection, then denoised and corrected using the ErgoLAB platform before analysis.

Subjective perception

The subjective perception measure employs the Semantic Differential (SD) method to gather psychological data from subjects. Based on Koga et al.‘s²⁶ original 15-item 5-point SD scale, the present study adapted it to a 7-dimension 7-point scale to reduce participant burden while focusing on the most relevant emotional and tactile dimensions for plant perception. The SD scale includes seven pairs of adjectives across seven dimensions: “ordinary - special,” “soft - hard,” “like - dislike,” “pleasant - unpleasant,” “friendly - unfriendly,” “stimulating - stable,” and “anxious - calm,” each rated on a bipolar 7-point scale. A higher score indicates a more positive restorative perception of the tactile sensation of plants, while a lower score suggests the opposite²⁵. The scale’s reliability is confirmed through Cronbach’s α coefficient, ensuring internal consistency.

Procedure

The subjects were thoroughly briefed on the purpose, procedures, and precautions of the experiment in detail, including standardized verbal instructions on the tactile procedure such as finger placement. After the subjects understood and signed the “Informed Consent Form for the Experiment”, participants donned the physiological monitoring equipment and assumed a seated, eyes-closed posture for a 5-minute acclimation period to stabilize their physiological states and adapt to the experimental environment. Stringent control measures were employed in the experimental design to ensure data accuracy and scientific rigor. Subjects were instructed to keep their hands still to prevent artifacts from electrode displacement and muscle activity. The distal phalanges of the left index and middle fingers, characterized by high sweat gland density and well-developed sweat glands, were selected as the sites for EDA measurements. During the experiment, subjects were explicitly required to remain sitting quietly without talking during data collection to prevent non-specific related electrodermal responses caused by vocalization movements. The subjects closed their eyes, placed their right hands on the table in front of them, with their wrists on the tabletop and fingers slightly hovering in the air. Subsequently, the subjects were asked to gently stroke the plant stimuli using the index, middle, and ring fingers of their right hand. To preserve ecological validity and ensure a naturalistic tactile experience, no specific constraints were imposed on stroking speed or pressure, or the number of strokes. Each touch lasted for 15 s, and the interval between two stimulus material switches was 15 s. The inter-trial interval (ITI) was used to avoid the overlap of non-specific responses. The experimental procedure is shown in Fig. 1. After the experiment, subjects completed the Semantic Differential Scale (SD) featuring seven pairs of bipolar adjectives to describe tactile sensations. A 7-point Likert scale was used for scoring (1 = extreme left, 7 = extreme right). The scale’s reliability was assessed using Cronbach’s α coefficient, which yielded a value of 0.73. The Kaiser-Meyer-Olkin (KMO) measure exceeded 0.7, and Bartlett’s test of sphericity was statistically significant (p < 0.001), indicating the questionnaire had structural validity.

Statistical analysis

The experimental dataset comprises EEG (α and γ bands), EDA and HRV. The experimental data were processed and exported using ErgoLAB 3.0 software, and statistical analyses were conducted using SPSS 27.0. All measurements underwent the Kolmogorov-Smirnov normality test and Levene’s test for homogeneity of variances. For data conforming to a normal distribution, a paired t - test was used to compare the changes in physiological and psychological indicators of the subjects when touching different types of plant leaves. Non-parametric tests, such as the Wilcoxon signed-rank test, were applied for non-normally distributed data. Differences were deemed statistically significant at p < 0.05. Data visualization was performed using GraphPad Prism 10.1.2 software.

Results

Changes in physiological indicators

Physiological responses were compared between touching hairy and hairless leaves. EEG (α and γ waves), HRV (pNN50 and pNN20), and Phasic SCR indices were analyzed to examine the effects of different leaf textures on participants’ psychophysiological states (Table 1).

Table 1.

T-test of EEG indicators (n = 30).

brain waves	Channel	Hairy(M ± SD)	Hairless(M ± SD)	Mean Difference	t	p
α wave	C3	8.106 ± 2.484	8.218 ± 2.322	-0.112(-0.565,0.341)	-0.504	0.618
	C4	8.836 ± 2.695	9.554 ± 2.518	-0.718(-1.151,-0.285)	-3.394	0.002**
γ wave	C3	-2.520 ± 1.143	-2.857 ± 1.171	0.337(0.011,0.663)	2.112	0.043*
	C4	-0.595 ± 0.943	-0.829 ± 1.089	0.234(0.025,0.444)	2.288	0.030*

* p < 0.05, ** p < 0.01.

The T-value is used to indicate the magnitude of the difference between hairy and hairless samples, while the p-value represents the significance level of the difference. A p-value < 0 0.05 signifies a significant difference, and a p-value < 0 0.01 denotes an extremely significant difference.

Figure 2 presents representative raw EEG recordings from a single participant during tactile contact with hairy and hairless leaves. Consistent with the group-level statistical results, the representative recording shows higher amplitude fluctuations across channels during contact with hairy leaves compared to hairless leaves.

Fig. 2.

Representative raw EEG waveforms during tactile contact with plant leaves. (a) hairy leaves; (b) hairless leaves.

Figure 3 presents representative EDA recordings from the same participant during tactile contact with hairy and hairless leaves, displaying the raw EDA signal, tonic component, and Phasic SCR. Consistent with the group-level statistical results, the representative recording shows more prominent SCR peaks during contact with hairy leaves compared to hairless leaves.

Fig. 3.

Representative EDA recordings during tactile contact with plant leaves. (a) hairy leaves; (b) hairless leaves.

Figure 4 compares α-wave activity during tactile contact with hairy versus hairless leaves. In the C3 channel, no significant difference in alpha power was observed between hairy and hairless leaves. In the C4 channel, α-wave power was significantly lower when touching hairy leaves compared to hairless leaves, indicating reduced relaxation during contact with hairy surfaces.

Fig. 4.

α waves between touching hairy and hairless leaves, ** p < 0.01.

Figure 5 presents γ-wave activity changes. In both C3 and C4 channels, gamma power was significantly higher when touching hairy leaves compared to hairless leaves, suggesting enhanced cognitive activation and attention during contact with hairy surfaces.

Fig. 5.

γ waves between touching hairy and hairless leaves, * p < 0.05.

Figure 6 compares HRV during tactile contact with hairy versus hairless leaves. For pNN20, a paired-samples t-test revealed significantly lower values when touching hairy leaves (M = 43.396, SD = 27.075) compared to hairless leaves (M = 49.969, SD = 30.188; t = -2.804, p = 0.009). The Wilcoxon signed-rank test for pNN50 showed a median (P25, P75) of 0.00 (0.00, 19.29) for hairy leaves and 0.00 (0.00, 36.40) for hairless leaves, with a statistically significant difference (z = -2.027, p = 0.043). These results indicate that tactile interaction with hairy leaves leads to decreased HRV, reflecting reduced parasympathetic activity.

Fig. 6.

HRV between touching hairy and hairless leaves, n = 30,* p < 0.05 ,** p < 0.01.

Figure 7 compares Phasic SCR during tactile contact with hairy versus hairless leaves. The median (P25, P75) Phasic SCR value was significantly higher when touching hairy leaves [0.16 (0.02, 0.26)] compared to hairless leaves [0.11 (0.01, 0.19); z = -2.604, p = 0.009]. This elevated electrodermal response indicates greater sympathetic nervous system activation during contact with hairy leaf surfaces.

Fig. 7.

Phasic SCR between touching hairy and hairless plant leaves, ** p < 0.01.

Subjective perception

The semantic differential results indicated significant differences in tactile perception (Table 2). In the “hard - soft” dimension, hairy leaves showed a significantly higher tendency towards “softness”, while hairless leaves were perceived as harder. In the “ordinary - special” dimension, hairy leaves scored higher for specificity, whereas hairless leaves were perceived as more ordinary. Hairless leaves exhibited more positive emotional associations in the emotional dimensions. Hairless leaves were more likely to be perceived as friendly, while hairy leaves tended to be perceived as unfriendly. In the “anxious - calm” dimension, hairless leaves were more likely to be perceived as calm, while hairy leaves were more likely to trigger anxiety. The perceived pleasantness and likeability of leaf surfaces were evaluated along the dimensions of “unpleasant - pleasant” and “dislike - like.” Hairless leaves were rated higher in pleasantness and likeability compared to hairy leaves, although these differences did not reach statistical significance. The tactile experience of hairy leaves was associated with a relatively high level of “stimulation”, while hairless leaves were more likely to be perceived as “stable”, suggesting that surface texture significantly influences the subjective evaluation of tactile attributes.

Table 2.

t-test analysis of subjective perception.

	Groups(M ± SD)		t	p
	Hairy(n = 30)	Hairless(n = 30)
Hard - Soft	6.20 ± 1.19	4.57 ± 1.10	5.520	0.000**
Unfriendly - Friendly	4.60 ± 1.40	5.40 ± 1.19	-2.379	0.021*
Stimulating - Stable	4.67 ± 1.67	5.50 ± 1.20	-2.224	0.030*
Anxious - Calm	4.87 ± 1.61	5.67 ± 0.99	-2.312	0.025*
Unpleasant - Pleasant	4.97 ± 1.61	5.23 ± 1.14	-0.742	0.461
Disliked - Liked	5.07 ± 1.87	5.33 ± 1.37	-0.629	0.532
Ordinary - Special	4.87 ± 1.93	3.37 ± 1.45	3.409	0.001**

Conclusions and discussions

The findings largely supported the proposed hypotheses. H1a was supported: hairy leaves induced significantly higher gamma activity in both C3 and C4 channels compared to hairless leaves. H1b was supported: hairy leaves elicited elevated EDA. H1c was supported: hairy leaves resulted in reduced HRV indices (pNN50 and pNN20). H2 was partially supported: hairless leaves induced enhanced alpha activity, although this effect reached significance only in the C4 channel. H3 was supported: subjective evaluations revealed that hairless leaves were perceived as more calming, stable, and friendly, whereas hairy leaves were perceived as more stimulating and special.

By integrating the methods of physiological feedback and subjective perception, this study explored the differences in tactile perception of leaves from two different types of ornamental plants. The findings systematically elucidate the differential effects of plant leaves with varying tactile properties on human emotional regulation and cognitive responses. Significant differences were observed among EEG, HRV, and EDA indicators when touching leaves with different leaf surface types. Leaves with different tactile sensations can cause emotional changes in people. Leaves with hairless surfaces bring more pleasant and relaxing emotional responses, while hairy leaves are more likely to stimulate attention and the desire for exploration. In conclusion, the tactile characteristics of plant leaves significantly affect human emotions and cognition by modulating EEG rhythms and autonomic nervous activity. The soothing effect of hairless leaves and the arousing property of hairy leaves support the regulatory role of plant tactile texture on psycho-physiological states at the neural level. These findings offer a neurophysiological foundation for future horticultural interventions leveraging plant surface structures to regulate human perception.

Physiological responses

The tactile sensation of hairless leaves notably increased α wave activity (8–13 Hz), aligning with Koga et al.‘s²⁶ conclusion that plant contact fosters relaxation. Enhanced α waves are generally associated with relaxed wakefulness and emotional stability, consistent with Thayer et al.‘s⁶⁷ findings that alpha power is positively correlated with parasympathetic activity. Notably, this effect reached significance only at C4 and not at C3. Given that participants used their right hand for tactile stimulation, the contralateral hemisphere (left, C3) would typically show the strongest somatosensory response. However, the observed right-hemisphere (C4) lateralization may reflect the well-documented role of the right hemisphere in emotional processing⁶⁸, suggesting that the relaxation effect of hairless leaves involves affective rather than purely sensory mechanisms. This indicates that the smooth texture of hairless leaves may induce a relaxed state by diminishing tactile complexity and cognitive demands. A reduction in α wave amplitude signifies heightened brain alertness or tension. The results (Fig. 4) indicate that when subjects touched hairy plant leaves, the marked decrease in α wave in the C4 channel likely reflects reduced brain relaxation, suggesting increased psychological tension or focus. Conversely, touching the hairless leaves elicited a significant increase in α waves, implying that this type of tactile sensation is more likely to induce a state of relaxation and positive affect in the subjects.

Alpha activity recorded at C3/C4 may overlap with the sensorimotor mu rhythm, which shares the same frequency range (8–13 Hz) but originates from distinct neural generators⁶⁹. The significant alpha modulation was observed at C4 (ipsilateral to the stimulated right hand) rather than at the contralateral C3, which would typically show stronger responses in purely somatosensory tasks. This ipsilateral pattern likely reflects the well-documented right-hemisphere specialization for emotional processing⁶⁸, suggesting that the observed effects are more closely related to affective responses than to primary somatosensory processing. It should also be acknowledged that participants actively explored the leaf surfaces, and the C3/C4 electrodes overlie both somatosensory and motor cortices; thus, the recorded signals may contain sensorimotor components that cannot be fully dissociated with the current 16-channel montage. Nevertheless, the converging evidence from autonomic measures (HRV, EDA) and subjective ratings supports the interpretation that the differential neural patterns reflect distinct emotional states elicited by the two leaf textures.

In contrast, hairy leaves notably increased gamma wave activity (30–100 Hz), particularly in the central C4 channel. Gamma oscillations are associated with a broad range of higher-order neural processes, including sensory integration, attentional engagement, cognitive processing, and emotional arousal, and their enhancement reflects the need for high-order brain regions to integrate complex tactile information⁷⁰. For example, in a fine tactile discrimination task near the limits of spatial acuity, gamma-band power was selectively elevated in the dorsolateral prefrontal cortex (dlPFC) and posterior parietal cortex, and coherence between these regions increased for correct discriminations⁷¹. Accordingly, the elevated gamma activity observed when touching hairy leaves likely reflects the engagement of multiple neural processes simultaneously, including the integration of complex tactile inputs, heightened attentional demands, and emotional arousal in response to the unfamiliar surface texture, rather than any single mechanism alone. Henderson J et al. reported that during the active exploration of textured surfaces, rough stimuli elicit greater neural engagement than smooth stimuli. Specifically, active exploration of a rough texture (e.g. coarse fabric) leads to stronger cortical responses (e.g. enhanced alpha-band desynchronization in contralateral somatosensory regions) compared to a smooth texture⁷². In an EEG study, the presence of friction (simulating a rough surface) not only boosted late-phase beta/gamma activity in frontal regions, indicating greater cognitive/sensory integration, but also significantly elevated self-reported arousal and other emotional response measures⁷³. The result of our research supports the view that irregular, hairy tactile stimuli provoke exploratory attention and emotional arousal, engaging higher-level integrative processing in the brain to a greater extent than uniform smooth stimuli. These findings complement the study of grassland textures by Hassan et al.³¹, further supporting the mapping relationship between tactile features and neural responses.

Meanwhile, the significant decrease in alpha waves under this condition indicates that when individuals touch hairy leaves, their state of relaxation decreases while their alertness or sense of tension increases. Previous studies have also pointed out that the energy of alpha waves decreases significantly in stressful situations, and this phenomenon is often accompanied by increased alertness and anxiety⁷⁴. Therefore, the EEG patterns of hairy leaves, marked by decreased α waves and increased γ waves, indicate an activated state necessitating attention and sensory processing, which contrasts with the calm, α-wave-dominated state induced by hairless leaves.

Differences in autonomic nervous system indicators, such as HRV and EDA, further support the conclusions. The HRV indices pNN20 and pNN50 primarily reflect parasympathetic nervous system regulation. A decrease in these indices suggests reduced parasympathetic activity, potentially indicating heightened emotional arousal, including tension, anxiety, or alertness. The alterations in pNN20 and pNN50 suggest that contact with hairy leaves, as opposed to hairless leaves, diminishes parasympathetic nerve activation and reduces emotional relaxation. Notably, pNN20, which reflects parasympathetic activity, significantly decreased, indicating that this tactile stimulation may provoke more tense or uncomfortable psychological responses. When subjects touch the surface of hairy plant leaves compared to the surface of hairless leaves, the regulation of the autonomic nervous system tilts towards sympathetic nerve dominance, accompanied by a higher level of emotional arousal and a lower level of relaxation. This result further supports the impact of the tactile texture of plants on emotional states. Specifically, hairy plant leaves often induce alertness and tension, whereas smooth leaves are more likely to promote relaxation and comfort, which is consistent with the comfort and relaxation subjectively experienced by individuals.

The results of EDA also confirm the difference between the hairy and hairless leaves. A notable rise in the Phasic SCR value (Fig. 7) suggests that tactile interaction with hairy plant leaves may activate the sympathetic nervous system, eliciting a mild stress response⁴², whereas touching hairless plant leaves results in relatively minor Phasic SCR fluctuations, indicating a relatively relaxed physiological state. Taken together, the convergent patterns across EEG, HRV, and EDA measures provide mutually reinforcing evidence for a unified psychophysiological response to leaf surface texture. Touching hairy leaves simultaneously activated central cortical networks, as reflected by elevated gamma and reduced alpha power, and peripheral autonomic pathways, as indicated by elevated Phasic SCR and reduced HRV indices, collectively indicating a state of heightened arousal and sympathetic dominance. In contrast, hairless leaves consistently elicited reduced cortical activation alongside increased parasympathetic activity, supporting a coherent pattern of relaxation across both central and peripheral physiological systems.

Dimensional differentiation and consistency of tactile perceptions

The results of subjective evaluation revealed that the hairless leaves significantly outperformed the hairy leaves in dimensions such as “stimulation - stability” and “anxiety - calmness”, which was highly consistent with the relaxation effect of physiological data. Notably, although the hairy leaves were perceived as “unfriendly” and “highly stimulating”, they scored higher in the “specialness” dimension, indicating that tactile complexity may stimulate cognitive interest. The rough tactile sensation likely enhances attention by activating gamma waves¹⁸, prompting subjects to deeply process unconventional stimuli and thereby generating a “sense of novelty” and a “desire for exploration.” This result expands the “affective touch” theory proposed by McGlone²¹, indicating that the valence (positive/negative) of tactile experience may be jointly regulated by surface features and cognitive evaluation. The tactile properties of smooth surfaces may elicit positive emotional responses by mitigating anxiety, aligning with previous studies on the emotional valence of smooth materials⁷⁵. Overall, hairless leaves are perceived as more comfortable, friendly, and calming in tactile perception, whereas hairy leaves are often seen as unstable and uncomfortable. This difference may be related to the influence of leaf surface characteristics on tactile sensations. The surface characteristics of hairy leaves may increase the complexity of tactile perception, resulting in greater perceptual variability, while the surface characteristics of hairless leaves are more likely to be perceived consistently.

Applications and design insights

Tactile perception of plants provides empirical evidence for the plant configuration of sensory gardens and therapeutic landscapes. For relaxation-oriented scenarios (such as anxiety-relief spaces), it is recommended to prioritize plants with hairless leaves, whose smooth touch can promote emotional calmness through α-wave enhancement and parasympathetic activation. In blind gardens or educational landscapes, moderately introducing plants with hairy leaves may enhance the interactive experience through γ-wave-driven curiosity stimulation⁸. In addition, individual differences should be taken into account. Although this study did not examine gender and age effects, due to the decline in tactile sensitivity in the elderly population³¹, multimodal stimuli (such as combining aroma and texture) may be required to enhance the perception effect.

Limitations and future research

First, only two plant species were examined to represent hairy and hairless leaf textures. Although this simplified design enhances experimental control, the two species differed not only in trichome presence but also in other tactile properties such as leaf thickness and surface morphology, which may have jointly contributed to the observed effects; future studies should include a broader range of plant species with systematically varied surface characteristics. Second, participants were not asked to rate perceived hairiness, limiting direct assessment of subjective tactile discrimination; future research should incorporate subjective ratings of perceived hairiness to validate the intended tactile contrasts. Third, the sample was restricted to females, which may limit the generalizability of the findings; future research should incorporate diverse genders, ages, and cultural groups. Fourth, the experiment used excised leaves, which may weaken the influence of plant dynamic characteristics (e.g., temperature, humidity) in real scenarios; virtual reality technology could be leveraged to simulate more naturalistic environments. Fifth, all participants touched hairy leaves before hairless leaves in a fixed order, which may have introduced potential order effects. Physiological arousal naturally declines over time as participants habituate to the experimental setting, and the 15-second inter-stimulus interval may have been insufficient for EDA and HRV to return to baseline levels following contact with hairy leaves, given that EDA recovery typically requires tens of seconds to minutes and HRV recovers more slowly than EEG. The elevated arousal observed during hairy leaf contact may partially reflect novelty effects or experimental anxiety associated with being the first stimulus, rather than exclusively reflecting leaf surface texture; counterbalanced designs should be employed in future studies. Sixth, the short-term tactile stimulation employed may not adequately capture the cumulative effects of prolonged contact; future research should examine extended exposure durations. Seventh, no specific constraints were imposed on stroking speed or pressure, and movement dynamics were not monitored, which may have introduced uncontrolled variability in the intensity of tactile stimulation across participants. Differences in stroking speed and pressure influence the degree of skin deformation and mechanoreceptor activation, potentially affecting EEG somatosensory responses, EDA sympathetic activation, and HRV autonomic regulation; motion tracking or standardized tactile delivery systems should be employed in future studies to quantify and control these parameters. Eighth, beta-band activity (13–30 Hz) was not included in the present analysis. While its exclusion was necessary due to the overlap between beta oscillations and the sensorimotor mu rhythm during active tactile exploration, this trade-off may risk overlooking potentially relevant neural processes, as beta oscillations have been associated with predictive coding, top-down cognitive control, and sensorimotor integration beyond purely motor functions^76,77. Future studies employing higher-density EEG with source localization or independent component analysis may better dissociate perceptual from motor contributions in the beta band during active plant tactile exploration. Additionally, multimodal neuroimaging techniques such as functional near-infrared spectroscopy (fNIRS) could help analyze the spatial dynamics of the tactile-emotion pathway, and the quantitative relationship between the microscopic structure of plant leaves and physiological responses warrants further investigation. Interdisciplinary collaboration will be essential to reveal the biological mechanisms of tactile interaction and promote practical applications for harmonious human-nature relationships.

Author contributions

X.W. and J.J. wrote the main manuscript text and prepared figures. All authors reviewed the manuscript.

Data availability

The data that support the findings of this study are not publicly available due to the sensitive nature of the physiological data collected from human participants and existing confidentiality agreements. These data were used under specific authorization for the current study and cannot be shared openly. However, data may be made available upon reasonable request, subject to approval through research ethics review and permission from the data provider.

Declarations

Competing interests

The authors declare no competing interests.

Informed consent statement

The participants provided their written informed consent to participate in this study.

Consent for publication

Participants have signed informed consent, and the experiments in the manuscript have ensured that no mental or physical harm is caused to the subjects and no harm is done to their safety and interests. They are consent for publication.

Institutional review board statement

Approval was obtained from the ethics committee of the North China University of Water Resources and Electric Power, China (Approval No. NCWU-SA-2023-002). The procedures used in this study adhere to the tenets of the Declaration of Helsinki. The use of plants in the present study complies with international, national, and/or institutional guidelines.