The Impact of Earth-Based Building in Residential Environments on Human Emotional Relief Using EEG + VR + LEC Method

Abstract

Urbanization exacerbates mental health challenges, prompting the exploration of biophilic design solutions. This study examined the therapeutic potential of raw earth through its thermal interactions in architecture. First, energy consumption simulations established distinct indoor temperature ranges for raw earth, concrete, and steel under identical energy constraints: low (22.8 ± 0.32 °C), medium (26.5 ± 0.39 °C), and high (30.1 ± 0.84 °C). The study then quantified the differences in physical and psychological perceptions across material-dominated spaces under controlled temperatures above. Nine scenes were constructed for emotional healing evaluation, incorporating the olfactory dimension into the Electroencephalogram (EEG) + Virtual reality (VR) + Laboratory environmental control (LEC) approach. The results indicated that raw earth materials were most effective in promoting emotional recovery under thermal stress conditions (low/high temperatures), as evidenced by a significant enhancement of α EEG rhythms. However, under moderate conditions, concrete environments produced the greatest relaxation effects, while steel environments were most conducive to enhancing focus. The core conclusion of this study is that the therapeutic effects of building materials are not static but are intricately linked to the surrounding thermal environment. This provides a new perspective for evidence-based healthy building design and underscores the importance of optimizing material selection based on specific environmental conditions and needs.

1. Introduction

1.1. Research Background

Global urbanization has been increasing significantly, with more than half of the world’s population now residing in urban areas. This trend is predicted to persist, with projections indicating that 68% of the global population will live in urban areas by the middle of the 21st century [1]. China is also experiencing unprecedented urbanization; the proportion of its population living in urban areas reached 67% in 2024 [2], and it is estimated that by 2030, China’s urban population will exceed one billion people [3]. A growing body of research suggests that urbanization may be detrimental to the mental health of urban dwellers, such as psychotic experiences, depression, and stress-related disorders, particularly among vulnerable groups [4]. Mental health problems, including depression, anxiety disorders, and chronic fatigue syndrome, are becoming increasingly prominent and represent a significant global public health concern. Factors such as social disparities, insecurity, pollution, and a lack of connection with nature are recognized as contributors to mental health challenges in urban settings [5].

The introduction of biophilic design is one of the potential solutions [6]. The Biophilia Hypothesis, proposed by E.O. Wilson in 1984 [7], posits that, due to evolutionary factors, humans have an “innate tendency to focus on life and lifelike processes”. This connection to nature is believed to enhance psychological well-being by reducing stress and increasing positive emotions. From an evolutionary psychology perspective, this innate biological predisposition likely conferred survival advantages to early humans who relied on natural ecosystems, resulting in a lasting genetic affinity for nature that continues to exist in modern populations [8]. Stress Recovery Theory (SRT) suggests that natural environments promote psychological recovery by reducing physiological stress responses [9], while Attention Restoration Theory (ART) elucidates that natural environments promote the recovery of unconscious attention through “soft charms” (e.g., flowing water, vegetation), which alleviate the cognitive fatigue induced by directed attention, thereby enhancing concentration and creativity [10].

Raw earth refers to building materials that can be utilized in construction without baking but only simple mechanical processing, using raw earth as the primary ingredient [11]. As a typical biophilic material, raw earth has been used for thousands of years and continues to be widely used across the globe. Contemporary statistics indicate that approximately 30% of the global population resides in earthen structures, with this figure exceeding 50% in certain developing nations [12,13]. Research has demonstrated that raw earth buildings can provide comfortable indoor temperatures and enhance air quality [14,15] through their natural hygrothermal regulation properties [16,17], thereby promoting a sense of relaxation [18]. Engaging with raw earth—through observation, tactile interaction, and olfactory stimulation—has been shown to increase human high-frequency heart rate variability (HF-HRV), decrease heart rate, alleviate stress, and improve emotional well-being [19]. Despite the benefits of raw earth, current research has focused on material modification [20,21,22,23,24] and construction processes [25,26,27], with fewer studies investigating its healing effects. Existing research is often constrained by limited spatial contexts and the subjective nature of ambiguous evaluations, resulting in a lack of systematic and reliable data collection methods.

Subjective human perception is influenced by numerous unmanipulated variables beyond the controlled experimental environment. To address the issues of vagueness and lack of accuracy in subjective evaluations, human factors engineering was introduced to employ quantitative methods [28,29,30]. In recent years, evidence-based design (EBD) methods, grounded in human factors engineering techniques, have been applied in architectural and environmental studies to investigate human perceptions of the building environment and to inform design decisions [31,32,33]. Design solutions are developed based on these data and are continuously monitored. To overcome the limitations of physical space and cost, Li et al. [34] combined virtual reality (VR) with laboratory environmental control (LEC) to create a controlled experimental environment. Their findings demonstrated that subjective feelings in virtual and real environments were in good agreement [35]. Recent studies have also demonstrated the validity of VR in simulating built environments and assessing human responses. For instance, Latini et al. [36,37] compared human performance in real and virtual environments regarding task performance, comfort, and interactive behavior, showing that virtual environments can effectively simulate real office environments and are suitable for productivity, comfort, and behavioral studies. Ozcelik et al. [38] found that users’ thermal comfort and satisfaction scores in virtual and physical environments did not differ significantly (p > 0.05), further confirming that virtual environments can effectively simulate the subjective experience of real thermal environments. Regarding stress induction, the VR environment was effective in eliciting stress by simulating real-world scenarios, resulting in significant changes in physiological indicators such as cortisol levels, heart rate, and galvanic skin response, which were comparable to those induced by real-world stressors [39]. In terms of emotional recovery, natural environments simulated by VR, such as forests, have been shown to reduce negative emotions and enhance positive emotions, such as vitality [40,41]. In summary, virtual reality offers a powerful platform for simulating biological environments and evaluating their effects on users’ psychological and physiological responses. The combination of VR and LEC can serve as an effective approach to studying the physical and mental experiences of individuals in raw earth environment.

Existing studies have established a connection between human cognitive performance and psychological assessments by monitoring physiological signals such as heart rate [42,43], blood oxygen levels [34], eye movement [44,45], and electroencephalogram results [46,47,48,49,50]. These physiological indicators serve as a basis for understanding human cognitive performance. EEG signals, known for high temporal resolution, capture physiological changes occurring within milliseconds, directly reflecting the activity of the central nervous system. Consequently, EEG has become a vital tool for assessing human cognition and perception in indoor environments [51]. Notably, Nayak et al. [52] found that a model trained using EEG signals to assess indoor thermal environment was 16 times more accurate than models based on other physiological parameters. Yu et al. [53] also found good prediction capabilities of EEG coupled with machine learning (ML) for various feedback items, achieving accuracies ranging from 0.85 to 0.96 for within-subject scenes and from 0.88 to 0.94 for all-subject scenes. These findings provide a strong rationale for selecting EEG as a method for evaluating physiological perceptions. Many studies have demonstrated the close relationship between the frequency domain characteristics of brainwave signals and the cognitive states of the brain. Delta waves (0.5–4 Hz), most commonly observed in infants and young children, are primarily associated with deep sleep [54]. Theta waves (4–8 Hz) emerge during states of intense arousal or light sleep and have been linked to fuzzy consciousness, distraction, and anxiety [55,56]. Alpha waves (8–12 Hz) are associated with daydreaming, an inability to concentrate, and deeply relaxed states. When suppressed, alpha waves may contribute to anxiety, high stress, and insomnia. Conversely, when alpha waves are prominent, they facilitate a state of relaxation [54]. Beta waves (12–35 Hz) are typically observed during wakefulness and are crucial for conscious focus, memory, and problem-solving. However, the prominence of beta waves can lead to anxiety, heightened arousal, an inability to relax, and stress [57]. Gamma waves (>35 Hz) are responsible for cognitive functioning, learning, memory, and information processing. An increase in gamma wave activity is also associated with anxiety, high arousal, and stress [54]. Regarding the influence of spatial physical elements on healing effects, thermal environment significantly impacts individuals’ physical and mental health recovery by affecting factors such as emotional responses [58], stress levels [59], and sleep quality [60]. Yang et al. [61] created three thermal environment conditions in a climate chamber (PMV = −1, PMV = 0, PMV = 1) and discovered that neutral and slightly cooler environment reduced stress, while slightly warmer environment exacerbated it. They also identified a synergistic effect between thermal environment and psychological stress. Kim et al. [62] investigated the effect of thermal sensation on emotional responses and found that indoor thermal environment significantly influenced emotional reactions through subjective thermal sensation. They observed that individuals experienced more positive emotions when they felt neutral or slightly warmer. Wang et al. [63] assessed the effects of three thermal environments (21.7 °C, 25.2 °C, 28.6 °C) on cognitive performance by measuring mental load with EEG. They found that the mental load required to achieve the same cognitive performance was higher in slightly warmer environment.

The studies mentioned above provide important foundational support for the current research question; however, there remains limited research on which specific temperatures and materials contribute to a more healing architectural space.

1.2. Objective of this Study

Due to their unique biophilic properties and low environmental impact, raw earth materials have shown significant physiological and psychological healing potential in the realm of biophilic design. However, current research on the perceptual feedback associated with raw earth architectural spaces remains insufficient. Firstly, the relationship between building materials, energy consumption, and the thermal environment is often underemphasized in existing studies, making it challenging to accurately assess the intrinsic properties of materials under consistent energy supply conditions. Secondly, the current findings lack a systematic quantitative analysis of human physiological and psychological responses in controlled environmental settings. Traditional research methods are limited by vague assessments of subjective feelings, hindering the understanding of the interaction mechanisms between raw earth buildings and human perceptual feedback. In this study, quantitative experimental comparisons were conducted to identify the differences in physical and psychological perceptions of raw earth buildings compared to other typical building types (concrete, steel) under different temperature conditions. This state-of-the-art study explores the mechanisms linking architectural spaces dominated by different materials to human psycho-physiological responses. Therefore, the objectives of this study are to address the following questions:

(1)

Does raw earth, as a biophilic material, provide advantages over other industrial materials (steel, concrete) in terms of its healing effects while maintaining an equivalent level of energy consumption?

(2)

What differences in psychological healing effects do subjects exhibit in various combinations of temperature environments and material spaces?

The integration of EEG, VR, and LEC was chosen to address the limitations of traditional subjective evaluations and to enable a controlled and immersive experimental setting. EEG provides high temporal resolution for capturing real-time neurophysiological responses associated with relaxation and stress [52,53]. VR offers a scalable and immersive medium to simulate material-dominated environments while maintaining experimental consistency [36,37,38]. LEC ensures precise control over thermal and other conditions. This multimodal approach allows for both objective physiological data collection and subjective experience assessment, enhancing the validity and reliability of the findings.

1.3. Glossary of Key Terms

To ensure clarity for a broad readership, some key technical terms and acronyms are defined as follows:

❖

EEG (Electroencephalogram): A non-invasive technique used to record the brain’s electrical activity via electrodes placed on the scalp. This study focused on the power of alpha (α, 8–12 Hz), beta (β, 12–35 Hz), and theta (θ, 4–8 Hz) rhythms, which are associated with relaxed/meditative states, active thinking/attention, and drowsiness/light sleep, respectively.

❖

VR (Virtual Reality): A computer-generated simulation of a three-dimensional environment that users can interact with in a seemingly real way using special electronic equipment

❖

LEC (Laboratory Environmental Control): Refers to the precise control and manipulation of physical environmental parameters (e.g., temperature, humidity) within a lab setting to simulate specific conditions.

❖

Biophilic Design: An approach to architecture that seeks to connect building occupants more closely to nature by incorporating natural materials, light, vegetation, and other experiences of the natural world.

2. Materials and Methods

In this study, the energy consumption simulation was used to determine the range of indoor temperatures corresponding to different building materials (raw earth/steel/concrete) at the same energy consumption level. On this basis, the experimental methodology of EEG + VR + LEC, previously developed by our team, was employed. The virtual environment, enhanced by VR and LEC, provided immersive experiences across various scenes for participants. The researchers investigated the relationship between spatial environment and human physical and mental health by collecting both physiological signal data and subjective questionnaire responses.

2.1. Experimental Environment Setting

Firstly, the experimental building scenes were based on real laboratory environment and were computer-modeled to accurately reflect actual conditions. The spatial prototype represented a typical small self-sustaining living module, measuring 6000 mm × 3000 mm. In this case, S1 utilized raw earth as the primary building material, while S2 and S3 employed steel and concrete, respectively (as shown in

Table 1. Three VR experimental building scenes (Source: Author).

Scene Number	S1	S2	S3
Construction material	Raw earth	Steel	Concrete
Window-Wall ratio	10%	80%	35%
Plan display
Model display
Indoor Scene

). This study did not isolate the effects of lighting. Instead, to maintain ecological validity, it embraced the holistic environmental changes, including varied window-to-wall ratios, that are inherently linked to these material choices. By reviewing relevant literature and normative atlases [64,65], the corresponding window opening rates for S1, S2, and S3 were determined to be 10%, 80%, and 35%, respectively, based on the characteristics of the materials and their construction properties. This approach reflected authentic architectural scenes, where material-driven spatial characteristics collectively shaped occupants’ experiences. Additionally, furniture and greenery that enhance the user’s living experience were incorporated into the scenes.

The envelope design of the three self-sustaining modules were optimized to meet the heat transfer coefficient requirements outlined in the Design standard for energy efficiency of residential buildings in severe cold and cold zones (JGJ26-2018 [66]). Each residential unit features full-coverage photovoltaic (PV) modules on the roof, with a constant efficiency of 0.15. Utilizing the Design Builder v6.1 platform, a parametric energy model was developed to simulate energy consumption based solely on the power supplied by the rooftop PV system. This simulation indicated a daily power generation of 9.36 kWh during the summer months in the Beijing area. The average indoor temperatures of the three types of enclosure structure modules (raw earth/concrete/steel) simulated under the data from a typical meteorological year are as follows: 22.8 ± 0.32 °C, 26.5 ± 0.39 °C, and 30.1 ± 0.84 °C. These temperature gradients correspond to low-temperature, medium-temperature, and high-temperature zones, respectively (see

Table 2. Three temperature ranges (Source: Author).

	Temperature Range
Low temperature zone	22.8 ± 0.32 °C
Middle temperature zone	26.5 ± 0.39 °C
High temperature zone	30.1 ± 0.84 °C

). The participants were required to engage in all three temperature intervals, resulting in a total of nine scenes (see

Table 3. Nine virtual reality experimental scenes (Source: Author).

	S1	S2	S3
Low temperature zone	L-S1	L-S2	L-S3
Middle temperature zone	M-S1	M-S2	M-S3
High temperature zone	H-S1	H-S2	H-S3

). The thermal environment was regulated by LEC. To provide the subjects with a more immersive spatial experience, the experimental scene was visually presented using VR equipment. Additionally, the inherent odors of different materials were conveyed through the odor player to enhance the participants’ olfactory perception. This approach aimed to emphasize the unique characteristics of raw earth, a biophilic material.

To record EEG signals and heart rate, participants were instructed to wear an EEG device and a heart rate monitor. Simultaneously, a physical environment monitoring device was employed to control environmental temperature, humidity, carbon dioxide levels, and air quality, ensuring compliance with preset standards (see

Table 4. Experiment equipment (Source: Author).

Contents	Tools	Tools’ Information	Test Program
Physiological signal monitoring	Emotive Flex 2 Saline	EMOTIV, San Francisco, CA, USA	EEG
	ViATOM O2 Ring	Shenzhen Viatom Technology Co., Ltd., Shenzhen, China	Blood oxygen, Heart rate
Physical environment monitoring	Tianjian Huayi WZY-1	Beijing Tianjian Huayi Technology Development Co., Ltd., Beijing, China	Temperature
	Tianjian Huayi WEZY-1		CO2
	Temtop-H3	Elitech Technology, Inc., San Jose, CA, USA	IAQ
Scene environment simulation	HTC VIVE/Pro Eye	HTC Corporation, Shanghai, China	Visual and Auditory Dimension
	X-Scent 3.0	Hangzhou Scentrealm Technology Co., Ltd., Hangzhou, China	Olfactory Dimension

). For data analysis, Emotiv PRO v4.2.0 was utilized to export the raw EEG data, which were filtered and processed in MATLAB (R2020b). Finally, relevant analyses were conducted using SPSS v25 (see Figure 1).

2.2. Experimental Process

A total of 90 experiments were conducted in this study, with participants openly recruited from the university student population (mean age: 22.47 ± 3.53 years; male-to-female gender ratio: 1:1). All participants provided written informed consent prior to the experiment. The inclusion criteria for participants were: (1) good physical and mental health with no history of major illness; (2) normal or corrected-to-normal vision; (3) no history of neurological or psychiatric disorders; and (4) no reported susceptibility to motion sickness in 3D environments. Participants were instructed to avoid caffeine and alcohol for 24 h before the experiment.

We acknowledge that the use of a university student cohort limits the generalizability of our findings to the broader, more age-diverse urban population. This specific sample was chosen for this foundational study to minimize potential confounding variables related to age, educational background, and health status, thereby allowing for a clearer assessment of the direct psycho-physiological impacts of the built environment. This limitation is further addressed in Section 4.

Subjects were randomly divided into three groups to participate in the experiment under three temperature conditions: low, medium, and high. Within each group, 30 participants were exposed to all three virtual material scenes (S1: Raw Earth, S2: Steel, S3: Concrete) in a counterbalanced, randomized order to mitigate order effects. The material scene thus served as a within-subjects factor. This mixed-design resulted in a total of 90 experimental trials. The experiment was conducted in following two main steps and the experimental procedure for a single trial is illustrated in Figure 2.

• Preparation Phase: Participants were first provided with an explanation of the experiment and then fitted with the experimental equipment (see Figure 3). Each participant was equipped with a 32-channel wireless EEG head cap system to measure EEG signals from their scalp. After donning the VR headset, pulse oximeter, and olfactory stimulation device, the participants were instructed to rest with their eyes closed for 30 s to minimize fluctuations caused by individual adaptation to the VR and EEG.
• Experimental Phase: In this phase, three scenes of repetitive experiments were conducted, with the order of experiments randomized for each scene. The psychological and physiological responses of the participants were measured across the different scenes. First, the subjects entered a predetermined virtual reality scene, designated as S0, which resembled a typical indoor environment. The baseline heart rate of each subject was monitored and recorded. To ensure that the subjects maintained a consistent state upon entering the test scene, stimuli were applied until their heart rates stabilized at a normal working level. The primary form of stimulus used in this experiment was numeric computation. Subsequently, the scene was randomly transitioned to one of the three test environments. Subjects adapted to the new scene for one minute before resting for three minutes. During this period, physiological parameters, including EEG, blood oxygen and heart rate, were recorded. After the resting phase, subjects were required to spend two minutes in each scenario to complete a subjective questionnaire corresponding to that particular scene.

2.3. Data Analysis

This study employed 90 multidimensional perceptual datasets, which included physiological signals and subjective evaluation data. The data were processed and analyzed according to the following steps: First, the research data were checked for validity to ensure quality and reliability. Second, the EEG data were processed and filtered for further analysis. Finally, the physiological signal data were analyzed with the subjective questionnaire responses from the participants in a spearman correlation analysis to explore the relationship between the building environment and human behavior. The specific data analysis process is outlined below.

2.3.1. EEG Data Processing

The manipulation of EEG data involved several key steps: data cleaning, processing, and analysis (see Figure 4). The first step was to filter the EEG data using band-pass filtering to eliminate interference from background signals, remove eye movement and electromyography (EMG) artifacts, use a 0.16 Hz first order high-pass filter to remove the DC offset from the data and retain EEG signals within the [−100, +100] range. Next, the experimental process was divided into a 1.5 min stimulation phase and a 3 min recovery phase, with each phase further divided into 30 s intervals for analysis (see Figure 5). Each 30 s segment from the moment of entering the VR scene forward was designated as a stimulation slice, while each 30 s segment from the moment of initiating rest in the VR scene backward was designated as a rest slice. In a single trial, a total of twenty-seven 30 s slices were analyzed per participant (3 scenes × 9 slices per scene). Finally, the EEG slice data were transformed from the time domain to the frequency domain using Fast Fourier Transform (FFT) to further analyze the distribution of EEG rhythms across different frequency bands and to observe trends in changes.

2.3.2. Self-Report

Based on the Stimulus-Organism-Response (S-O-R) theoretical framework in environmental psychology, this study developed a three-dimensional environmental perception assessment scale (see

Table 5. Test Structure of the Subjective Questionnaire (Source: Author).

Test Item	Test Subitem	Evaluation Scale	Parameter
Affective Dimension	My mood is more relaxed	Not fit–very fit	Core [−3, −2, −1, 0, 1, 2, 3]
	My emotion	Very irritable—very calm
	Heart rate than usual	Faster—slower
	Concentration	Focused—distracted
	Willing to stay for a long time	Unwilling—willing
Spatial Cognition	Space size	Narrow–spacious
	Space permeability	Closed-transparent
	Spatial smell	Uncomfortable—comfortable
	Light perception	Dark-bright
	Thermal perception	Cold—hot
Environmental Interaction	VR environmental adaptability	Not adapted—adapted
	VR environmental satisfaction	Dissatisfied—satisfied
	Thermal acceptability	Uncomfortable—comfortable
	Light environment acceptability	Uncomfortable—comfortable
	Overall acceptability	Uncomfortable—comfortable

for details). The scale comprises three dimensions: Affective Dimension, Spatial Cognition, and Environmental Interaction. It systematically evaluates the psychophysiological response mechanisms in virtual spatial environments through 15 measurement items.

The construction of the scale strictly adhered to the principles of the Semantic Differential Method (SDM), which is widely utilized by international scholars. It employed Bipolar Adjective Pairs that have been tested for reliability and validity as the basis for assessment. In designing the measurement scale, a balanced seven-point continuous scale [−3, +3] was implemented to accurately quantify individual differences in participants’ psychological experiences. The zero point was anchored to the state of Neutral Experience. The two extremes of the scale represented the maximum validity of “Inhibitory Experience” (−3) and “Facilitative Experience” (+3), while the middle gradient indicated the intensity of psychological feelings. Additionally, for each assessment question, the abstract valence was translated into more intuitive perceptual indicators, using specific assessment terms such as “uncomfortable/comfortable” and “narrow/spacious”.

2.3.3. Statistical Method

All statistical analyses were performed using SPSS version 25.0, with the significance level (α) set at 0.05.

To analyze the effects of the built environment on physiological data (EEG rhythms, heart rate, blood oxygen), a series of repeated-measures analyses of variance (ANOVAs) were conducted. This method was chosen as it is well-suited for this mixed-design study, which includes both a between-subjects factor (Temperature: Low, Medium, High) and a within-subjects factor (Material Scene: Raw Earth, Steel, Concrete). Mauchly’s test was used to assess the assumption of sphericity. In cases where this assumption was violated, the Greenhouse-Geisser correction was applied to the degrees of freedom. For significant main effects, post hoc pairwise comparisons were conducted using the Bonferroni correction.

To investigate the relationship between objective physiological measures (EEG rhythms) and subjective psychological perceptions (questionnaire scores), the Spearman rank correlation coefficient was calculated. This non-parametric test was selected as it does not require the assumption of a linear relationship between variables and is robust to outliers, making it appropriate for analyzing Likert-scale questionnaire data.

3. Results

3.1. Overall Trends of α/β/θ Rhythms Across Three Temperature Ranges

To investigate whether the overall effect of temperature on brain rhythms is statistically significant, this study conducted repeated-measures ANOVA on the alpha, beta, and theta rhythms under low, medium, and high-temperature conditions, using Greenhouse-Geisser correction. The results showed that the main effects of the three brain wave bands under different temperature conditions were statistically significant (as illustrated in

Table 6. Repeated-measures ANOVA results for rhythms at different temperature ranges (Source: Author).

Rhythm	F Value	p Value	η2 Value
Alpha	57.157	0.000	0.648
Beta	99.337	0.000	0.762
Theta	86.629	0.000	0.736

). By analyzing the mean values of α/β/θ rhythms from all participants at rest in different temperature ranges, the following findings can be derived (as illustrated in

Table 7. Overall trends of α/β/θ rhythms across three temperature ranges (Source: Author).

Rhythm	Average			S1
	Low Temp Zone (22.8 ± 0.32 °C)	Middle Temp Zone (26.5 ± 0.39 °C)	High Temp Zone (30.1 ± 0.84 °C)	Low Temp Zone (22.8 ± 0.32 °C)	Middle Temp Zone (26.5 ± 0.39 °C)	High Temp Zone (30.1 ± 0.84 °C)
Rhythm	S2			S3
	Low Temp Zone (22.8 ± 0.32 °C)	Middle Temp Zone (26.5 ± 0.39 °C)	High Temp Zone (30.1 ± 0.84 °C)	Low Temp Zone (22.8 ± 0.32 °C)	Middle Temp Zone (26.5 ± 0.39 °C)	High Temp Zone (30.1 ± 0.84 °C)

):

(1)

Under different temperature conditions, the most significant activation of α rhythms was observed in the occipital region of the brain. The primary areas of change for β rhythms were predominantly located in the temporal lobe. Significant activation of θ rhythms primarily occurred in the frontal regions of the brain, suggesting that different EEG rhythms may exhibit distinct patterns of response to temperature fluctuations across various brain regions.

(2)

In the resting state, both the α and β bands exhibited varying degrees of increase with rising temperatures. When the ambient temperature was elevated from the low temperature zone (=22.5 °C) to the high temperature zone (=30.5 °C), the maximum increase in α rhythms was 28.8%, while the maximum increase in β rhythms was 13.7%. These findings suggested that an increase in temperature may have a modulatory effect on the neuroelectrical activity of the brain, resulting in enhanced power of the α and β rhythms.

(3)

Compared to the α/β bands, the θ rhythm (4–7 Hz) demonstrated a decreasing trend as temperature increased, with a maximum reduction of 24.2% observed in the T7 channel. Furthermore, EEG recordings conducted under hypothermic conditions revealed that θ rhythm activity was more pronounced in the frontal and left parietal regions.

(4)

The average rate of change in temperature was 10.8% for the α rhythm, 6.5% for the β rhythm, and 10.1% for the θ rhythm. This indicated that the temperature sensitivity of α and θ rhythms was greater.

3.2. Trends in α/β/θ Rhythms over Time

To further quantify the dynamic changes in brain rhythms from the stimulated state to the resting state, we performed repeated repeated-measures ANOVA on the α, β, and θ rhythms with time slices (Sim-R6) as the independent variable. The analysis showed that the time factor had a significant main effect on the α, β, and θ rhythms, as shown in

Table 8. Repeated-measures ANOVA results for rhythms in different time slices (Source: Author).

Rhythm	F Value	p Value	η2 Value
Alpha	493.187	0.000	0.941
Beta	179.982	0.000	0.853
Theta	19.071	0.000	0.381

. Based on the analysis of EEG data, this study further subdivided the stimulation and resting phases into 30 s time slices for a comparative analysis of EEG rhythms. As illustrated in Figure 6, during the transition from the stimulated state to the resting state, both the α rhythms and β rhythms exhibited an increasing trend, ultimately converging to a steady state in the R2 slice. Notably, the α rhythms responded more rapidly and increased to a greater extent, with a maximum increase of 46% for α and 9% for β. This indicates that all three spatial scenes in the experimental design effectively promoted the subjects’ mental relaxation. In contrast, the θ rhythms exhibited the least fluctuation, showing a gradual rebound after an initial decrease of approximately 5% at the onset of the state transition.

3.3. Correlation of α/β/θ Rhythms with Questionnaires

To investigate the relationship between EEG rhythms and the perception of spatial physical elements, this study analyzed the spearman correlation between EEG rhythms and the subjective questionnaire scores of participants in various scenes during a resting state. The questionnaire data primarily focused on two perceptual dimensions: the first pertained to the subjects’ physiological perceptions, such as temperature (cold or hot) and light intensity (bright or dark) (SPV, SOV, TSV, LSV), while the second related to their psychological assessments, including comfort levels regarding temperature and light (comfortable or uncomfortable) (TCV, LSV). The study found that, as illustrated in Figure 7:

(1)

Overall, there was a positive correlation between thermal comfort vote (TCV) and α/β/θ rhythms. In addition, the β rhythms exhibited a positive correlation with spatial odor vote (SOV), while θ rhythms were significantly positively correlated with lighting comfort vote (LCV) and spatial permeability vote (SPV).

(2)

The α rhythms exhibited a strong positive correlation (R = 0.33, p = 0.071) with thermal comfort vote (TCV) in the M-S1 scene. In contrast, in the H-S1 scene, the α rhythms demonstrated a more significant negative correlation (R = −0.44, p = 0.014) with lighting sensitive vote (LSV).

(3)

The β rhythms exhibited a stronger correlation with the perception of spatial physical elements in the mid-temperature environment. It was observed that β rhythms showed a positive correlation with thermal sensitive vote (TSV, R = 0.36, p = 0.050) and lighting comfort vote (LCV, R = 0.42, p = 0.020) in M-S2.

(4)

In terms of spatial perception, in M-S3, the θ rhythms were highly correlated with the lighting comfort vote (LCV, R = 0.54, p = 0.002), suggesting that θ rhythms were more active when people perceived the light environment as comfortable. Additionally, regarding the properties of space, θ rhythms exhibited a positive correlation with spatial permeability (SPV, R = 0.47, p = 0.009) and thermal sensitive vote (TSV, R = 0.43, p = 0.019) in M-S3.

3.4. Changes in α/β Rhythms in Nine Scenes During State Transition

Changes in α/β rhythms before and after stimulation of the resting state were analyzed to explore differences in EEG rhythm alterations within the temperature-material matrix (see

Table 9. Changes in α/β rhythms during state transition (Source: Author).

Scene	αs (Proportion)	αr (Proportion)	A (%)	βs (Proportion)	βr (Proportion)	B (%)
L-S1	0.106	0.141	33.2%	0.243	0.264	8.8%
L-S2	0.111	0.143	28.7%	0.252	0.266	5.2%
L-S3	0.113	0.140	23.4%	0.250	0.266	6.3%
M-S1	0.103	0.155	50.9%	0.256	0.282	10.3%
M-S2	0.104	0.162	55.8%	0.254	0.284	12.1%
M-S3	0.100	0.158	59.0%	0.262	0.283	8.2%
H-S1	0.102	0.159	55.1%	0.263	0.282	7.3%
H-S2	0.108	0.157	45.4%	0.253	0.285	12.9%
H-S3	0.109	0.156	42.9%	0.260	0.277	6.4%

). The rate of change in EEG rhythms was calculated as follows:

$$\mathrm{A}=(\mathsf{\alpha }\mathrm{r}-\mathsf{\alpha }\mathrm{s})/\mathsf{\alpha }\mathrm{s}$$

$$\mathrm{B}=(\mathsf{\beta }\mathrm{r}-\mathsf{\beta }\mathrm{s})/\mathsf{\beta }\mathrm{s}$$

where αr represents the average proportion of α rhythms to total rhythms in the resting state, and αs represents the average proportion of α rhythms to total rhythms in the stimulated state. Similarly, βr denotes the average proportion of β rhythms to total rhythms in the resting state, while βs denotes the average proportion of β rhythms to total rhythms in the stimulated state.

To compare whether there are significant differences in the α/β rhythm change rates in different temperature–material scenarios, repeated-measures ANOVA was performed on the above data. The results showed that there were significant differences in the α/β rhythm change rates in different scenarios (α: p = 0.001; β: p = 0.000). When considering temperature alone, the degree of change in the α rhythm was ranked as follows: middle temperature (A = 1.66) > high temperature (A = 1.43) > low temperature (A = 0.85). This indicated that temperature significantly influenced variations in the α rhythm. Both medium and high temperatures enhance the α rhythm and promote emotional recovery. Additionally, it was observed that the most substantial changes in α occurred at both low and high temperatures during S1 (L-S1: 33.2%, H-S1: 55.1%). This suggested that emotional recovery in response to the raw earth scene was most pronounced at these temperature extremes. In contrast, at middle temperature, the largest change in α occurred at S3 (58.9%), indicating a more significant emotional recovery associated with the concrete scene. Compared to the α rhythm, the β rhythm exhibited a smaller degree of change.

In this study, researchers developed a coordinate system to assess the rate of change in α/β rhythms before and after stimulating the resting state (see Figure 8), and analyzed the characteristics of EEG rhythm responses to environmental factors across different scenes. The results indicated that the α rhythm exhibited the least increase in low-temperature environment, was centered in high-temperature environment, and showed the most significant increase in middle-temperature environment. Furthermore, when considering the scenes alone, the mean value of the rate of change in the α rhythm was highest in S1 (${\overline{\mathrm{A}}}_{\mathrm{S}1}$ = 0.46 > ${\overline{\mathrm{A}}}_{\mathrm{S}2}$ = 0.43 > ${\overline{\mathrm{A}}}_{\mathrm{S}3}$ = 0.41). The β rhythm did not demonstrate a significant association with temperature; however, it exhibited differential effects based on scene types, with S2 scene being more conducive to active the β rhythm (${\overline{\mathrm{B}}}_{\mathrm{S}2}$ = 0.10 > ${\overline{\mathrm{B}}}_{\mathrm{S}1}$ = 0.08 > ${\overline{\mathrm{B}}}_{\mathrm{S}3}$ = 0.06). Additionally, the largest growth in α rhythm was observed in M-S3 (A = 0.59), while the largest increase in β rhythm was observed in H-S2 (B = 0.129). Under the same energy supply, the maximum α increase was observed in the concrete scene, and the maximum β increase was observed in the steel scene.

3.5. Blood Oxygen Concentration and Heart Rate

The blood oxygen concentration and heart rate of all subjects were monitored under different environmental conditions using the O2 Ring device (Shenzhen Viatom Technology Co., Ltd., Shenzhen, China). The experimental data were averaged and compared across nine different environments during a resting state. First, repeated measures ANOVA was performed on the above data, corrected using Greenhouse-Geisser, and the results showed that blood oxygen concentration remained stable overall in the nine environments (p = 0.479), with no significant differences. Specifically, subjects exhibited the highest blood oxygen concentration in the low-temperature steel scene (L-S2: 96.723%) and the lowest in the medium-temperature concrete scene (M-S3: 96.308%) as shown in Figure 9. Notably, the heart rate parameters of the subjects demonstrated varying changes in response to environmental temperature (p = 0.003), as follows: high temperature zone (76.775 bpm) > medium temperature zone (75.157 bpm) > low temperature zone (70.388 bpm). As the temperature increased, the heart rate of the subjects also increased.

4. Discussion

This study aimed to quantify the psychophysiological impact of raw earth building environments compared to steel and concrete under varying thermal conditions, using the EEG + VR + LEC methodology developed by the team in the initial phase. Building upon this foundation, the study incorporated an olfactory dimension to enhance the immersive experience and explored the operational mechanisms between building spaces constructed from different materials and human perceptual feedback through quantitative experiments. A key finding is that the restorative effects of materials are highly context-dependent and linked to thermal conditions. The raw earth environment (S1) prompted the most significant increase in α rhythm activity—a reliable indicator of a relaxed mental state—under both low (22.8 °C) and high (30.1 °C) temperature conditions. This supports the central tenets of Biophilic Design and Stress Recovery Theory (SRT). The natural, unprocessed nature of raw earth may act as a “soft fascination,” as described in Attention Restoration Theory (ART), helping to reduce physiological stress responses, particularly when the body is experiencing thermal stress [9,10]. The material’s inherent hygrothermal regulation properties, even when simulated, may create a perceived sense of environmental stability and natural connection that is most potent when individuals are feeling physically uncomfortable (too cold or too warm) [16,17]. This suggests raw earth’s healing potential is not merely symbolic but is amplified through its interaction with physical stressors.

However, under neutral, mid-temperature conditions (26.5 °C), the concrete scene (M-S3) elicited the largest increase in α-rhythm, while the steel scene (H-S2) prompted the greatest increase in β-rhythm (associated with active concentration) [57]. Plausibly, in a thermally neutral state where cognitive load is minimal, the familiar and unadorned esthetic of a modern material like concrete may facilitate mental relaxation more efficiently than the complex sensory information processing required for natural textures. The steel environment, conversely, may foster a state of focused attention rather than relaxation. This indicates that the “best” material for well-being depends on both the desired psychological outcome and the ambient environmental conditions.

The positive correlation observed between thermal comfort votes (TCV) and all three brain rhythms (α, β, θ) further reinforces the profound link between thermal perception and overall cognitive-emotional state, aligning with prior studies by researchers like Yang et al. and Kim et al. [61,62]. Our study extends this work by demonstrating how specific materials mediate this relationship. The strong correlation between θ rhythm and perceptions of lighting and spatial permeability, particularly in the concrete and steel scenes, suggests that in the absence of strong natural material cues, occupants’ cognitive and emotional responses may be more heavily influenced by other architectural elements like spatial volume and daylighting.

Several limitations of this study must be acknowledged. First and foremost, the reliance on a homogeneous university student cohort significantly limits the generalizability of our findings. This sample, while advantageous for initial exploratory research due to its relative uniformity in age and cognitive function, is not representative of the broader urban population. Physiological and psychological responses to thermal and material stimuli are known to vary considerably across the lifespan. For instance, older adults may have different thermal comfort sensitivities and physiological regulatory mechanisms, while children’s attentional and emotional responses to spaces may differ fundamentally from those of young adults. Furthermore, individuals from diverse socioeconomic and cultural backgrounds may hold different perceptions and expectations of comfort and well-being in the built environment, which this study could not capture. Second, the experiment was conducted in a controlled VR environment, and while validated, it cannot fully replicate the long-term, multisensory experience of inhabiting a real building. Factors such as the subtle aging of materials and long-term adaptation were not captured. Finally, to maintain ecological validity, material choices were linked to corresponding window-to-wall ratios, introducing a potential confounding variable. While this reflects real-world design constraints, future research should employ factorial designs to isolate the distinct effects of material from those of daylighting and views.

Despite these limitations, the practical implications for architectural design are significant. This suggests a practical application in architectural design: a “material-thermal zoning” strategy. Architects could specify raw earth for restorative spaces (e.g., bedrooms, lounges) in buildings with passive thermal designs, where occupants are likely to experience greater temperature fluctuations. Conversely, in highly controlled office or learning environments where stable temperatures are maintained and cognitive focus is paramount, concrete and steel may be more appropriate. This study provides an empirical basis for moving beyond a purely esthetic application of biophilic principles toward a performance-based approach that optimizes material selection for both human well-being and energy efficiency. Future research could integrate psychophysiological feedback on material perception into generative architectural design systems [53,67], enabling AI-driven design tools to optimize both spatial layout and material selection based on human well-being metrics.

5. Conclusions

This study investigated the restorative potential of raw earth architecture by analyzing human psychophysiological responses to different materials under controlled thermal conditions. The study collected 90 multidimensional perceptual datasets based on three scenes and three temperature intervals to examine the physiological and psychological differences experienced by participants when temperature interacts with material types. The following conclusions were drawn from the analysis:

(1)

During the transition from the stimulated state to the resting state, both the α and β rhythms exhibited an ascending trajectory, culminating at a peak one minute subsequent to the onset of the resting state. Thereafter, these rhythms exhibited a tendency towards stabilization. Notably, the α rhythm responded more rapidly and demonstrated a greater increase (α maximal increase: 46%; β maximal increase: 9%), indicating that the experimental design scenes were effective in promoting the mental relaxation of the subjects. In contrast, the θ rhythm displayed the smallest fluctuation amplitude and the longest fluctuation period, showing a gradual recovery trend after an initial decrease of approximately 5% at the onset of the state transition.

(2)

Correlations between EEG rhythms and subjective questionnaires related to different scenes were analyzed. Overall, the thermal comfort vote exhibited positive correlations with both α/β/θ rhythms. This indicates that an optimal thermal environment may promote relaxation and enhance cognitive functions in the brain. Notably, the θ rhythm demonstrated strong correlations with lighting comfort vote (R = 0.54), thermal sensitive vote (R = 0.43), and spatial permeability vote (R = 0.47). This indicates that suitable light and thermal environment may enhance θ rhythm activity.

(3)

By comparing the dynamic responses of EEG rhythms during the transition from stimulus to rest, it was found that temperature significantly influenced changes in the α rhythm. Middle temperatures were found to activate the α rhythm the most, followed by high temperatures, while low temperatures had the least effect. This suggests that middle- and high-temperature environments are more conducive to promoting emotional recovery. Additionally, the raw earth scene (S1) induced the most pronounced changes in α rhythm under both low and high-temperature conditions, with L-S1 reaching a maximum of 33.2% and H-S1 reaching a maximum of 55.1%. This demonstrates the neuroadaptive advantages of raw earth materials in passive thermoregulation.

(4)

By establishing a coordinate system to analyze the rates of change in α/β rhythms before and after stimulating the resting state, it was found that the mean value of the α rhythm change rate was highest in S1 when evaluated solely from the perspective of the scene (${\overline{\mathrm{A}}}_{\mathrm{S}1}$ = 0.46 > ${\overline{\mathrm{A}}}_{\mathrm{S}2}$ = 0.43 > ${\overline{\mathrm{A}}}_{\mathrm{S}3}$ = 0.41). This finding implies that raw earth offers superior emotional restoration compared to steel or concrete under the tested conditions. Although not significantly correlated with temperature, β rhythms were influenced by the type of scene, with S2 scenes promoting a more active β rhythm (${\overline{\mathrm{B}}}_{\mathrm{S}2}$ = 0.10 > ${\overline{\mathrm{B}}}_{\mathrm{S}1}$ = 0.08 > ${\overline{\mathrm{B}}}_{\mathrm{S}3}$ = 0.06). Raw earth was not dominant across all temperature environments; the most significant increase in α rhythm was observed in the concrete scene, and the largest increase in β rhythm occurred in the steel scene under the same level of energy consumption.

The primary innovation of this study is the empirical demonstration that the healing effects of building materials are not inherent, static properties but are instead dynamically modulated by the thermal environment. Under low/high-temperature conditions, raw earth can induce the greatest emotional recovery, whereas traditional materials such as concrete and steel elicit better recovery or focused responses under thermally neutral conditions. In summary, by elucidating the synergistic relationship between materials and temperature, this study provides a robust scientific framework for optimizing mental health-friendly built environments. As urbanization continues, the strategic use of materials like raw earth will be crucial not only for their low environmental impact but also as a key medium for alleviating the “nature deficit disorder” associated with urbanization.