📄 earth_science

Soundscape categories shape recreational responses across independently characterized bamboo forest settings: a two-phase audio– visual experiment

This two-phase audio–visual experiment demonstrates that acoustic conditions significantly shape recreational responses in bamboo forests, with natural sounds eliciting the highest subjective relaxation and pleasure while traffic noise produces the strongest negative effects, although these subjective preferences show weak correspondence with EEG measures and are largely mediated by perceived pleasantness and annoyance.

Original authors: Hao Li, Yi Peng, Zongsheng Li, Xinyu Du, Chenmingyang Jiang, Qibing Chen, Bingyang Lv, Cong Ma

Published 2026-08-14

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Hao Li, Yi Peng, Zongsheng Li, Xinyu Du, Chenmingyang Jiang, Qibing Chen, Bingyang Lv, Cong Ma

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you're walking through a beautiful forest. Your eyes see towering trees and dappled sunlight, but your ears are just as busy. Science tells us that our brains don't just process what we see; they blend sight and sound into one big experience. This field of study is called soundscape research. It's not just about how loud something is (decibels); it's about what that sound means to us. Is it the chirp of a bird, which feels peaceful? The hum of a conversation, which feels social? Or the roar of a car, which feels intrusive? Researchers care about this because forests aren't just for timber anymore; they are places for people to relax, play, and recharge. If a forest looks perfect but sounds like a highway, can it still help you feel calm? That's the big question this study tackles.

To answer this, a team of researchers from Sichuan Agricultural University decided to play a high-tech game of "match the sound to the scene." They wanted to see if the type of noise in a bamboo forest changes how much people enjoy it, regardless of how pretty the bamboo looks. They set up a clever two-part experiment. First, they took photos of 50 different bamboo spots and had people rate how "restorative" or relaxing they looked. This gave them a solid baseline for the visual quality of each spot. Then, in the second phase, they brought 113 volunteers into a lab. These volunteers wore headphones and watched videos of those same bamboo spots, but with a twist: the original sound was muted, and they played one of three types of audio over the video. The sounds were: Nature Sounds (birds, wind, rustling leaves), Human Activity (people chatting, footsteps, playing), and Traffic Noise (cars, horns, trains).

While the volunteers watched, the researchers didn't just ask them how they felt; they also hooked them up to an EEG machine. This device measures brain waves, specifically looking at the ratio of "beta" waves (often linked to active thinking or alertness) to "alpha" waves (linked to relaxation). The goal was to see if the brain's electrical signals matched what the people said they felt.

The results were clear and surprisingly consistent. When it came to how people felt (their relaxation, pleasure, and preference), the order was exactly what you'd expect: Nature sounds were the best, Human sounds were in the middle, and Traffic noise was the worst. Specifically, compared to nature sounds, human activity sounds reduced relaxation scores by about 17.5 points and pleasure by 18 points on a 0–100 scale. Traffic noise was a real mood-killer, dropping relaxation and pleasure by over 41 points and preference by nearly 47 points.

However, the brain waves told a slightly different story. The EEG data showed the reverse pattern: the brain was more active (higher beta/alpha ratio) when listening to traffic, then human sounds, and least active when listening to nature. This suggests that while traffic noise makes us feel terrible, it also keeps our brains in a state of high alert or stress, whereas nature sounds allow the brain to truly "power down."

Here is the most interesting part: the researchers tested if a really beautiful view could "fix" a bad sound. They found that even if the bamboo forest looked stunning, it couldn't cancel out the negative effects of traffic noise. The bad feelings from the traffic sound persisted, no matter how pretty the trees were. This means that a forest manager can't just plant pretty trees to hide a noisy road; they have to deal with the noise itself.

The study also looked at whether age or how anxious a person felt before the experiment changed the results. They found no strong evidence that older people or younger people reacted differently, nor did pre-existing anxiety levels change the outcome. The bad effect of traffic and the good effect of nature seemed to hit everyone pretty much the same way.

One final twist: the brain waves and the people's feelings didn't always match up perfectly on a trial-by-trial basis. Just because a person's brain showed a little more activity didn't always mean they said they felt less relaxed. This suggests that while brain scans are useful for seeing general trends, they aren't a perfect "lie detector" for how much someone is enjoying a moment.

In short, this paper suggests that for bamboo forests to be truly great places for recreation, they need to be quiet. A beautiful view is not enough to save a noisy experience. Nature sounds are the gold standard for relaxation, human sounds are an acceptable middle ground, but traffic noise is a major barrier to enjoying the forest, regardless of how green the leaves look.

Technical Summary: Soundscape Categories Shape Recreational Responses Across Independently Characterized Bamboo Forest Settings

Problem Statement
While forests are increasingly managed for cultural ecosystem services such as recreation and psychological well-being, assessments of forest experience often prioritize visible stand characteristics over acoustic conditions. Existing research frequently relies on visual stimuli (photographs or videos) without adequately characterizing the visual context, or it fails to separate auditory and visual contributions when both are assessed concurrently. Field studies often suffer from confounding variables where vegetation structure, traffic exposure, and sound-source composition vary simultaneously. Consequently, there is a need to determine whether acoustic conditions represent a distinct component of recreational quality that persists independently of visual stand structure, specifically within bamboo forest settings which offer a coherent vegetation identity but structurally heterogeneous environments.

Methodology
The study employed a rigorous two-phase experimental design with independent participant samples to decouple visual context characterization from audio–visual outcome measurement.

Phase 1 (Visual Characterization): Researchers collected 50 matched photograph–plot units from 10 distinct visual blocks across five bamboo-space types (ecological conservation, productive–economic, protective–greenbelt, landscape–recreational, and understory–composite). Participants rated relaxation, pleasure, and preference for the photographs. Field measurements provided structural indicators. Based on these independent findings, three block-level visual descriptors were derived: VisualBase (mean relaxation rating), vertical stratification (VS), and groundcover coverage (GC).
Phase 2 (Audio–Visual Experiment): A separate sample of 113 adults participated in a laboratory-based incomplete repeated-measures experiment. They were exposed to 50-second video clips (derived from the Phase 1 locations) paired with one of 18 specific soundtracks. The soundtracks belonged to three categories: Natural sounds (NAT), Human-activity sounds (HUM), and Traffic noise (TRA), with six distinct soundtracks per category.
Procedure: Each participant completed six trials, receiving two exposures from each soundscape category. Outcomes included subjective ratings (relaxation, pleasure, preference, auditory pleasantness, and annoyance) on 0–100 Visual Analogue Scales (VAS) and electroencephalography (EEG) data. The primary EEG metric was the change in the beta-to-alpha power ratio ( $\Delta(\beta/\alpha)$ ) from baseline to exposure.
Statistical Analysis: Linear mixed-effects models were used to analyze the 678 observations. Models included crossed random intercepts for participants and visual blocks (or specific soundtracks for EEG) and adjusted for trial position and the Phase 1-derived visual covariates. Robustness was tested through sensitivity analyses, including leave-one-out stimulus omission, alternative visual specifications, bootstrap inference, and hierarchical modeling of participant-specific slopes.

Key Contributions

Independent Visual Characterization: By deriving visual context measures from an independent sample (Phase 1) and applying them as external descriptors in the outcome phase (Phase 2), the study minimized common-source bias and allowed for a cleaner attribution of soundscape effects.
Granular Soundscape Categorization: The study distinguished human-activity sounds from traffic noise, treating them as distinct categories rather than a monolithic "anthropogenic" group.
Multi-Modal Robustness: The integration of subjective ratings with physiological EEG data, combined with extensive sensitivity analyses (including CR2 inference and bootstrap methods), provides a comprehensive view of the stability of soundscape effects.

Results

Subjective Responses: Subjective ratings of relaxation, pleasure, and preference followed a consistent ordering: NAT > HUM > TRA. Compared to NAT, HUM reduced relaxation, pleasure, and preference by approximately 17–22 points, while TRA reduced them by 41–46 points. All contrasts remained significant after Holm adjustment.
EEG Responses: The EEG $\Delta(\beta/\alpha)$ showed the reverse ordering: NAT < HUM < TRA. Both HUM and TRA were associated with significantly higher $\Delta(\beta/\alpha)$ values compared to NAT, indicating increased relative cortical activation.
Visual Context Independence: The soundscape contrasts remained highly stable (changes < 1%) after adjusting for VisualBase, vertical stratification, and groundcover coverage. No significant interactions were found between soundscape categories and visual context descriptors.
Appraisal Overlap: When concurrent auditory pleasantness and annoyance ratings were added to the models, the subjective contrasts were substantially attenuated. The HUM–NAT contrasts were reduced by 95.2%–99.6% (becoming non-significant), while TRA–NAT contrasts were reduced by 81.3%–90.0% but remained negative. This indicates a strong statistical overlap between immediate appraisal and subjective response, though not necessarily mediation.
EEG-Subjective Correspondence: While category-level trends were opposite, the correspondence between EEG and subjective responses at the trial or participant level was weak. No significant within-participant or between-participant associations survived multiplicity correction.
Moderation: Neither age nor pre-exposure state anxiety (STAI-S) showed multiplicity-adjusted evidence of moderating the soundscape contrasts.

Significance and Claims
The paper claims that acoustic conditions constitute a distinct and independent component of recreational quality in bamboo forests, separate from visual stand structure. The findings suggest that favorable visual conditions (e.g., high vertical stratification or groundcover) cannot compensate for the adverse effects of traffic noise.

The study concludes that forest management should not treat all non-natural sounds as a single disturbance category. Instead, a differentiated approach is warranted:

Traffic Noise: Represents a distinct adverse factor that degrades recreational quality regardless of visual stand quality.
Human-Activity Sounds: Produce an intermediate profile, less favorable than natural sounds but significantly less adverse than traffic noise. These sounds may indicate legitimate social use and accessibility and should not be managed equivalently to external mechanical noise.

The authors emphasize that their findings are specific to the tested soundtrack sets and bamboo-space visual contexts. They do not claim a universal ranking for all natural or anthropogenic sounds but provide robust evidence that the observed contrasts persist across independently characterized visual heterogeneity. The study advocates for the inclusion of acoustic-source composition alongside stand structure in the assessment and management of forest cultural ecosystem services.

Technical Summary: Soundscape Categories Shape Recreational Responses Across Independently Characterized Bamboo Forest Settings

More like this