Effects of Soundscape Complexity on Urban Noise Annoyance Ratings: A Large-Scale Online Listening Experiment

Andrew Mitchell, Mercede Erfanian, Christopher J. Soelistyo, Tin Oberman, Jian Kang, Robert W Aldridge, Jing‐Hao Xue, Francesco AlettaInternational Journal of Environmental Research and Public Health 2022 · DOI

TL;DR

This study reveals that the number of distinct sounds in an urban environment, not their specific combinations, is a key driver of noise annoyance. Interestingly, annoyance is minimized with two distinct sounds, increasing with either too few (monotony) or too many (chaos), and traffic sounds consistently increase annoyance while nature sounds decrease it.

What Problem Does This Paper Address?

Environmental noise is a significant public health concern, contributing to various health issues and widespread annoyance. While traditional research has focused on the impact of specific noise sources like traffic or aircraft, and their overall sound levels, there's a growing recognition that urban acoustic environments are complex soundscapes. This complexity, involving multiple sound sources, their interactions, and their semantic meaning, likely plays a crucial role in how people perceive and react to noise. However, previous studies haven't fully explored how the number of distinct sound sources (soundscape complexity) or the combinations of these sources, beyond individual source effects, influence noise annoyance. This paper aims to fill this gap by investigating these nuanced aspects of urban soundscapes.

Methods

The researchers conducted a large-scale online listening experiment using the Gorilla Experiment Builder platform. Participants first underwent a browser and headphone check to ensure audio quality. The core of the experiment involved participants listening to 15-second binaural audio recordings of urban soundscapes. For each recording, they first performed a sound source recognition task, selecting from a list of 24 labels. Following this, they rated their perceived noise annoyance on a 10-point scale. To analyze the data, they employed hierarchical agglomerative clustering and association rule learning to explore source combinations. They also categorized recordings by the number of identified sound sources to assess "soundscape complexity" and conducted independent-samples t-tests to examine the effect of individual sound sources.

Datasets

The study utilized audio recordings from the International Soundscape Database v0.5 (ISD) [29,33]. This database comprises 2890 audio recordings of urban acoustic environments, originally collected in London and Venice. For this study, the original recordings were segmented into 15-second chunks. The data from the experiment, including sound recognition and annoyance ratings, is openly available on Zenodo [37].

Key Insights and Ideas

  • Complexity Over Combination: The specific mix of sounds matters less than the sheer number of distinct sounds present. This challenges the idea that certain combinations might mitigate or exacerbate annoyance beyond the sum of their parts.
  • The "Sweet Spot" of Two Sounds: Noise annoyance appears to be minimized when there are exactly two clearly distinguishable sound sources. This suggests a balance between monotony and overwhelming chaos.
  • Nature as an Annoyance Buffer: The presence of nature-related sounds (like bird tweets or rustling leaves) consistently decreases annoyance, even when more annoying sounds (like traffic) are also present. This highlights the potential restorative power of natural elements in urban soundscapes.
  • Traffic's Unwavering Annoyance: Conversely, traffic-related sounds reliably increase annoyance, regardless of other sounds present. This reinforces their well-established negative impact.
  • The "Non-Identifiable" Effect: Interestingly, "non-identifiable" sound sources were associated with lower annoyance. This could be interpreted as a sign of a lively, albeit indistinct, urban buzz, potentially perceived positively in the context of reduced social interaction during the COVID-19 pandemic.

Key Results

  • Soundscape Complexity: A second-degree polynomial regression showed that annoyance scores were highest with 0 or 8 identified sound sources, and lowest with 2 identified sound sources. The ANOVA confirmed a significant effect of soundscape complexity on annoyance (F(8, 2881) = 15.354, p < 0.001).
  • Individual Sound Source Effects: Independent t-tests revealed that most sound sources, particularly traffic-related ones (Aircraft, Bus, Car, General traffic, Horn, Motorcycle, Rail, Screeching brakes, Siren), significantly increased annoyance when present compared to when absent. Conversely, nature-related sources (Bird tweet, Rustling leaves) and "Non-identifiable" sources significantly decreased annoyance.
  • Combinations Not Significant: Clustering and association rule learning analyses did not yield statistically significant differences in annoyance scores based on sound source combinations.

Limitations and Future Work

The authors acknowledge that while their online experiment offers ecological validity by using real urban soundscapes and a large, diverse participant pool, it's still a semi-controlled environment. They also note that the definition of "annoyance" can be multifaceted and that their study focused on "relative annoyance" within the experimental context.

For future research, they suggest focusing on models that incorporate sound sources with higher "ecological frequency" (i.e., those commonly encountered in everyday life). They also highlight the ongoing efforts to bridge theoretical stances on noise annoyance, particularly between community noise assessment standards and soundscape approaches, which could lead to a more unified understanding of this complex perceptual construct. The finding that only 7-8 sources could be reliably identified by participants also offers empirical support for the ISO/TS 12913-2:2018 protocol's implicit limit on the number of sound sources to consider.

Summary updated

Abstract

Noise annoyance has been often reported as one of the main adverse effects of noise exposure on human health, and there is consensus that it relates to several factors going beyond the mere energy content of the signal. Research has historically focused on a limited set of sound sources (e.g., transport and industrial noise); only more recently is attention being given to more holistic aspects of urban acoustic environments and the role they play in the noise annoyance perceptual construct. This is the main approach promoted in soundscape studies, looking at both wanted and unwanted sounds. In this study, three specific aspects were investigated, namely: (1) the effect of different sound sources combinations, (2) the number of sound sources present in the soundscape, and (3) the presence of individual sound source, on noise annoyance perception. For this purpose, a large-scale online experiment was carried out with 1.2k+ participants, using 2.8k+ audio recordings of complex urban acoustic environments to investigate how they would influence the perceived noise annoyance. Results showed that: (1) the combinations of different sound sources were not important, compared, instead, to the number of sound sources identified in the soundscape recording (regardless of sound sources type); (2) the annoyance ratings expressed a minimum when any two clearly distinguishable sound sources were present in a given urban soundscape; and (3) the presence (either in isolation or combination) of traffic-related sound sources increases noise annoyance, while the presence (either in isolation or combination) of nature-related sound sources decreases noise annoyance.
BibTeX
@article{mitchell2022effects,
  author = {Andrew Mitchell and Mercede Erfanian and Christopher J. Soelistyo and Tin Oberman and Jian Kang and Robert W Aldridge and Jing‐Hao Xue and Francesco Aletta},
  title = {{Effects of Soundscape Complexity on Urban Noise Annoyance Ratings: A Large-Scale Online Listening Experiment}},
  journal = {International Journal of Environmental Research and Public Health},
  year = {2022},
  doi = {10.3390/ijerph192214872},
}
Figure