The occlusion effect is typically described as an amplified and distorted perception of physiological sounds of one’s own body, such as speech, chewing, and heartbeat. In many cases, the occlusion effect induced by hearing protection devices leads to acoustical discomfort that can result in increased noise exposure due to improper use or removal of the protector by the wearer as a way to alleviate discomfort. Despite numerous studies over the past decades, several aspects relating to the measurement of the occlusion effect and the assessment of the discomfort it causes remain to be investigated.
In this thesis, the occlusion effect caused by earplug-type hearing protection devices is investigated to achieve a better understanding of the discomfort it produces through an experimental study with human participants in laboratory conditions. For this purpose, this thesis focuses on (i) proposing a methodology for measuring the objective occlusion effect induced by speech, (ii) identifying and prioritizing the physical and psychosocial characteristics of the environment–person–protector triad that influence the discomfort experienced by a user, and (iii) objectivizing the occlusion effect experienced by a user using indicators derived from microphonic measurements. In addition, the formulation of questions used to evaluate occlusion effect-related discomfort, as well as the methodology for its assessment, are also investigated.
The data necessary to achieve these objectives were collected through two measurement campaigns conducted in laboratory conditions with 63 human participants, along with a survey of 21 respondents.
In the first part, various aspects related to the objective measurement of the occlusion effect are examined, namely the indicator, the stimulation source, and the measurement method. The quantification of the objective occlusion effect based on frequency-dependent indicators remains the most suitable approach to analyze the phenomenon across the frequency range. However, single-value indicators based on the averaged 160–500 Hz band allow trends to be observed, are simple to compute, and show lower variability than others. Moreover, inducing the occlusion effect through continuous speech enables easy and reproducible measurement of the objective occlusion effect, while remaining representative of the situation for which this discomfort is generally reported by users. Finally, a method based on simultaneous measurements inside and outside the occluded earcanal, analogous to the noise reduction method for sound attenuation assessment, allows from a single measurement the derivation of an objective occlusion effect indicator in each ear independently, without the need for an open-ear measurement.
In the second part, key aspects related to assessing discomfort through a questionnaire are highlighted. Although various terms can be used to assess discomfort in French, the word “gêne” appears suitable and well understood, provided that its definition is given to participants before testing. The comparison between the “experienced occlusion effect” and the “perceived occlusion effect” revealed significant differences on the rating scale, highlighting the importance of clear, precise, and easily understandable questions when evaluating discomfort.
In the third part, the most influential characteristics of the environment–person–hearing protection device triad are identified using a statistical approach based on mixed linear models. User-related factors included acquaintance with the experimenter (associated with higher discomfort), non-circular earcanal morphology at the second bend (associated with greater discomfort), and hearing loss at 250, 500, and 1000 Hz (which influenced discomfort to a lesser degree). The earplug model did not significantly affect discomfort, whereas environmental background noise level significantly increased discomfort. Regarding the interaction between the user and the protector, insertion depth had a significant effect, with deeper insertions leading to less discomfort than shallow ones.
In the fourth part, discomfort associated with the occlusion effect is objectivized using a statistical approach based on mixed linear models. When relying solely on microphonic-based indicators, discomfort was linked to increasing noise level in the occluded ear within the 125 Hz octave band and to the overall background noise level of the test environment. However, this first model presented low statistical power and limited predictive accuracy. When the influential characteristics of the environment–person–hearing protection device triad were combined with these objective indicators, the model’s statistical power and predictive accuracy improved. This second model also highlighted the stronger influence of users’ physical and psychosocial characteristics compared to objective indicators.
The work conducted in this thesis contributes to a deeper understanding of the discomfort associated with the occlusion effect induced by earplug-type hearing protection devices and proposes methods to improve its measurement using microphones and its assessment using questionnaires. The results of this work will also contribute to the development of earplugs that generate less occlusion effect and are therefore more comfortable, as well as to improved earplug selection in the field by taking this aspect of comfort into account.
| Date | 16 Dec 2025 |
|---|
| Original language | American English |
|---|
| Awarding Institution | - École de technologie supérieure
|
|---|
| Supervisor | Olivier Doutres (Supervisor) & Hugues Nélisse (Co-supervisor) |
|---|