2 - (SYM 24) Affective Responses to Semantic Sound Suppression in Adults with Misophonia

Misophonia is a newly recognized disorder characterized by behavioral, attentional, somatic, interpersonal, and/or cognitive responses to specific aversive sounds (e.g., chewing, throat-clearing, sniffling, tapping, clicking) and associated stimuli, resulting in significant distress and functional impairment. Because scientific knowledge and clinical guidance remain limited, individuals often rely on coping strategies such as noise cancellation or masking sounds to manage their symptoms. However, traditional noise-cancellation approaches can create challenges, as there is still a need to hear other non-triggering environmental cues to effectively engage in daily life.

This study evaluated the feasibility and acceptability of a neural network-based semantic sound suppression model designed to selectively suppress targeted auditory cues based on their semantic class. Adults who identified as having misophonia consistent with the consensus definition, reported eating/chewing as a primary trigger, and met the clinical cut-off on the Duke Misophonia Questionnaire (DMQ) were eligible. Participants were presented with ten audio samples: five originals containing a mixture of environmental sounds including eating/chewing sounds, and five corresponding suppressed versions with the selected trigger sounds removed. They rated their responses before and after listening to each sample and provided feedback on the technology.

Significant within-participant differences emerged across all outcomes. Participants reported lower distress and arousal and higher valence when listening to the suppressed audio samples compared to the originals. Participants also rated the sound suppression technology as highly feasible and acceptable, with ratings well above the scale midpoint. These findings support further exploration and development of personalized, real-time sound-modification interventions for individuals with misophonia.