Clustering individuals based on performance in audio-visual testing

Audio-visual interactions play a crucial role in shaping individuals' perception of their surroundings and have been proposed as an effective strategy for improving environmental noise perception.However, the relative influence of visual and auditory inputs, as well as the degree of audio-visual integration, will vary across individuals.To investigate these influences, a dataset of 196 valid participations in an audio-visual performance test was collected under controlled laboratory conditions.Using Uniform Manifold Approximation and Projection (UMAP) for dimensionality reduction and k-means clustering, three distinct participant groups were identified based on six relative performance indicators.These clusters were categorized as visual-first responders (cluster A), balanced integrators (cluster B) and visually dominated individuals (cluster C).Visual-first responders and the visually dominated group show hardly any audio-visual integration in the audio-visual acuity test, while the balanced integrators clearly do.Notably, clusters A and C, accounting for 68% of the test population, are expected to benefit most from noise annoyance mitigation strategies that incorporate green window views.