Backward recognition masking of similar and dissimilar environmental sounds

Abstract

Backward Recognition Masking (BRM) of sound occurs when two sounds are presented close in time to one another and the second sound hinders the recognition of the first sound. Previous studies on BRM used either white noise or sine tones. Here, we present backward recognition thresholds using environmental sounds. 34 young normal hearing individuals were presented with 6 combinations of similar or dissimilar environmental sounds from the categories of household appliances, automobiles, and power tools. The sounds were selected based on their multi-dimensional scaling distances (Rosen et al., 2017). The sounds were presented at 20 dB SL re: 3 frequency PTA and the inter stimulus interval was fixed at 22.2 ms. A two-down one-up adaptive procedure was used to identify the target duration at which the listeners could identify that the target is different from the masker. Initial data analyses indicated that similar sounds needed significantly longer presentation durations to be identified as different compared to dissimilar sounds, indicating more backward masking for similar sounds compared to dissimilar sounds. These results potentially highlight how critical safety information in real-world environments may be missed in complex listening scenarios.