UBC Theses and Dissertations
Identification of environmental sounds Spanik, Christiane Susan
The purpose of this study was to investigate the processing of environmental sound, and to compare this to the processing of spoken language. This was done by conducting an identification experiment using the gating paradigm to assess the on-line processing of environmental sounds. In this experiment, twenty participants identified brief, isolated segments of environmental sounds and reported a confidence rating for their responses. Additional context was provided by gating the original recordings in both the preceding and following directions. Eight different recordings of environmental sound were presented, with each listener hearing additional segments of each recording in either the preceding or following direction. It was found that identification performance improved with addition of context in both directions of context, suggesting that as with language, there are context effects in environmental sound processing. Unlike language, however, there is no clear effect of direction of context. There was some evidence that high-context sound sequences were more easily identified than low-context sound sequences, but these results were not consistent. An error analysis provided strong evidence that the process of environmental sound identification entails the activation of a cohort, which is compelling evidence that there may be some similarities in the auditory processing language and environmental sound. It appears that item identification in these two types of auditory input may be similar, with bottom-up processes activating an initial cohort based primarily on acoustic information, and top-down processes narrowing the size of the cohort as more information becomes available.
Item Citations and Data