Abstract
BACKGROUND: Accent detection studies have shown native listeners to be highly sensitive to the presence of nonnative accents. This study examined the robustness of this sensitivity. METHODS: We evaluated listeners' accent discrimination performance when presented with a stimulus set consisting of multiple nonnative accents, as well as words and sentences that were unique in each trial. Listeners heard pairs of talkers reading the same word or sentence and indicated whether the talkers' native languages were the same or different. Talkers included two native talkers and six nonnative talkers from three native language backgrounds. RESULTS: Listeners were highly sensitive to the difference between native and nonnative accents, confirming earlier findings, but were much less sensitive to the difference between two nonnative accents. Furthermore, while stimulus length affected listeners' sensitivity to the difference between native and nonnative accents, this factor had a minimal effect on their sensitivity to the difference between two nonnative accents. CONCLUSION: The findings suggest that task and stimulus characteristics have a significant effect on the degree of sensitivity to nonnative accents.