Despite the ambiguous many-to-many mapping between acoustic signals and target phonemes, human listeners quickly overcome and adapt to inconsistencies during speech perception. However, the processing cost of speech perception increases when a change of the talker occurs and preceding context speech was found to reduce the processing cost. The magnitude of the response time difference between the single- and mixed-talker condition is called the interference effect. The literature indicates that that the interference effect is reduced by increasing length of speech context preceding target speech, but the quantity of the information embedded in that speech context does not further impact processing cost. In this study, we further explored the relationship between the duration of the preceding speech context and its facilitative effect in talker adaptation. The results indicated that even though response times were always shorter in the single-talker condition than the mixed-talker condition, the facilitative effect of preceding context speech became constant for durations longer than beyond 600ms, rather than eventually eliminating the interference effect at some carrier phrase of sufficient duration. Therefore, wo mechanisms are proposed to subserve talker adaptation: a feed-forward, extrinsic process that reduces the interference effect by integrating prior speech context of up to 600ms, and a top-down, intrinsic process that unfolds over longer timescales by allocating cognitive resources to cope with potential talker variability, leading to a global processing time increase.
| Identifer | oai:union.ndltd.org:bu.edu/oai:open.bu.edu:2144/38124 |
| Date | 08 September 2019 |
| Creators | Kou, Sio Nga |
| Contributors | Perrachione, Tyler K. |
| Source Sets | Boston University |
| Language | en_US |
| Detected Language | English |
| Type | Thesis/Dissertation |
Page generated in 0.0019 seconds