|
The dynamics of spoken-word recognition in context| old_uid | 9082 |
|---|
| title | The dynamics of spoken-word recognition in context |
|---|
| start_date | 2010/09/24 |
|---|
| schedule | 11h-12h30 |
|---|
| online | no |
|---|
| summary | Theories of speech comprehension have assumed that the recognition of spoken words consists of mapping the speech signal onto context-free representations. These representations, it is argued, abstract from most of the phonetic consequences of utterance context, such as prosodic structure or talker identity. However, my work using listeners’ saccadic eye movements to visual objects during speech comprehension shows that listeners utilize (rather than neutralize) the acoustic correlates of the context in which a spoken word occurs. Word forms are not abstract phonological prototypes, nor are they collections of past exemplars. Instead, I argue, the forms that listeners expect words to take are dynamically updated to their current context. |
|---|
| responsibles | Rämä |
|---|
| |
|