Myrna F. Schwartz, Linguistics Colloquium: Spontaneous speech-error monitoring and incremental lexical learning

Colloquium | November 20 | 3:10-5 p.m. | 370 Dwinelle Hall

 Myrna F. Schwartz, Institute Scientist Emerita, Moss Rehabilitation Research Institute

 Department of Linguistics

Clinicians have long appreciated that people with aphasia (PWA) who self-monitor their errorful speech have better outcomes. It has been assumed, with empirical justification, that this is because successful monitoring is associated with a stronger language system and a strong system is more likely to improve (Strength hypothesis). A second possibility, not incompatible with the first, is that monitoring success in some way causes those language systems to undergo adaptive change (Learning hypothesis). My talk will review recent evidence from PWA that supports both the Strength and Learning hypotheses. We obtained novel evidence that errorful items that were spontaneously self-corrected had a higher probability of being named correctly on a future occasion (forward prediction). The effect was also significant in the backwards direction, which is evidence for the Strength hypothesis. The evidence for learning – restricted to semantic errors – was that self-correction produced bigger change in the forward than backwards direction. This evidence, along with a follow-up study of monitoring latencies, will be discussed within an explanatory framework that integrates aspects of monitoring theory with models of lexical access and incremental learning.

 terry.regier@berkeley.edu