Does naming accuracy improve through self-monitoring of errors?
| Autor: | Erica L. Middleton, Adelyn Brecher, Maureen Gagliardi, Kelly Garvey, Myrna F. Schwartz |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2016 |
| Předmět: |
Male
Cognitive Neuroscience Feedback Psychological Experimental and Cognitive Psychology Semantics 050105 experimental psychology Article 03 medical and health sciences Behavioral Neuroscience 0302 clinical medicine Speech Production Measurement Phonetics Aphasia medicine Humans Learning Speech 0501 psychology and cognitive sciences Psychological testing Aged Psychological Tests 05 social sciences Linear model Regression analysis Middle Aged Random effects model Self Concept Self-monitoring Linear Models Female medicine.symptom Psychology Error detection and correction Social psychology 030217 neurology & neurosurgery Cognitive psychology |
| Popis: | This study examined spontaneous self-monitoring of picture naming in people with aphasia. Of primary interest was whether spontaneous detection or repair of an error constitutes an error signal or other feedback that tunes the production system to the desired outcome. In other words, do acts of monitoring cause adaptive change in the language system? A second possibility, not incompatible with the first, is that monitoring is indicative of an item’s representational strength, and strength is a causal factor in language change. Twelve PWA performed a 615-item naming test twice, in separate sessions, without extrinsic feedback. At each timepoint, we scored the first complete response for accuracy and error type and the remainder of the trial for verbalizations consistent with detection (e.g., “no, not that”) and successful repair (i.e., correction). Data analysis centered on: (a) how often an item that was misnamed at one timepoint changed to correct at the other timepoint, as a function of monitoring; and (b) how monitoring impacted change scores in the Forward (Time 1 to Time 2) compared to Backward (Time 2 to Time 1) direction. The Strength hypothesis predicts significant effects of monitoring in both directions. The Learning hypothesis predicts greater effects in the Forward direction. These predictions were evaluated for three types of errors – Semantic errors, Phonological errors, and Fragments – using mixed-effects regression modeling with crossed random effects. Support for the Strength hypothesis was found for all three error types. Support for the Learning hypothesis was found for Semantic errors. All effects were due to error repair, not error detection. We discuss the theoretical and clinical implications of these novel findings. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|