Sentence Processing in Traumatic Brain Injury: Evidence from the P600

• Published in: Journal of speech, language, and hearing research language, and hearing research, 2016-08, Vol.59 (4), p.759-771
• Main Author: Key-DeLyria, Sarah E
• Format: Article
• Language: English
• Subjects: Adult; Agreements; Aphasia; Behavior; Brain; Brain - physiopathology; Brain Hemisphere Functions; Brain Injuries, Traumatic - physiopathology; Brain research; Cognition - physiology; Cognitive Measurement; Cognitive Tests; Communication Problems; Communication Skills; Comprehension; Comprehension - physiology; Control Groups; Electroencephalography; Evoked Potentials; Executive Function; Female; Head Injuries; Humans; Injuries; Intellectual Disability; Judgment - physiology; Language; Language Impairments; Language Processing; Linguistics; Listening comprehension; Male; Neurological Impairments; Neuropsychological Tests; Nonparametric Statistics; Pattern Recognition, Visual - physiology; Reaction Time; Reading; Reading Comprehension; Reading Difficulties; Sampling; Semantics; Sentences; Signal Processing, Computer-Assisted; Speech production; Statistical Inference; Stimuli; Traumatic brain injury; Violations; Written Language; Young Adult; Young Adults
• ISSN: 1092-4388; 1558-9102
• DOI: 10.1044/2016_JSLHR-L-15-0104

Purpose: Sentence processing can be affected following a traumatic brain injury (TBI) due to linguistic or cognitive deficits. Language-related event-related potentials (ERPs), particularly the P600, have not been described in individuals with TBI history. Method: Four young adults with a history of closed head injury participated. Two had severe injuries, and 2 had mild-moderate injuries more than 24 months prior to testing. ERPs were recorded while participants read sentences designed to be grammatically correct or incorrect. Participants also completed cognitive and sentence comprehension measures. Results: One participant with TBI was significantly different than the control group on several behavioral sentence measures and 1 cognitive measure. However, none of the participants with TBI had a reliable P600 effect. Nonparametric bootstrapping indicated that the ERP was reliable in 10 control participants but no participants with TBI history. Conclusions: There were few behavioral differences between individuals with TBI history and the control group, though all reported subjective difficulty with reading. The P600 was absent in the TBI group in this study. Given the heterogeneity of individuals with TBI and the difficulty in assessing subtle language impairments, exploring the P600 further may provide useful insight into language processing difficulties.