Brain potentials reveal deficits of language processing after closed head injury.

Abstract

To delineate deficits in language processing after closed head injury with use of behavioral measures and event-related brain potentials. Case-control design. All subjects participated in three verbal event-related brain potential experiments, and the resulting measures were compared both within and between groups. PATIENTS/CONTROLS: Eleven patients at least 2 years after severe closed head injury (Glasgow Coma Scale score < 8 at admission and duration of posttraumatic amnesia > 48 hours) were compared with a control group matched for age and educational level. None. Reaction times and percentage correct as behavioral measures in the three experiments (sentence verification, semantic and repetition priming with lexical decision task, and continuous word recognition). Event-related brain potentials were quantified by area measures in successive time windows for the different experimental conditions and for different experiments. The reaction times of the patient group were significantly longer than those of the controls (P < .005). Similarly, the patients' accuracy was significantly worse in all experiments (P < .03). The event-related brain potentials of the controls showed a clear and significant reduction of a negative component (N400) to terminal words of true sentences (sentence verification experiment), semantically primed words and repeated words (lexical decision experiment), and recognized words (continuous word recognition). For the patients, a clear N400 effect was seen only in the sentence verification task (delayed by about 100 milliseconds), while only later event-related brain potential modulations were seen in the other tasks. Language functions are disturbed after closed head injury. The electrophysiologic data suggest difficulties in the integration of incoming linguistic stimuli with the previous context as a possible underlying cause.