Evaluation of mild traumatic brain injury: ERP processing in auditory and visual novelty oddball tasks

Abstract

Is automatic registration of deviance affected by mild traumatic brain injury (TBI)? Is working memory affected in recent survivors of mild TBI? Does stimulus modality play a role? To address these questions, we evaluated the P3a and P300 components of the eventrelated brain potential elicited during auditory and visual novelty oddball tasks. Participants were service members with mild TBI (mTBI), injured service members negative for TBI (injured controls, IC), and healthy civilians (HC). Standard (p=.60) and target (p=.20) stimuli were 2000 and 1940 Hz tones in the auditory task and two complex geometric patterns in the visual task. Novel stimuli (p=.20) comprised unique environmental sounds in the auditory task and unique images from the International Affective Pictures System in the visual task. Participants were instructed to press a button to target stimuli. Accuracy and speed of performance on the tasks did not differ among groups. A 62-channel electrode array was used to collect ERP data. Novel stimuli elicited a large P3a (mean over groups: 16.0 (mu)V at FCz in the auditory task and 7.6 (mu)V at Cz in the visual task). P3a latency averaged 326 ms and 392 ms in the auditory and visual tasks, respectively. Target stimuli elicited prominent P300 components in all groups in the auditory (11.7 (mu)V at CPz at a latency of 420 ms) and visual (16.6 (mu)V at CPz at a latency of 444 ms) tasks. Neither the amplitude nor the latency of P3a elicited by novel auditory or visual stimuli differed among the three groups. The amplitude of P300 in both tasks appeared to be reduced in the injured servicemembers, but the differences were not statistically significant. In contrast, there were large and significant (p=.021) group differences in the latency of P300 elicited by auditory and visual targets. Post hoc comparisons confirmed that P300 latency in both tasks was significantly longer in the two injured groups (mTBI=ICNHC). The results suggest that automatic processing, as reflected by P3a to novel stimuli, is relatively intact after blast-induced injury. In contrast, group differences in P300 indicate that higher-level processing, involving working memory, is delayed in service members with recent mild TBI. The finding of delayed processing in injured service members negative for TBI was unexpected. If this preliminary finding is replicated as our sample size increases, it suggests that sequelae of blast injury may include altered information processing even when subjects screen negative for TBI.