Abstracts

RATIONALE: Left-hemispheric specialization in speech perception may be based on temporal processing (Tallal & Newcombe, 1978 ). Data from studies using intracerebral depth electrodes (Liegeois-Chauvel et al., 1999)suggest that the left auditory cortex is preferentially involved in the temporal (ie. sequential) processing of different acoustic elements in speech. The aim of the present study was to explore such temporal processing using scalp-recorded auditory evoked potentials (AEPs), compared to depth-recorded AEPs in epileptic patients. METHODS: To establish hemispheric dominance, a WADA test was administered to a group of epileptic patients who had previously undergone a pre-surgical evaluation (SEEG). AEPs to voiced and voiceless natural-speech stop-consonant syllables[/ba/ (350 ms) and /pa/ (220 ms)] were recorded using intracerebral electrodes implanted in Heschl s gyrus (HG) and surface electrodes from a 64-electrode cap. Stimuli were presented monoaurally. AEPs were recorded separately in each of the two sessions and were later compared. RESULTS: AEPs differed significantly according to stimuli for both techniques. For the voiced /ba/, four different components were observed (mean latencies: 100 ms;178 ms; 227 ms; 280 ms). Only two components ( 100 ms; 182 ms), however; were observed for the voiceless /pa/. The two supplementary components (227 ms; 280 ms) for /ba/ were observed more often on right-ear presentations and corresponded to the temporal processing of voiced stop-consonant stimuli. These components originating specifically in the left HG for subjects with LH dominance for language were predominantly observed over the left hemisphere with scalp recordings CONCLUSIONS: These results suggest that temporal processing of the LH may be evaluated using scalp recordings. Scalp-recorded AEPs may represent an effective, non-invasive method to establish hemispheric dominance for language before epilepsy surgery.