TARGET SIDE AND SCALP TOPOGRAPHY OF THE SOMATOSENSORY P300

We recorded somatosensory P300 to target electrical shocks (10% probab ility) delivered to the left or right hand concurrently with non-targe t stimuli applied to the opposite hand. Target stimuli evoked a widely distributed P300, which was found to be double-peaked in 60% of the s ubjects investigated. The mean latencies of the ''early'' and ''late'' P300 peaks were respectively 302+/-8 msec and 353+/-12 msec. Both com ponents (but especially the earlier one) were consistently lateralized over the hemiscalp contralateral to the target stimuli, whichever the target side. Although on theoretical grounds P300 lateralization migh t depend upon the sustained attentional activation of the hemisphere c ontralateral to targets, this hypothesis was contradicted in one compl ementary experiment in which we dissociated the side receiving the P3- evoking, rare, stimuli from the side to which sustained attention was directed. Under these conditions the P3 evoked by rare, but ignored, s hocks corresponded to the ''early'' peak of target P300 and was also p redominant contralateral to the evoking stimulus, i.e., over the hemis phere not activated by sustained attention. Therefore the lateralizati on of the somatosensory P300, or at least that of its ''early'' compon ent, is that of the hemisphere receiving the PS-evoking volley, and no t that of the one involved in maintaining sustained attention. The P3 to rare-ignored shocks and the ''early'' P300 to targets had identical latencies; both are likely to reflect an automatic processing of devi ant stimuli and thus belong to the family of ''novelty P3s'' (or ''P3a '').