-It has been hypothesized that the human cortical responses to nociceptive and nonnociceptive somatosensory inputs differ. Supporting this view, somatosensory-evoked potentials (SEPs) elicited by thermal nociceptive stimuli have been suggested to originate from areas 1 and 2 of the contralateral primary somatosensory (S1), operculo-insular, and cingulate cortices, whereas the early components of nonnociceptive SEPs mainly originate from area 3b of S1. However, to avoid producing a burn lesion, and sensitize or fatigue nociceptors, thermonociceptive SEPs are typically obtained by delivering a small number of stimuli with a large and variable interstimulus interval (ISI). In contrast, the early components of nonnociceptive SEPs are usually obtained by applying many stimuli at a rapid rate. Hence, previously reported differences between nociceptive and nonnociceptive SEPs could be due to differences in signal-to-noise ratio and/or differences in the contribution of cognitive processes related, for example, to arousal and attention. Here, using intraepidermal electrical stimulation to selectively activate A␦-nociceptors at a fast and constant 1-s ISI, we found that the nociceptive SEPs obtained with a long ISI are no longer identified, indicating that these responses are not obligatory for nociception. Furthermore, using a blind source separation, we found that, unlike the obligatory components of nonnociceptive SEPs, the obligatory components of nociceptive SEPs do not receive a significant contribution from a contralateral source possibly originating from S1. Instead, they were best explained by sources compatible with bilateral operculo-insular and/or cingulate locations. Taken together, our results indicate that the obligatory components of nociceptive and nonnociceptive SEPs are fundamentally different.pain; nociception; event-related potentials; primary somatosensory cortex; laser-evoked potentials; intraepidermal stimulation FOR THE PAST 30 YEARS, a large number of studies have aimed at understanding the neural mechanisms underlying the processing of nociceptive input and the perception of pain in the human cortex. Most of these studies have relied on noninvasive techniques to sample brain activity, such as electroencephalography (EEG), magnetoencephalography (MEG), functional magnetic resonance imaging (fMRI), and positron emission tomography (PET) (Bushnell and Apkarian 2005;Garcia-Larrea et al. 2003;Kakigi et al. 2005;Peyron et al. 2002;Treede et al. 1999). With these techniques, it has been shown repeatedly that nociceptive stimuli elicit responses in a wide array of brain areas, including postcentral, operculo-insular, and cingulate areas. A number of investigators have considered that the combined activation of these brain regions is responsible for the transformation of nociceptive input into a conscious perception of pain, in particular, the coding of pain intensity and pain unpleasantness (Boly et al.