RAPID COMMUNICATION Attention Modulates Both Primary and Second Somatosensory Cortical Activities in Humans: A Magnetoencephalographic Study

Mima, Tatsuya, Takashi Nagamine, Kaori Nakamura, and Hir1990, 1991; Sinclair and Burton 1993) showed that neuronal oshi Shibasaki. Attention modulates both primary and second soactivity in SII shows spatial as well as modality convergence matosensory cortical activities in humans: a magnetoencephaloand greater stimulus specificity dependent on SI activity. This graphic study. J. Neurophysiol. 80: 2215–2221, 1998. To clarify is in accord with the hypothesis that SII serves as a higher the role of primary and second somatosensory cortex (SI and SII) order region for somatosensory information processing. These in somatosensory discrimination, we recorded somatosensory observations also support the concept of a ventral somatosenevoked magnetic fields during a stimulus strength discrimination sory pathway from SI via SII to the insula, which finally task. The temporal pattern of cortical activation was analyzed by projects to limbic structures related to long-term memory dipole source model coregistered with magnetic resonance image. (Friedman et al. 1986; Mishkin 1979). This study addresses Stimulus intensity was represented in SI as early as 20 ms after the stimulus presentation. The later components of SI response the question about the time course of SI and SII activation as (latency 37.7 and 67.9 ms) were enhanced by rarely presented well as the responsiveness of these regions to passive versus stimuli (stimulus deviancy) during passive and active attention. active attention to stimulus change, with a whole head magnetThis supports an early haptic memory mechanism in human prioencephalography (for review see Hämäläinen et al. 1993). mary sensory cortex. Contraand ipsilateral SII responses followed Previous human studies on somatosensory recognition mainly the SI responses (latency 124.6 and 138.3 ms, respectively) and utilized a spatial attention task stimulating different fingers were enhanced by attention more prominently than the SI re(Desmedt and Tomberg 1989; Garcia-Larrea et al. 1995; Hari sponses. Active attention increased SII but not SI activity. These et al. 1990; Kekoni et al. 1996; Mauguière et al. 1997; Yamaresults are consistent with the concept of ventral somatosensory guchi and Knight 1991). In these tasks, however, cortical pathway that SI and SII are hierarchically organized for passive receptive areas are different for attended and nonattended stimand active detection of discrete stimuli. uli because of somatotopic organization. Accordingly, the rate of impulse arrival at the somatosensory cortex is also different I N T R O D U C T I O N for stimuli delivered to different digits. Thus it is difficult to disentangle the effect of attention from that of stimulus Somatosensory information is represented in multiple areas presentation rate or stimulus deviancy (Hari et al. 1990; Naof human brain. Among them, the primary and second somatogamine et al. 1998). To address this problem, we employed sensory cortices (SI and SII) are the principal regions for a stimulus intensity detection task (Michie 1984; Michie et tactile perception. Activation studies with positron emission al. 1987) in which the activated brain areas shared somatotopy. tomography in human reported that somatosensory stimulation We examined the effects of both passive and active attentions increases regional blood flow in SI and the parietal opercular on neural activity in SI and SII. cortex (including SII) (Bonda et al. 1996; Burton et al. 1997; Fox et al. 1987; Seitz and Roland 1992). However, there is