Somatosensory system deficits in schizophrenia revealed by MEG during a median-nerve oddball task

Abstract

Although impairments related to somatosensory perception are common in schizophrenia, they have rarely been examined in functional imaging studies. In the present study, magnetoencephalography (MEG) was used to identify neural networks that support attention to somatosensory stimuli in healthy adults and abnormalities in these networks in patient with schizophrenia. A median-nerve oddball task was used to probe attention to somatosensory stimuli, and an advanced, high-resolution MEG source-imaging method was applied to assess activity throughout the brain. In nineteen healthy subjects, attention-related activation was seen in a sensorimotor network involving primary somatosensory (S1), secondary somatosensory (S2), primary motor (M1), pre-motor (PMA), and paracentral lobule (PCL) areas. A frontal-parietal-temporal "attention network", containing dorsal- and ventral-lateral prefrontal cortex (DLPFC and VLPFC), orbitofrontal cortex (OFC), anterior cingulate cortex (ACC), superior parietal lobule (SPL), inferior parietal lobule (IPL)/supramarginal gyrus (SMG), and temporal lobe areas, was also activated. Seventeen individuals with schizophrenia showed early attention-related hyperactivations in S1 and M1 but hypo-activation in S1, S2, M1, and PMA at later latency in the sensorimotor network. Within this attention network, hypoactivation was found in SPL, DLPFC, orbitofrontal cortex, and the dorsal aspect of ACC. Hyperactivation was seen in SMG/IPL, frontal pole, and the ventral aspect of ACC in patients. These findings link attention-related somatosensory deficits to dysfunction in both sensorimotor and frontal-parietal-temporal networks in schizophrenia.

Fig. 1

Median-nerve oddball paradigm. Block 1: 15% of the stimuli were delivered to the right wrist and 85% to the left wrist. Subjects were instructed to count silently the rare stimuli to the right wrist. Block 2: the rare and frequent stimuli were reversed, and subjects were instructed to silently count the rare stimuli to the left wrist. For each side, oddball MEG signals were obtained by subtracting the frequent responses from the rare responses across different blocks as indicated by the arrows. Block 3 was identical to Block 1, and Block 4 was identical to Block 2

Fig. 2

MEG sensor waveforms from 204 gradiometer channels (102 pairs of co-located orthogonal pairs; top row) and 102 magnetometer channels (bottom row) in a representative healthy control participant evoked by right median-nerve oddball stimuli averaged over “frequent” and “rare” trials. a Gradiometer waveforms for the first 200 ms interval evoked by frequent stimuli. The two arrows indicate the sharp N20m and P30m components. b Gradiometer waveforms for rare stimuli. Note the marked increase in averaged signal amplitude. c Gradiometer waveforms of rare minus frequent oddball responses. d Same as c, but for the whole 500 ms interval. The bottom row displays magnetometer waveforms for the frequent (e), rare (f), rare minus frequent responses for the first 200 ms (g), and for the whole 500 ms interval (h), respectively

Fig. 3

Attention-related brain activations (rare minus frequent responses) in the control group derived from the VESTAL method. a Left-hemisphere activation, evoked by contralateral median-nerve stimulation of the right arm. b Left-hemisphere activation, evoked by ipsilateral stimulation of the left arm. c Right-hemisphere activation, evoked by contralateral stimulation of the left arm. d Right-hemisphere activation, evoked by ipsilateral stimulation of the right arm. FWER was controlled at P = .01 level

Fig. 4

Group differences in attention-related brain activation (rare minus frequent responses). a Left-hemisphere group differences in activation, evoked by contralateral (right arm) median-nerve stimulation. b Left-hemisphere group differences in activation, evoked by ipsilateral stimulation (left arm). c Right-hemisphere group differences in activation, evoked by contralateral stimulation (left arm). d Right-hemisphere group differences in activation, evoked by ipsilateral stimulation (right arm). FWER is controlled at P = .05 level. The red-yellow color indicates statistically stronger activation in the control group than in the patient group (hypo-activation for the patients), whereas the blue-cyan color indicates stronger activation in the patient than control group (hyper-activation for the patients)