Locating primary somatosensory cortex in human brain stimulation studies: systematic review and meta-analytic evidence

Transcranial magnetic stimulation (TMS) over human primary somatosensory cortex (S1), unlike over primary motor cortex (M1), does not produce a measurable output. Researchers stimulating S1 with TMS must therefore rely on one or more indirect methods to position the TMS coil over the target area. One common method is to find the optimal location on the scalp at which twitches in a hand muscle (typically the first dorsal interosseus, M1-FDI) are produced, then move the coil several centimetres posteriorly. Here, we ask whether this method is optimal for stimulation of the index finger area of S1 (S1-index). First, we systematically reviewed 117 TMS studies that targeted S1-index, and 54 functional magnetic resonance imaging (FMRI) studies involving passive stimulation of the index finger. Second, we re-analysed FMRI data in which all ten fingers received vibrotactile stimulation, and constructed a probabilistic atlas of the central sulcus from 100 healthy adult MRIs. Third, we conducted two experiments mapping the effects of TMS across the scalp on tactile discrimination performance, and re-examined all available data from our laboratory pertaining to the scalp locations of S1-index and M1-FDI. Contrary to the prevailing method, we show that S1-index is very close to the C3/C4 electroencephalography (EEG) electrode locations, approximately 7-8cm lateral to the vertex. Importantly, S1-index is approximately 2cm lateral and 0.5cm posterior to M1-FDI, measured on the scalp. These results suggest that an immediate revision to the most commonly-used heuristic to locate S1-index is required, and the results of many TMS studies of S1-index need reassessmen