Magnetic stimulation over human primary somatosensory cortex interferes selectively over time with tactile detection

Introduction

Detecting and discriminating between sensory stimuli are fundamental functions of the nervous system. Electrophysiological and lesion studies in the macaque brain suggest that primary somatosensory cortex (S1) is critically involved in discriminating between somatosensory stimuli, but is not required simply for detecting those same stimuli. By contrast, classical studies in humans using transcranial magnetic stimulation (TMS) over the sensorimotor cortex show near-complete disruption of perception when TMS is delivered around the same time as the somatosensory target. To resolve this discrepancy, we used single, and dual-pulse TMS over human S1, guided by single- subject structural and functional brain images.

Methods

Participants (N=12 per experiment) underwent a series of MRI scans (localisers) to produce somatotopic maps of S1. The maps were used to stimulate over S1 with TMS during subsequent experiments. We measured the sensitivity and the decisional criterion of participants in a tactile detection task in which vibrotactile stimuli were presented at threshold level in a one-interval forced-choice (1IFC) design while fMRI-guided TMS was applied over S1, over a control site not activated by vibrotactile stimuli (supramarginal gyrus, SMG), or away from the head (sham) (Experiment 1). We also measured tactile thresholds using QUEST (i.e., an adaptive staircase procedure) in a series of criterion-free two-interval forced-choice (2IFC) vibrotactile detection and discrimination tasks with tactile stimuli (50ms, 200Hz sinusoidal vibration) applied to the middle and/or index fingertips of one hand (Experiments 2-7).

Results

Results showed that in the 1IFC design TMS increased participants' likelihood of reporting 'no' target present regardless of site (Figure 1c-d), but TMS over S1 also impaired detection only over time (i.e., in the latter half of the experiment) (Figure 1a-b). In the 2IFC design we found that thresholds for detecting vibrotactile stimuli were not influenced by single- or double-pulse TMS over S1 as compared to over SMG (Figure 2a-b). By contrast, TMS over the median nerve at the wrist increased thresholds relative to TMS over finger extensor muscles, and TMS over S1 increased thresholds in a frequency-discrimination task (Figure 2c-f).

Conclusions

We conclude that, in accordance with macaque studies (LaMotte and Mountcastle, 1979; Romo et al., 2012), S1 is required for discriminating between tactile stimuli, and for holding stimulus representations over time, but may not be required for simple detection of tactile stimuli at the fingers.