P 134. Causal and directional information flow from visual to motor areas during resting-state revealed by a TMS-EEG co-registration

Abstract

Our brain is spontaneously active and constructs a resting-state network even when there are no sensory inputs or no motor output responses. Such networks connect many distributed brain regions, however but it is not well known how clear about causal and directional information flow occurs within the resting-state network. Such causal information flow could be evaluated by perturbing electroencephalograph (EEG) oscillations using transcranial magnetic stimulation (TMS). A single-pulse TMS can reset the phase of ongoing cortical oscillations. Here we analyzed global propagation of phase resetting across different brain regions to elucidate causal information flow. More specifically, we applied single-pulse TMS to the resting-state brain and analyzed spatial spread of TMS-induced phase resetting in the human brain. Eleven right-handed subjects took part in this experiment. A single-pulse TMS was delivered to left primary motor cortex or visual cortex at 2.5–3.5 s intervals during the eye-closed resting states. Each subject completed four separate sessions; two TMS intensities (95% and 50%; 100% means motor threshold) × two TMS-target locations (left primary motor cortex (C3) and visual cortex (Oz)). Each session consisted of 50 times of TMS applications. To confirm their arousal, they were asked to response by their right index finger when they felt sensed a weak flashlight which was sometimes induced by a white flash square on a computer display during inter-pulse interval periods. EEGs were measured from 67 scalp electrodes. To elucidate the cortical activity free without errors from volume conduction, we applied current source density analysis. Next, we removed the EEG epochs that would be affected by TMS artifacts, using a linear interpolation. Finally, to identify the time–frequency (TF) phases, we applied wavelet transforms using Morlet’s wavelets. TMS-induced evoked phase resetting was assessed calculated by phase locking values (PLV) at each electrode, time point and frequency. The time–frequency EEG results showed enhanced PLV for of both the low and high frequency bands at the TMS target electrodes at the instant of TMS applications. In particular, the theta (4–8 Hz) PLV increased from the onset of TMS applications around the TMS target locations. Under visual area-targeting TMS, phase resetting (i.e. enhanced PLV) of the theta oscillations were transmitted from the visual areas to the motor areas (in particular to the left motor area). Under high largeTMS intensity, the left motor electrode showed the highest theta PLV from about 150 ms after the onset of TMS applications, whereas PLV at the visual electrode reached peaking values just when TMS was applied. Under low small TMS intensity, such regional transmissions of the phase resets decreased and disappeared. In contrast, the TMS to the left motor areas increased only the theta PLV in the left and central motor areas. High PLV was observed from about 100 ms after TMS applications. In contrast, the visual areas showed low small PLV. Similar to the visual area-targeting TMS, the regional transitions of the theta PLV disappeared when the TMS intensity was low small. We succeeded in evaluating directional information flow in the resting state. Our results indicated the existence of potential networks from the sensory input regions to the motor output regions in the resting state. The sensory–motor mechanisms might be related to a spontaneous preparation of perception–action coupling; e.g. seeing a visual stimulus, making a decision, and then responding by the motor output.