Transcranial magnetic stimulation of frontal oculomotor regions during smooth pursuit

Abstract

Both the frontal eye fields (FEFs) and supplementary eye fields (SEFs) are known to be involved in smooth pursuit eye movements. It has been shown recently that stimulation of the smooth-pursuit area of the FEF [frontal pursuit area (FPA)] in monkey increases the pursuit response to unexpected changes in target motion during pursuit. In the current study, we applied transcranial magnetic stimulation (TMS) to the FPA and SEF in humans during sinusoidal pursuit to assess its effects on the pursuit response to predictable, rather than unexpected, changes in target motion. For the FPA, we found that TMS applied immediately before the target reversed direction increased eye velocity in the new direction, whereas TMS applied in mid-cycle, immediately before the target began to slow, decreased eye velocity. For the SEF, TMS applied at target reversal increased eye velocity in the new direction but had no effect on eye velocity when applied at mid-cycle. TMS of the control region (leg region of the somatosensory cortex) did not affect eye velocity at either point. Previous stimulation studies of FPA during pursuit have suggested that this region is involved in controlling the gain of the transformation of visual signals into pursuit motor commands. The current results suggest that the gain of the transformation of predictive signals into motor commands is also controlled by the FPA. The effect of stimulation of the SEF is distinct from that of the FPA and suggests that its role in sinusoidal pursuit is primarily at the target direction reversal.

Figure 1.

Transverse (A) and coronal (B) sections illustrating the location for each subject (n = 6) of the lFPA, rFPA, and SEF as determined by fMRI comparing smooth pursuit to fixation. The left hemisphere is shown on the left side of each image. The locations are superimposed on a single subject's MRI, which has been transformed into standard Talairach space. For illustrative purposes, the horizontal slice (z = 53) was selected based on the average z coordinates of activation for the FPA and SEF as determined by fMRI. The coronal section (y =–7) was selected based on the average y coordinates of activation for the FPA and SEF. Note that the greatest activity in the FPA was consistently found at or near the junction of the superior frontal and precentral sulci (A) and that increased activity in the SEF was regularly found in the medial wall near the paracentral sulcus (B), both established anatomical land marks for these regions in humans (Paus, 1996; Grosbras et al., 1999).

Figure 2.

Experiment 1. The effect of TMS on eye velocity when applied at peak target velocity in one subject. The x-axis is time (milliseconds), and the y-axis is position. The panels depict pursuit at midcycle of a target moving horizontally at 0.4 Hz with a sinusoidal velocity profile across 20° of visual angle. The dotted line indicates the point of peak target velocity, the moment that TMS was applied in B. The gray portion of each panel illustrates the 100 ms analysis window that begins 25 ms after peak target velocity. In A (No TMS), eye velocity is close to target velocity (average eye velocity in window, 23.0°/s; average target velocity, 24.4°/s). In B (TMS of lFPA), eye velocity in the window is reduced (average eye velocity, 20.0°/s; average target velocity, 24.4°/s). R, Right; L, left.

Figure 3.

Experiment 1. Main effect of TMS on eye velocity when applied at peak target velocity for each ROI for each individual. Each individual's eye velocity with and without TMS is connected by a straight line. TMS applied at peak target velocity to the lFPA and rFPA decreased eye velocity (for the rFPA, the third diamond from the top in the no-TMS condition represents two subjects who have almost identical data in the two conditions). There is no consistent effect on eye velocity of TMS applied to the SEF or rosPAR.

Figure 4.

Experiment 2. An example of the effect of TMS on eye velocity when applied at the target direction reversal in one subject. The x-axis is time, and the y-axis is position. The panels depict pursuit at the target direction reversal; the target is moving horizontally at 0.4 Hz with a sinusoidal velocity profile across 20° of visual angle. The dotted line is placed 75 ms before the target direction reversal, the moment at which TMS was applied in B. The gray portion of each panel illustrates the 100 ms analysis window that began the moment the eye reversed direction (see Materials and Methods). A, No TMS: the average eye velocity in this analysis window was 2.7°/s (average target velocity, 3.5°/s). B, TMS of the lFPA: the average eye velocity in this analysis window was 4.8°/s (average target velocity, 3.2°/s). R, Right; L, left.

Figure 5.

Experiment 2. Effect of TMS on eye velocity when applied at the target direction reversal. Each individual's eye velocity with and without TMS is connected by a straight line. TMS of each ROI significantly increased eye velocity.

Figure 6.

Experiment 2. Effect of TMS on the time required for the eye to reverse direction when applied at the target direction reversal. Each individual's direction reversal latency with and without TMS is connected by a straight line. There was no effect of TMS on direction reversal latency for any ROI and no interaction of TMS with direction (all p > 0.05).

Figure 7.

Experiment 3. Effect of TMS on eye velocity when applied to the lFPA and leg-S1 at peak target velocity and the target direction reversal. Each individual's eye velocity with and without TMS is connected by a straight line. Consistent with the stimulation results of the lFPA in both experiments 1 and 2, TMS applied at peak target velocity significantly decreased eye velocity and TMS applied at the target direction reversal significantly increased eye velocity. TMS applied to the leg-S1 at peak target velocity or the target direction reversal did not affect eye velocity (p > 0.05). The outlying trace in the two top panels belong to the same subject.