Abstract
Despite internal processing delays monkeys and humans perform accurate interceptive saccades toward moving targets. This is possible through an estimate of the target position at saccade end through the combination of position and velocity information. We investigated interceptive saccades of two macaque monkeys and 4 human subjects in response to step-ramps of eight different relative angular directions between their step and ramp components. The directions differed by 45° covering the whole 360° direction space. A target appeared at an eccentric location, then a step to the center occurred and the target immediately moved at 10°/s in one of the eight directions. In addition we presented static targets for each direction with matched saccadic amplitudes. Despite different average saccade latencies (humans: 150 ms; monkeys 120 ms) monkeys and humans showed highly similar oculomotor behavior. The accuracy of initial saccades depended on the direction difference between steps and ramps. Initial saccades to orthogonal step-ramps reflected the stimulus movement quite accurately. Saccades to co-directional step-ramps appeared to lag ~1° behind the stimulus, while saccades to opposite-directional step-ramps were ahead by ~1°, probably due to general saccadic undershoot. We found - in agreement with Guan et al. (EBR, 2005) - that initial saccades towards moving targets differed in their main sequence dependent on the relative directions of step and ramp. For co-directional step-ramps the peak saccade velocity was significantly lower, than for step-ramps pointing in opposite directions. The saccade velocities for orthogonal step-ramps were in-between the two. This result is in line with findings in monkeys that neuronal activity in a subcortical saccade control area - the superior colliculus - does not fully account for the amplitudes of interceptive saccades. As a next step, we will investigate the contribution of the parietal eye field – monkey area LIP - to accurate interceptive saccades.
Acknowledgement: Deutsche Forschungsgemeinschaft CRC/TRR-135 A1