The Neural Control of Visually Guided Eye Movements
D. Temporal Factors in Visually Guided Saccadic Eye Movements
          1. The conditions under which express saccades can and cannot be generated.

i. When successive saccades are generated, does the frequency with which express saccades are made depend on the direction of the saccadic vector made relative to the initial fixation?

It has been reported by Carpenter (Vision Research, 2001, 41, 1145-1151) that when humans make saccades to two successive targets, express saccades are frequent to the second target when it is further away than the first (away condition) but are not when it appears at a location at or near the initial fixation spot (return condition). We examined this in monkeys and could not replicate the reported effect. To determine what might be going on, we examined the performance of the monkey whose data had been shown in Figure 37. This animal made profuse express saccades with a vector of 225/3 and few with a vector of 45/3. We therefore arranged the two successive targets so that the monkey made his second saccade either toward or away from fixation with either 225/3 or 45/3 vector saccades. This was accomplished by strategically placing the fixation spot to appropriate locations. The distribution of saccadic latencies obtained to the second target is shown in Figure 39 and established the following: (1) The monkey made few express saccades for the away condition when the second saccade had a vector of 45/3 (Figure 39A) and (2) made many express saccades for the away condition when the second saccade had a vector of 225/3 (Figure 39D). (3) For the return conditions (Figure 39B&C) the monkey again made few express saccades with a vector of 45/3 and many with a vector of 225/3. Thus the crucial factor in express saccade generation is not whether the second saccade is toward or away from the fixation spot but what the direction and amplitude is of the saccadic vector.