Over the past 20 years, a considerable amount of research has been conducted to determine the relationship between the visual attention and oculomotor systems. Specifically, researchers have attempted to determine if a movement of attention precedes endogenous (voluntary) saccadic eye movements. Previous studies that have examined this question have typically used a dual-task paradigm in which the primary task is an eye movement task and the secondary task is a manual response task, sensitive to attentional allocation. However, the results of these dual-task studies have provided conflicting evidence.

The present study used a novel priming paradigm, in which a saccade is a necessary part of a single task, to examine the relationships between attention and eye movements. Participants were required to identify an object, located in the periphery, to left or right of fixation. The location of the target was indicated by a central arrow (the location cue) and the target was so small that the only way to make the discrimination was to make a saccade to the target. In Experiment 1 a prime that was compatible or incompatible with regard to target identity was presented at the target location simultaneously with the location cue. The prime was either predictive (80% valid) or unpredictive (50% valid) of the target identity. The results showed that there were priming effects (lower RT's when prime and target were compatible than when they were incompatible) in both validity conditions and that the priming effect was larger when the prime was predictive. This suggests an attentional shift to the target location prior to the execution of the saccade. In Experiment 2 the prime (now always unpredictive) was presented at the target or non-target location and it was presented simultaneously with (0 ms SOA), or prior to (-250 ms SOA), the onset of the location cue. The results showed that there was a priming effect when the prime appeared at the target location, but not when it appeared at the non-target location. Furthermore, the priming effect at the target location was larger when the prime appeared simultaneously with the location cue, than when it appeared prior to the location cue. Experiment 3 used the same design but with an additional control condition in which participants were to remain centrally fixated. Priming effects were found in the eye movement conditions, with greater priming effects with the 0 ms SOA than with the -250 ms SOA. Only the 0 ms SOA yielded priming effects in the fixation conditions.

Overall, these results suggest that a shift of attention precedes saccades and that a shift of attention accompanied by oculomotor programming is faster than an endogenous shift of attention without oculomotor programming. These results further support the view that the visual attention system and the oculomotor system are intimately related.