Abstract
Background: A large number of studies have investigated observers' ability to make targeted saccades in the absence of other stimuli. In these studies the observers' task is to saccade quickly and accurately to the target, usually presented in one of a limited number of locations. Here we investigated observers' ability to make accurate saccades when they first must select the target on the basis of its color, and when the target was presented alone and in the presence of other stimuli. Methods: Observers made saccades to an oriented-c of a specified color when it was presented alone or flanked on both sides by one or two c's of other colors. The spacing between the target and the flanks was either near or far. The target was randomly presented to the left or right of fixation. The display on the other side of fixation was the same, except that it did not contain the target color. Targets and flanks were about 0.33 deg in diameter. Eye movements were tracked with a dual-Purkinje-image eye tracker sampled at 1000Hz. Trials were judged correct if the eye landed within a window centered on the target of ±1.5x its size. Results: Observers performed poorly on this task. In the absence of flanks they landed on the target only 54% of the time. The addition of one flank reduced performance to 43%; two flanks further reduced performance to 30%. There was little effect of spacing; 38% for the near spacing and 34% for the far spacing. Reaction times parallel these data. Conclusions: We were surprised by the poor performance of the observers in this study. It is possible that the choice component of the task or the displays we chose interfere with one's ability to program accurate saccades. Experiments are under to test these possibilities.