Abstract
Last year, we reported that temporary inactivation of a portion of the superior colliculus (SC) in rhesus monkeys results in a target selection deficit in a color-oddity search task. Specifically, after a microinjection of lidocaine or muscimol into the intermediate layers of the SC, saccades can still be made to visual targets in the affected part of the field when distractors are absent. However, when the target must be selected from distractors, monkeys make a greater proportion of erroneous saccades to distractors. We have now observed a similar, although less pronounced, target-selection deficit after temporary muscimol inactivation of a portion of the frontal eye field (FEF). Saccade target selection has been hypothesized to involve a winner-take-all competition across a map of visual salience. Our results suggest that the salience of stimuli in the affected field for the saccadic system is effectively reduced by inactivation of SC or FEF. To test this idea, we manipulated the perceptual salience of the search target by varying the difference in color between target and distractors. After inactivation of either SC or FEF, the magnitude of the search deficit was strongly affected by target salience, consistent with a salience-based target selection deficit, rather than a purely motor impairment. Although the effect on target selection of inactivation of either SC or FEF was similar in most respects, we found that performance after SC inactivation grew markedly worse when more distractors were present, while the opposite trend was seen after FEF inactivation. This suggests that the selective mechanisms in the SC depend more strongly on local inhibition among neighboring stimuli than in FEF. Overall, our results indicate that both SC and FEF play a functional role in saccade target selection.