Abstract
PURPOSE. The superior colliculus (SC) plays an important role in the control of voluntary gaze shifts toward visual targets. Tonic neural activity preceding a saccadic burst appears to be related to the probability that a given object will be chosen as a saccade target. To test whether an alteration of this presaccadic activity can alter a monkey's target choice, we microstimulated in the SC during a 2-AFC task. METHODS. We trained two macaque monkeys to saccade to one of two 0.2 × 0.4° bars (one white, one gray) which appeared on opposite sides of a central fixation point at fixed horizontal eccentricities (3.5°). During the fixation interval, a central cue flashed either white or gray, specifying the color of the upcoming target. After a 1 – 1.7 s delay, both bars appeared; the monkey was rewarded for making a saccade directly to the bar matching the color of the cue. During randomly interleaved microstimulation trials, biphasic current was applied to the SC beginning either 100 or 200 ms prior to target and distracter appearance and persisting for ~ 400 ms. We chose subthreshold stimulation parameters (10–30 A, 100–200 Hz) to minimize stimulation-evoked saccades. RESULTS. Although somewhat variable, the effects of stimulation depended in a predictable way on the stimulation intensity and the site of stimulation with respect to target location. Saccades to bars ipsilateral to the site of stimulation had longer latencies than contralateral saccades. Stimulation resulted in an increase in the number of correct responses when the target was contralateral to the site of stimulation and a decrease in correct responses when the target was ipsilateral. CONCLUSION. Using the 2-AFC design, we found that microstimulation can change target choice in addition to the well-known change in saccade metrics. These results demonstrate that presaccadic activity in the SC is causally related to the probability that a given object will become a saccade target.