Abstract
A single saccade paradigm was developed to study search eye movements across the visual field of the macaque. We measured both visually and memory guided saccades during a search task performed monocularly within the central 30 deg field of view. Monkeys were required to make an eye movement to a cued target among zero to seven distractors. Positional jitter was used to make target and distractor locations unpredictable from trial to trial. Target and distractors were black discs presented on a gray background and targets could be either smaller or larger than the distractors.
Saccade metrics, such as latency and accuracy, were analyzed. Saccadic latency was defined as the duration from initial fixation offset to the beginning of the eye movement. Visually guided search showed shorter latencies than memory guided and did not differ with eccentricity. Memory guided search latencies decreased with eccentricity.
Saccade accuracy was evaluated by the distance from the saccadic endpoint to the target. Accuracy decreased with search eccentricity, and compared to visually guided search, memory guided eye movements were less accurate with more variability in saccadic endpoint location and became more hypometric with increasing eccentricity. In general, saccade endpoint variability increased with greater eccentricity.
Eye movements were also examined in each major visual field region. No consistent nasal-temporal field differences were found. Test subjects displayed the most accurate visually-guided saccades in the superior field. For both visually and memory guided trials, search eye movements to the superior visual field were predominately hypermetric, while saccades in the other visual field regions were predominately hypometric.
Supported by NIH grant P30 EY01319 and Research to Prevent Blindness