The programming and reprogramming of oculomotor responses to double-step and single-step targets was investigated in 5-6-year-old and 10-12-year-old children and in adults. The independent variables in Experiment 1 were intertarget interval (50, 100, 150, and 200 msec) and target location. The number of trials on which a saccade was made to both first and second targets increased with age and intertarget interval, but the two factors did not interact. On trials where responses were made only to the second target, the children responded slower than the adults but showed generally similar patterns of response latencies. In Experiment 2, a warning signal was presented 0, 100, or 300 msec prior to the first target. For adults the 100-and 300-msec warning intervals reduced the latency of single-step responses and the first saccade of double-step responses, whereas only the 300-msec warning interval was similarly effective with children.In both experiments subjects in all age groups exhibited amplitude transition functions, indicating that the modifiability of saccadic programming is basically similar for adults and children. A comparison of simultaneous programming characteristics of adults and children suggested possible age differences, but the data were not conclusive. The results indicated age differences in the rate of programming and reprogramming saccades but no qualitative age-related differences in these processes.

Vernier acuity (delta v) as a function of two-dot separation (s) was measured at five retinal locations between the foveal center and 2.0 deg of eccentricity. We compare these results with average cone spacing at each of the corresponding retinal eccentricities and find that the angular dot separations at which the delta v versus s function intersects the angular cone spacing at each eccentricity remain nearly constant. Further, we define a vernier Weber fraction delta v/s and find that the average Weber fraction increases by nearly a factor of 2 from the fovea to 2.0 deg of retinal eccentricity. We suggest that both receptor factors and postreceptor factors contribute to limits of vernier acuity within 2 deg of the foveal center.

Conventional methods of measuring retinal function are dependent on the optics of the eye. Therefore, an optical opacity such as a cataract that obstructs the normal optical transmission of light can prevent measurement of retinal function. In some cases a laser interferometer designed to bypass the effects of the optics of the eye may be employed to measure retinal function without interference due to the optical components. In this study we compare contrast sensitivity functions determined by laser interferometry and conventional display methods for cataract and normal eyes. Comparison of laser and monitor contrast sensitivity functions indicates that for these cataract patients interferometric measurements show a normal retinal function, whereas similar measures using the traditional display system (monitor) show considerable loss of visual function. The extent to which contrast sensitivity was limited by the lens of the eye is indicated by the ratio of laser and monitor contrast thresholds. In general, the laser interferometry technique demonstrates that relatively minor cataracts decrease the contrast transfer function of the eye' optics over a range of spatial frequencies.

Spatial discriminations that are finer than the center-to-center spacing of foveal photoreceptors (hyperacuity) are readily observed for 1-D spatial tasks such as Vernier offset, spatial frequency and line separation discriminations, and bisection tasks. We introduce a new task where the just- noticeable difference in offset from the center of circle Δc is measured. Circle center discriminations were measured over a range of circle radii and compared to just-noticeable differences in bisection between two dots Δb with separations equal to the circle diameters. The data show that Δb and Δc were equivalent for small stimuli (radii or bisection distance were approximately ≤0.025° of visual angle). For stimuli larger than this critical size, performance on the 2-D circle task Δc progressively exceeds performance on the 1-D bisection task Δb. These results demonstrate that the human visual system is able to exploit a 2-D spatial parameter (area) to achieve a hyperacuity advantage.