Background

Dynamic visual acuity is measured using acuity optotypes, such as Landolt C's, Sloan or Snellen letters, and checkerboard patterns, but under conditions in which these optotypes are moving and the observer must track them. During the past 30 years some 73 researchers have generated 81 reports on this topic, but, apart from the pioneering work by Ludvigh and Miller (1953, 1958; Miller and Ludvigh, 1953, 1962) and subsequent applications by Burg and coworkers (Burg, 1966, 1967, 1968; Burg and Hulbert, 1961; Henderson and Burg, 1973), there has been little sustained, programmatic effort to further develop measurement methodology or to understand the basis of dynamic visual acuity. There is general agreement that it decreases as a function of the acuity target's angular velocity with respect to the observer (Miller and Ludvigh, 1962; Morrison, 1980). This decrease in acuity is found for horizontal target movement (Ludvigh and Miller, 1953, 1958), vertical target movement (Miller and Ludvigh, 1953; Miller, 1958), and circular target movement (Ludvigh, 1949; Miller, 1956). Decrease in acuity is also found when a moving observer views a stationary target (Miller, 1958; Goodson and Miller, 1959).

Other general findings are that dynamic visual acuity decreases with decreased exposure duration (Elkin, 1962; Miller, 1959; Mackworth and Kaplan, 1962; Crawford, 1960a, 1960b, 1960c) and increases with increased target contrast (Mayyasi et al., 1971). Dynamic visual acuity continues to improve with increasing luminance well above levels for which static acuity has reached an asymptote (Ludvigh, 1949; Miller, 1956, 1958). Males have slightly better dynamic visual acuity than females (Burg, 1966; Burg and Hulbert, 1961; Weissman and Freeburne, 1965). Dynamic visual acuity declines more rapidly with age than does static visual acuity (Burg, 1966; Reading, 1972a, 1972b). The correlation between dynamic and static visual acuity is generally low--i.e., there are large individual differences in dynamic visual acuity among subjects with similar static visual acuities (Ludvigh and Miller, 1954, 1958). The correlation between dynamic and static visual acuities is increased with lower target speeds, binocular viewing conditions, increased exposure durations, and free head movement (DeKlerk et al., 1964; Burg, 1966; Burg and Hulbert, 1961; Weissman and Freeburne, 1965).

Most likely, dynamic visual acuity tends to be poorer than its static counterpart because, at high target velocities, the eyes fail to track the moving target accurately. This explanation was offered by Ludvigh and Miller (1958; Ludvigh, 1949), even though they did not have the benefit of direct measurements of tracking eye movements. Although this explanation has occasionally been questioned (Westheimer and McKee, 1975), it has enjoyed considerable experimental support (Murphy, 1978; Morgan et al., 1983). Of particular practical importance is the fact that Ludvigh's hypothesis seems to be able to account for individual differences in dynamic visual acuity as well as for the substantial changes in it that accompany practice (Murphy, 1978).

Implications

Although the relationship between dynamic and static visual acuities is not well understood and although there has not been an effective standardization of measures of dynamic visual acuity, the working group found much evidence that dynamic visual acuity is often more predictive of real-world task performance than are static acuity measures of vision. So it may be that combining measurement of contrast sensitivity function with dynamic, moving-target testing conditions can lead to more powerful measures of visual assessment that are predictive of visual task performance. DeKlerk et al. (1964) tested 30 pilots in their performance on in-flight measures of instrument, formation, and night flying ability. They reported that dynamic visual acuity correlated more highly than did static visual acuity with each of these performance measures. Burg (1967, 1968) investigated the relationships between a battery of seven vision tests (including dynamic visual acuity, static visual acuity, visual fields, and lateral phorias) and people's automobile driving records. He reported that dynamic visual acuity had the strongest relationship to automobile driving records of all the vision measures studied. Henderson and Burg (1973) studied the accident record of truck and bus drivers and found that there was a significant inverse relationship between dynamic visual acuity and accident record.

Although the relationships between dynamic visual acuity and task performance discussed above are not particularly strong, they are stronger than those found with static measures. The working group concludes that these relationships found despite the coarseness of the measures, the large within-subject variability on dynamic visual acuity tests, and the lack of standards for measuring dynamic visual acuity warrant further investigation. The working group concludes that dynamic visual acuity has real potential for the assessment of vision.

Recommendations

The working group recommends that a study be conducted to investigate the relationship between dynamic visual acuity and some important real-world task such as flying ability. The working group further recommends that a dynamic contrast sensitivity function be measured using grating targets moving with a range of angular velocities, and that it be compared with measurements of static contrast sensitivity function in their relationship to flying ability. Sinusoidal grating targets having the general form of Gabor functions (see Appendix C) would seem useful in the dynamic contrast sensitivity task. It would be highly desirable to measure both eye movements and accommodative states in taking both static and dynamic contrast sensitivity measurements in order to better understand the relationship between them.

Background

In many real-world visual motor tasks, the retinal image of an object expands or contracts, as a result of object motion toward or away from the observer, observer motion toward or away from the object, or a combination of both types of motion. Recent psychophysical (Beverly and Regan, 1975, 1980) and physiological data (Regan and Cynader, 1982) suggest the existence of specialized visual mechanisms sensitive to object movement in depth. Perimetric analysis of sensitivity to motion in depth (Beverly and Regan, 1983) reveals large individual differences and considerable heterogeneity. Some individuals are unable to discriminate an expanding retinal image fromSome individuals are unable to discriminate an expanding retinal image froma a lateral motion, while others are highly sensitive to the difference. Because of these individual differences and because this sensitivity to depth may be highly correlated with a pilot's flying ability (see below), the working group concludes that this area should be more extensively investigated.

In one study, pilots were measured on their sensitivity to motion in depth (Kruk et al., 1981). The pilots viewed a square on a display screen that randomly expanded or contracted in size, simulating motion in depth. The pilots were instructed to keep the size of the square as constant as possible using a control that counteracted the size change. Accuracy in performing this task was taken as an index of depth sensitivity. The pilots were also tested on an Air Force flight simulator on flying tasks such as landing in fog, formation flying, and low-level bombing accuracy under counterattack. Pilots who performed well on the depth-tracking task also performed well on the flying tasks. In another study, sensitivity to depth motion was found to correlate well with flying performance in an actual airplane that was tracked by means of telemetry (Kruk and Began, 1983).

Conclusions and Recommendations

The working group concludes that the depth motion tracking task has potential for screening pilots and others involved in precise visual-motor tasks and that its potential should be further developed and explored. Like measures of dynamic visual acuity, dynamic depth tracking involves both the visual system and the oculomotor system. Superior performance on both of these tasks probably involves the integration of these two systems. The working group believes that the existence of a rather simple dynamic tracking task, which is rather easy to administer and which predicts the ability to perform complex flight tasks, would be a significant development in the assessment of vision.

The working group recommends that further studies of this technique be undertaken. These studies should focus on finding the optimal test conditions giving the highest predictability of actual flight performance as well as other visual-motor tasks such as vehicular driving.