VISUAL ACUITY 



35 



luminance is less than that of the task, 

 visual performance is not consistently- 

 affected one way or the other. Cobb's 

 data (11) on luminance sensitivity — the 

 smallest luminance difference which can be 

 seen — are in essential agreement with those 

 of Johnson, except that Cobb appears to 

 find a somewhat more consistent, though 

 still very small, decrease in visual perform- 

 ance for surrounds which have a luminance 

 lower than that of the task. Although raw 

 data are not given, the present writer 

 questions whether the differences are 

 statistically significant. 



Fisher's study (27) is interesting because 

 it attempted to provide quantitative data on 

 the influence of the size of the surrounds. 

 His central task area had a constant diam- 

 eter of 2°. Five different surround sizes 

 were used: 2.5°, 5°, 7.5°, 12.5°, and 20° in 

 radial width measured beyond the edge of 

 the central task area. His results show 

 that when the surrounds are brighter than 

 the task, acuity decreases with increasing 

 size of the surrounds. When the surrounds 

 are dimmer than the central task, acuity 

 increases with increasing size of surrounds. 

 When the surrounds are equal to the central 

 task in luminance, increasing the size of the 

 surrounds has no consistent effect on 

 acuity. It should be noted, incidentally, 

 that if his data are analyzed in terms of the 

 task/surround luminance ratio, in only a 

 few instances was acuity best when the 

 surround and task luminances were equal. 



This problem is discussed here in some 

 detail because it is a good illustration of the 

 need for more complete studies of many 

 visual functions. Once specifications have 

 been prepared for illuminances to be used in 

 various seeing situations, they have an air 

 of finality about them. The reader gets the 

 impression that the specifications stem from 

 sound experimental evidence. It is rather 

 startling at times, therefore, to discover how 

 meager the evidence really is. 



Chromatic Contrast. A factor which is 

 largely ignored in studies of visual acuity is 



chromatic contrast. Walls (93, p. 496) puts 

 the situation in an amusing fashion : 



A gentle indictment can be laid against all 

 the investigators who have wrestled with the 

 problems of resolution and come away from the 

 strife proudly clasping theories to their bosoms. 

 One and all, they seem to be color-blind. 



The gap in our knowledge on this point is 

 serious. One gets the impression from 

 reading most authors that luminance and 

 luminance contrast are the really important 

 factors in determining visual acuity. But 

 it is easy to demonstrate that visual acuity is 

 possible when a red object is placed on a 

 green background of the same luminance. 



A start was made on this problem during 

 the war in a research contract held by the 

 Eastman Kodak Company (101). It was 

 admittedly only a beginning and this work 

 needs to be pursued further. A tentative 

 conclusion which came out of the Eastman 

 work was that luminance contrast and 

 chromatic contrast both contribute to 

 visual acuity according to this equation : 



Ce = VCz.- + Cc', 



where Ce is the effective total contrast, Cl 

 is the luminance contrast, and Cc is the 

 chromatic contrast. Whether this gen- 

 eralization will be valid when more data are 

 accumulated is difficult to say. One thing 

 is clear, however; if we do not take chromatic 

 contrast into account, we are probably 

 ignoring an important determinant of visual 

 acuity. 



Visual Acuity as a Function of Color of 

 Illumination. A special case of the problem 

 of chromatic contrast concerns visual acuity 

 as a function of the color of illumination. 

 Usually, the problem is one involving black 

 or gray test objects seen against a white 

 background illuminated with different col- 

 ored lights. This means, in general, that 

 the background takes on the color of the 

 light, so that chromatic contrast varies. 

 In addition, the reflectance of the black test 

 object is usually much less than that of the 



