A forced-choice test improves clinical contrast sensitivity testing.

Decreased contrast sensitivity has been demonstrated in early glaucoma, but the deficit in not regularly observed. We designed a prototype for a forced-choice printed test and evaluated it with several other measures of contrast sensitivity. The results also bear on the pattern of loss and the variables which effect performance. Mildly glaucomatous patients show at 6 db (50%) loss of contrast sensitivity at all spatial frequencies tested compared with age matched controls. Moving gratings give the same information as stationary ones, and practice effects are negligible. Contrast sensitivity at or below 2 cycles/degree is poorly correlated with visual acuity and does not change with age in the forced-choice test. Subjective judgment made the apparent contrast threshold higher, age dependent, and more variable, particularly at higher spatial frequencies. The pattern of variability can explain some reports of insignificant effects and why low spatial frequency contrast sensitivity detects glaucoma better than visual acuity. Methods correlated so poorly, despite high reliabilities, that uncontrolled biases must be suspected in subjective measures. Our new forced-choice format was superior to all other tests on at least one formal criterion and always at least equal to them. Improvements in contrast sensitivity screening tests are thus indicated.