Retest variability of human infant contrast sensitivity: How many tests are sufficient?

Retest variability of a new infant contrast sensitivity (CS) card procedure was assessed by binocular measurement of a group of 20 6-month-olds twice within a 1-week period. Coefficient of reliability analyses showed that within-subject variability between tests was only slightly less than variation across subjects, which suggests that results from a single test are a poor predictor of an infant's "true" visual functioning. To determine how many tests are needed to estimate when infant CS stabilizes to within an acceptable (0.15 log unit) criterion, a second experiment was conducted in which a small group of subjects was tested repeatedly over a 2-week period. The results showed that averaging performance on 2 to 3 tests was required before an accurate estimate of the subject's performance could be obtained. Our results suggest that caution should be taken in the interpretation of a single measurement of infant visual functioning.