The project's purpose was to determine whether attention to the task during testing was a confounding variable in measures of visual perception ability. Samples of 30 perceptually handicapped (PH) and 30 normal subjects (N) were randomly selected from children so classified on the Frostig DTVP, providing they had IQ scores between 85 and 115 on the Peabody Picture Vocabulary Test. The two samples were further stratified on sex and race. A test of visual memory for letter-like forms was administered to both samples in two presentations, one a group paper-and-pencil test and the other a machine presentation which provided reinforcements for correct responses. An "X Design" controlled for practice effect between the two presentations. Analysis of covariance was performed co-varying the visual memory and IQ scores for both presentations. No significant differences in performance were found on either presentation of the visual memory tests, between the PH and N groups, between the sexes, or in the interaction of perception ability and sex. There was insufficient evidence to conclude that attention to the task was an important variable in perception testing. However, the use of DTVP in testing perceptual ability was found to be highly questionable. For related documents, see TM 002 117, 118.)