Four-, five-, and six-year-old children were used as subjects in this investigation. There were 192 experimental and 96 control children used, divided equally between the three age groups. The experimental children received a 1-minute pretest exposure procedure in which 1/4 of the children observed 4 two-dimensional stimuli (irregular pentagons), 1/4 observed 4 three-dimensional stimuli, 1/4 manipulated the two-dimensional stimuli, and 1/4 manipulated the three-dimensional stimuli. The control group received no pretest condition. The test condition given to all children involved choosing the one of four comparison stimuli that matched a standard stimulus. For half of each group, experimental and control, the stimuli were two-dimensional, and for the other half, three-dimensional. The results indicated that there was a significant age main effect; that is, the older they were, the faster the children learned the test task. Also, children found it easier to discriminate among two-dimensional stimuli than among three-dimensional stimuli. The absence of a significant main effect of pretest exposure to stimulus dimensionality and/or of exposure to a manipulation condition was contrary to the experimental hypothesis. No significant difference was found between the experimental group and control group in discrimination learning ability. (WD)