Numerous validity studies have shown that the PPVT consistently overestimates mentally retarded persons' IQs. One possible interpretation is that this phenomenon is an outcome of the dissociation between their cognitive level and their experience. Indeed, compared to intellectually average subjects of the same mental age, they have had more learning opportunities, simply because they have lived longer. In order to validate this hypothesis, the French version of the PPVT, Raven's Coloured Progressive Matrices (RCPM), and the 1966 version of the Binet-Simon were administered to 90 subjects of various chronological ages matched on mental age (30 nonretarded 5 year olds and two groups of 30 retarded subjects aged 10 and 16 years, respectively). The results indicate that CA exerts a strong effect on vocabulary, but not on RCPM performance. The research implications of this finding are discussed.