Testing 20 boys with the fragile X (or Martin-Bell) syndrome with the Kaufman Assessment Battery for Children (K-ABC) showed a consistent pattern of strengths and weaknesses that may be useful in predicting a fragile X positive result from cytogenetic testing. This K-ABC pattern included 1) Sequential Scale score less than the Simultaneous Scale score; 2) Mental Processing Composite less than the Achievement Scale score; 3) Spatial Memory subtest score less than the Matrix Analogies subtest score; and 4) Arithmetic subtest score less than the mean of the Achievement subtest scores. A comparison group of 20 boys did not demonstrate such a pattern. Testing with the K-ABC should be considered for boys who present as learning disabled, hyperactive with attentional problems, or mildly retarded. Boys with three or four of the four features of the K-ABC fragile X pattern should be considered for medical evaluation and cytogenetic testing.

Previous studies have reported a drop of IQ in males with Martin-Bell or fragile X syndrome during childhood and adolescence. It is uncertain when and why this drop occurs and whether it affects all fragile X males. We have analyzed longitudinal IQ data on all 24 fragile X males who have been followed for at least 2 years through the Child Development Unit in Denver and who have been evaluated with a Stanford-Binet Form L-M assessment. A significant drop in IQ occurred for the group as a whole, but an individual z analysis demonstrated that only 7 of 24 had a significant decline in IQ. Visual inspection demonstrates an initial decline in IQ beginning in middle childhood and continuing through adolescence. We postulate that this drop occurs because of relatively greater weaknesses with abstract reasoning and higher symbolic language skills that are stressed in the cognitive testing of later childhood and adolescence.

We investigated the possibility that fra(X) heterozygotes had a distinct or specific set of mental deficits ("cognitive profile") which would allow for accurate diagnosis. Wechsler Intelligence Scale for Children-Revised (WISC-R) subtest scores obtained on 8 fra(X) school age girls were compared with similar scores obtained on 8 "learning-disabled" non fra(X) girls matched on the basis of Full Scale IQ (FSIQ). The Block Design subtest score was significantly lower in fra(X) girls. In a larger sample of 22 fra(X) females, a characteristic combination of low Arithmetic, Digit Span, and Block Design subtest scores was observed. The mean discrepancy between these 3 subtest scores from the total Verbal or Performance subtest means was significant for the fra(X) group but not for a comparison group of 20 learning-disabled females. Verbal IQ (VIQ) and Performance IQ (PIQ) discrepancy was not significant in fra(X) females. Percent fra(X) positive cells was negatively correlated with VIQ and FSIQ but not with PIQ.