The present French-German investigation of fragile-X syndrome (fra-X) was undertaken to disentangle genetic from environmental effects on cognitive performance as assessed with the following measures: Wechsler Adult Intelligence Scale-Revised (WAIS-R), Wisconsin Card Sorting Test, Trail-Making Test, Tower of Hanai, Verbal Fluency Test, Stroop Test, short-term and consolidation memory, and the d2 task. Groups with different genotypes (n=11 mothers with a full mutation in the FMR-1 gene of fra-X children; n=65 mothers with a premutation in the FMR-1 gene of fra-X children; n=18 siblings of these mothers with normal CGG repeats) and with different psychosocial stressors from fra-X families (n=14 siblings with a premutation but without affected children of their own) were examined. A group of mothers of non-fra-X autistic children (n=39) formed an external control group. Previous findings were replicated concerning cognitive performance of FMR-1 full-mutation carrier mothers, who were characterized by lower overall IQ and poorer performance than the group of mothers with the FMR-1 premutation in verbal and performance subtests of the WAIS-R, tests of executive-frontal lobe functioning, and tests of sustained attention. Carriers of the FMR-1 premutation, whether they were mothers of affected children or not, performed in a similar way on all neuropsychological tasks to the intrafamilial control group without CGG amplification. On the basis of these results, it is concluded that there is no neuropsychological evidence of reduced cognitive performance of FMR-1 premutation carriers compared with performance of two control groups with normal CGG repeats. Furthermore, the psychosocial burden of raising fra-X children does not exert an environmental effect on neuropsychological test performance. Copyright (C) 1999 Elsevier Science Ireland Ltd.