Using non-preferred hand skill to investigate pathological left-handedness in an unselected population.

This paper considers whether pathological influences affect hand-preference among children with epilepsy or mental retardation, for whom there is no evidence of gross motor defect. Data on two tasks of manual skill were analysed for a group of over 12,000 children. The predicted association between poor skill with the non-preferred hand and left-handedness was confirmed for the match-sorting test, and was particularly strong among children with a neurological abnormality (excluding cerebral palsy). No association between poor skill and left-handedness was found for the preferred hand. The data agreed well with a model which estimates hand preference it it occurs on the previously preferred side. The probability of pathological left-handedness among left-handers in an unselected population is about 0.05--this is much higher for children with a history of neurological disease, epilepsy or mental retardation, and for children with poor performance of the non-preferred hand.