The proposed objectives have the following aims: (1) to determine whether a shift in the distribution of handedness exists in an unselected sample of autistic children and adults; (2) if so, to determine its underlying constructs; (3) to determine whether the constructs are specific to autism or whether they relate more generally to overall mental retardation irrespective of autistic symptomatology and (4) to determine whether these phenotypes provide potential neurobiological markers of different etiological subgroups within or between disorders. Subjects will consist of two unselected autistic samples as follows: (a) UCLA Sample (n=50) which comprises a younger and higher functioning outpatient group; (b) Camarillo Sample (n=80) which comprises an older and lower functioning inpatient group. Two lower and higher functioning MR non-autistic samples will be selected from UCLA and Camarillo State Hospital as comparison matches for the two target samples. The research design involves a rigorous one week test-retest assessment of handedness using multiple items normed for developmentally retarded Ss. These measures will be used to investigate Aims 1-3 which essentially involve the identification of various handedness phenotypes. A unique feature of this project, including the assessment of handedness, concerns an extension of the PLH model (Satz, 1972) to deal with multiple handedness phenotypes. This revised model provides a conceptual and quantitative framework for dealing with potential etiological subgroups of autistic and/or MR subjects. The subtypes will first be validated against external behavioral criteria to determine whether they correlate with different neurobehavioral outcomes (e.g., adaptive behavior, intelligence, familial handedness, trophic changes in extremities). If group differences occur, as preliminary results suggest, then a random subset of each phenotype will be given high resolution computerized tomography to determine whether these phenotypes confer any significance as biological markers of etiological subtypes in infantile autism. The proposal seeks to explain some of the known heterogeneity that has long plagued research in this disorder(s).