It has been 60 years since the definitive descriptions of autism, yet it is only in the past decade that related advances in cognitive and basic neuroscience have begun to be incorporated in clinical practice. Some of the resultant clinical advances, which include a trend toward international standardization of diagnosis on the basis of behavioral criteria and which, in turn, seem to allow for earlier, more secure diagnosis and the application of behavioral therapy in early childhood, as well as more thorough genetic studies, are briefly reviewed. The three major defects in thought processing that are postulated by cognitive neuropsychologists to result in aberrant autistic behaviors are also reviewed and linked to recent functional imaging studies in autistic patients and some animal and bench research suggestive of both cortical and subcortical developmental vulnerabilities in autism. Overall it seems at least possible that neuroscientific research may yield results applicable to prevention or remediation of autism, a condition heretofore considered irremediable.