Identification of meaningful endophenotypes may be critical to unraveling the etiology and pathophysiology of autism spectrum disorders (ASD). We investigated whether impressions of "humanness" for android robot might represent a candidate characteristic of an ASD endophenotype. We used a female type of android robot with an appearance similar to that of a real person. Significant differences in overall impressions of "humanness" for android robot were found between adolescents with ASD and typical development (TD) controls, as well as parents of children with ASD and parents of TD controls. Our current work does suggest robotic systems could potentially play an intelligent role in dissecting ASD heterogeneity.