The aim of this study is to examine the validity of the Autism Spectrum Quotient (AQ) to differentiate high-functioning autistic spectrum disorder (ASD) from schizophrenia (SCH). The AQ was developed by Baron-Cohen et al. to measure autistic traits. In addition to the original AQ items, we created self-administered questions about psychotic symptoms (S-scale). We administered the modified AQ to 51 ASD patients and 46 SCH patients, and we compared these two groups in terms of total AQ score, AQ subscale scores and S-scale score. We applied receiver operating characteristic (ROC) curves to examine the discriminating power of the AQ. The mean total AQ score of the ASD group (32.6; SD=6.8; range: 8-48) was significantly higher than that of the SCH group (21.8; SD=7.4; range: 10-39) (p<0.001). All AQ subscale scores of the ASD group were significantly higher than those of the SCH group. By using a cut-off score of 29 for the AQ total score, we were able to correctly classify 80% of the subjects. At this cut-off, the positive and negative predictive values were 0.83 and 0.78, respectively. Inclusion of additional questions of the S-scale did not increase the power of differentiation. These results indicate that the usefulness of the AQ in differentiating high-functioning ASD from SCH is limited.