Disenchantment with traditional intelligence test has prompted an examination of alternative assessment procedures that accurately and adequately measure children\'s intellectual ability. In view of this important concern over better assessment and intervention approaches, the present paper examine the effectiveness of error-analysis based intervention on the enhancement of cognitive function on Raven\'s Standard Progressive Matrices of children with intellectual disabilities. Forty-five children with intellectual disabilities from the second and third grades children in three elementary schools were assigned to three groups. One group was control group;a second, experimental group 1 received a traditional didactic mediation with Raven\'s Standard Progressive Matrices for training materials;and third group, experimental 2 group received an error-analysis based intervention Raven\'s Standard Progressive Matrices. Results indicated that two experimental groups out-performed the control, and experimental 2 group received an error-analysis based intervention was not significantly different from that of the experimental group 1 received a traditional didactic mediation