This research aims to prove whether the ability to recognize body parts for a second-grade child with Down syndrome at SLB YPPLB Padang can improve using puzzle media. This study employs a quantitative research type with an experimental approach in the form of Single Subject Research (SSR) with an A-B design. The subject of this research is a second-grade child with Down syndrome at SLB YPPLB Padang. The research was conducted at school and at home. The data analysis technique in this study is visual data analysis by presenting data graphs. The results of the data analysis obtained in the Baseline condition (A) indicate a horizontal tendency of the subjects with a horizontal change level of (+0). The stability tendency in the initial condition is stable with a percentage of 100% over 3 meetings: 37.50%, 37.50%, 37.50%. In the baseline condition (A), it is evident that the direction of data change is stable with no changes before the intervention was conducted. Observations during the intervention condition (B) showed a change level of (+25) which indicates an improvement in the ability to recognize body parts with percentages from 7 meetings: 68.75%, 87.50%, 87.50%, 90.625%, 93.75%, 93.75%, and 93.75%. The data direction shows a significant increase and the percentage of data overlap is at 0%. This illustrates a clearer development of the abilities of children, thus the interventions provided have a positive impact on the variables.