An alternating treatments design with a best treatments phase was used to compare two active student response (ASR) conditions and one on-task (OT) condition on the acquisition and maintenance of social studies facts during computer-assisted instruction. Each week for six weeks, five students were provided daily computer-assisted instruction on 21 unknown facts divided randomly into Clicking-ASR (active responses with computer mouse), Repeating-ASR (active oral responses) or Listening-OT (on task or passive responses). For all five students, Repeating-ASR resulted in more facts correct on same-day, next-day, and one-and two-week maintenance tests. During weeks 7 and 8 with implementation of the best treatment condition, Repeating-ASR produced higher scores than all conditions (including Repeating-ASR) during the first 6 weeks. (Contains 8 figures and 2 tables.)