The purpose of this study was to investigate how the use of CAI tutorial programs, incorporating advance organizers and involving various sizes of groups of subjects, would affect students' achievement scores, retention scores, and rates of learning. Used as subjects were 100 suburban high school physics students running interactive tutorial physics programs focusing on strobe simulation and displacement-time and velocity-time graphs. For fifty students (experimental group) an advance organizer program preceded each tutorial. The remaining fifty students (control group) had an advance non-organizer program preceding each tutorial. While pursuing the tutorials the students worked individually or together in groups of two, three, or four. Five days were allocated for the students to repeat the four tutorials until 90% competency level was attained. Achievement and retention were measured by individually administered paper-pencil teacher-made tests sampling the content of all four tutorials. Rate of learning was determined for groups by the number of times the first three tutorial programs were executed in order to attain 90% competency. In the two-way analysis of variance the only significant result at the 0.05 level pertained to group size. Results of the Tukey Test revealed that students working in groups of three and four on CAI tutorials had significantly better rates of learning than students working alone. No significant differences in achievement or retention were observed for the various groups. The implications for using CAI tutorials in physics point to grouping students in fours as a time saving and economic method of presenting material without significant loss of achievement or retention.