Educable retarded (64) and nonretarded (64) subjects (mean age 15 years) heard two stories. After each story, performance was assessed on recall and comprehension. All Ss were randomly assigned to one of four treatments: (1)knowledge (Ss were told concepts critical to processing each story); (2)purpose statements (Ss were guided in organizing incoming information); (3)knowledge plus purpose statements; and (4)control (no information). The comprehension measures included comprehension of central, noncentral, and implied story content; recall measures included recall of central and noncentral units. Comprehension factors were: (1)population (retarded vs. nonretarded); (2)treatment (knowledge, purpose statements, knowledge plus purpose statements, control); (3) placement of information (before vs. during); and (4)order of placement (before first vs. during first); and (5)question type (central, noncentral and implied). The independent variables in the analysis of recall were identical to the analysis for comprehension, except that recall unity type (central vs. noncentral) was substituted for comprehension question type. Results indicated that nonretarded Ss performed better than retarded Ss on both comprehension and recall measures. Comprehension of central questions was significantly better than comprehension of noncentral questions, which was significantly higher than comprehension of implied question. Recall of central units was superior to recall of noncentral units. Results indicated that among both retarded and nonretarded Ss comprehension scores were significantly higher in the knowledge plus purpose statement condition than in any of the other experimental conditions. Treatment effects were not significant for the recall analysis. However, several significant interactions were obtained. It was concluded that knowledge plus purpose statements has a facilitative effect on the comprehension abilities of both retarded and nonretarded Ss, but not on their recall abilities. (Author/CL)