IN A STUDY OF THE EFFECTS ON LEARNING OF A FILM ON OCEANOGRAPHY, 477 GRADE 6 STUDENTS WERE RANDOMLY ASSIGNED TO A CONTROL GROUP NOT VIEWING THE FILM AND TO 12 TREATMENT GROUPS. INDEPENDENT VARIABLES WERE RESPONSE MODE, FEEDBACK (IN FORMS THAT PRESENTED AUDIOVISUAL OR AUDIO-ONLY FEEDBACK, PRESENCE OR ABSENCE OF CORRECT ANSWERS, AND SPACED OR DELAYED MASSED FEEDBACK), AND REVIEW MODE (FOR WHICH NO STUDENT RESPONSES WERE PERMITTED). AN ADDITIONAL GROUP THAT VIEWED A NON-EXPERIMENTAL VERSION OF THE FILM LEARNED AS MUCH AS ANY OF THE OTHER GROUPS IN HALF THE TIME. SCORES ON AN IMMEDIATE POST-TEST OF GROUPS RECEIVING FEEDBACK WERE SIGNIFICANTLY SUPERIOR TO THOSE FOR NO FEEDBACK, BUT NO RESPONSE MODE WAS SUPERIOR UNDER ANY FEEDBACK CONDITION. THERE WERE NO SIGNIFICANT DIFFERENCES AMONG ANY OF THE REVIEW CONDITIONS. SUBJECTS' MENTAL ABILITIES DID NOT AFFECT SCORES, BUT LOW VOCABULARY ACHIEVEMENT DID SHOW AN INTERACTION EFFECT. THE LACK OF SIGNIFICANT EFFECTS WAS INTERPRETED THIS WAY--FREQUENT REPETITION, EVEN IN THE NONEXPERIMENTAL VERSION OF CONCEPTS TO BE LEARNED MAY HAVE MASKED DESIRED EFFECTS OF THE EXPERIMENTAL VARIABLES. (LH)