DEPENDENCY LEARNING, THE ACQUISITION AND DEVELOPMENT OF PERSONAL INTERRELATIONSHIP VALUES, WAS STUDIED AS A TECHNIQUE FOR FOSTERING AUDIOVISUAL ACADEMIC LEARNING AMONG 54 YOUNG, EDUCABLE MENTAL RETARDATES. SOME OF THESE SUBJECTS WERE TAUGHT TO VALUE SIMULATED DEPENDENCY MODELS. THESE MODELS WERE CONSISTENTLY PAIRED WITH SUCH REWARDING STIMULI AS GAMES, TOYS, AND EDIBLES. AFTER SEVERAL WEEKS OF ACQUIRING DEPENDENCY WITH THE MODELS, THESE CHILDREN AND OTHER SUBJECTS, HAVING NO EXPOSURE TO THE MODELS, WERE SHOWN A SERIES OF AUDIOVISUAL PRESENTATIONS IN WHICH A MODEL FIGURE PERFORMED NOVEL RESPONSES. MEASURES OF THE EXTENT TO WHICH THE SUBJECTS LEARNED THESE RESPONSES WERE MADE IMMEDIATELY FOLLOWING THE AUDIOVISIUAL PRESENTATIONS, AND 1 WEEK LATER. STATISTICAL COMPARISONS WERE MADE OF THE LEARNING AND RETENTION SCORES WHEN TASKS WERE PRESENTED (1) BY A VALUED MODEL VERSUS AN UNFAMILIAR MODEL, (2) IN INDIVIDUAL VERSUS GROUP SETTINGS, (3) UNDER INTENTIONAL VERSUS INCIDENTAL LEARNING CONDITIONS, AND (4) BY A TEACHER VERSUS A VALUED MODEL. RESULTS INDICATED THAT SUBJECTS WHO VALUED THE FILM MODEL PERFORMED BETTER THAN SUBJECTS WHO WERE EXPOSED TO AN UNFAMILIAR FILM MODEL. ALL SUBJECTS, HOWEVER, WERE ABLE TO LEARN FROM THE AUDIOVISUAL PRESENTATIONS. GROUP VERSUS INDIVIDUAL VIEWING CONDITIONS AND INTENTIONAL VERSUS INCIDENTAL LEARNING CONDITIONS, WHEN COMPARED, APPEARED TO HAVE LITTLE AFFECT ON THE CHILDREN'S OVERALL PERFORMANCE. ALL SUBJECTS OBTAINED HIGHER SCORES UNDER THE TEACHER PRESENTATION AS COMPARED TO THE AUDIOVISUAL PRESENTATIONS USED IN THIS STUDY. (JH)