THIS REPORT DESCRIBES A SERIES OF EXPERIMENTS TO ASSESS THE USEFULNESS OF TELEVISION RECORDINGS IN IMPROVING TEACHING PERFORMANCE. OBJECTIVES OF THE STUDY ARE (1) TO COMPARE THE EFFECTS OF SELF-EVALUATION OF A TEACHING PERFORMANCE WITH FEEDBACK PROVIDED BY A SUPERVISING INSTRUCTOR, (2) TO COMPARE THE EFFECTS OF REINFORCEMENT DELAY, AND (3) TO COMPARE THE EFFECTS OF A PERCEPTUAL MODELING DEMONSTRATION OF A DESIRED BEHAVIOR WITH THOSE PRODUCED BY PROVIDING A WRITTEN DESCRIPTION OF THE BEHAVIOR AND TO COMPARE THE EFFECTS OF COMBINING REINFORCEMENT WITH EACH. EACH OBJECTIVE WAS THE SUBJECT OF A SEPARATE EXPERIMENT CONDUCTED UNDER HIGHLY CONTROLLED, LABORATORY-LIKE CONDITIONS. IT IS CONCLUDED THAT THE RESULTS OF THIS STUDY SUPPORT THE ASSUMPTION THAT THE RATE AND LEVEL OF LEARNING A GIVEN TEACHING STRATEGY VARY AS A FUNCTION OF THE MODE OF MODEL PRESENTATION. THERE IS EVIDENCE TO INDICATE THAT PERCEPTUAL MODELING PROCEDURES ARE CHARACTERIZED BY DISTINCTIVE CUING PROPERTIES WHICH TEND TO RECOMMEND THEM OVER SYMBOLIC MODELING PROCEDURES FOR USE IN TRAINING CONTEXTS ANALAGOUS TO THOSE DESCRIBED IN THE EXPERIMENT. (HW)