THIS STUDY INVESTIGATED THE EXTENT TO WHICH SOUND FILM RECORDINGS OF PUPILS IN TEACHING-LEARNING SITUATIONS CAN BE USED AS A BASIS FOR TRAINING TEACHERS TO IMPROVE THEIR ACCURACY IN INTERPRETING NONVERBAL COMMUNICATIONS, AND THE EXTENT OF DIFFERENCES AMONG TEACHERS, NEW TEACHERS, AND NONTEACHERS IN THEIR ABILITY TO INTERPRET NONVERBAL COMMUNICATIONS. EQUIPMENT AND METHODS WERE DEVISED TO RECORD INSTANCES OF STUDENT CLASSROOM RESPONSE AND SEVERAL EXPERIMENTS CONDUCTED. SUBJECTS WERE SHOWN SOUND FILM CLIPS OF TEACHER-STUDENT INTERACTIONS AND WERE ASKED TO RATE THE AMOUNT OF STUDENT COMPREHENSION IN THE CLIPS. SOME SUBJECTS WERE SHOWN ONLY THE PICTURE TRACK, SOME HEARD ONLY THE SOUND, AND OTHERS SAW AND HEARD THE COMPLETE RECORDING. THE RESULTS INDICATE THAT VERBAL CUES CONTRIBUTE SIGNIFICANTLY TO THE ACCURACY OF JUDGMENT, WHILE NONVERBAL CUES, SPECIFICALLY VISUAL ONES, DO NOT INCREASE ACCURACY OF JUDGMENT. THERE WAS NO EVIDENCE THAT THE USUAL TEACHER TRAINING PROGRAM IMPROVED ACCURACY OF JUDGMENTS. FURTHER RESEARCH IN THE AREA WAS RECOMMENDED. (JC)