AN EXPLORATION OF THE INTERACTION BETWEEN PROGRAMED TEACHING DEVICES AND INDIVIDUAL CHARACTERISTICS OF HIGH SCHOOL STUDENTS WAS UNDERTAKEN. THERE WERE FOUR PARTS -- PREPARATION OF MATERIALS ON GEOMETRY, THE ESTABLISHMENT OF CRITERION MEASURES, PRELIMINARY PILOT EXPERIMENTS, AND THE MAIN EXPERIMENT. THE MAIN EXPERIMENT WAS PERFORMED IN 27 CLASSES IN 3 HIGH SCHOOLS. THE VARIABLES TESTED WERE ACHIEVEMENT CHARACTERISTICS (OVER, NORMAL, AND UNDERACHIEVERS) OF APPROXIMATELY 250 STUDENTS, AND THE ROTE AND CONCEPTUAL FORMS OF THE PROGRAM. THE CONCLUSIONS WERE (1) THE DIMENSIONS FOR SPECIFYING THE LEARNING TASK, METHOD, AND STUDENT HAVE NOT BEEN IDENTIFIED, (2) A PERIOD OF INFORMAL HYPOTHESIS DEVELOPMENT SHOULD PRECEDE FORMAL EXPERIMENTS, AND (3) THE POTENTIAL ADVANTAGES OF PROGRAMED INSTRUCTION, AND PARTICULARLY OF HIGHLY RESPONSIVE PROGRAMED INSTRUCTION, ARE NOT EFFECTIVE AT THE SECONDARY SCHOOL LEVEL. ATTEMPTS TO KEEP METHODS WITHIN THE PROGRAMED INSTRUCTION FRAMEWORK HAVE IMPAIRED THE EFFECTIVENESS OF THE INSTRUCTION. FUTURE RESEARCH WAS SUGGESTED TO DETERMINE THE EFFECTIVE COMBINATION OF INSTRUCTIONAL TECHNIQUES AND METHODS FOR SPECIFIED TASKS AND FOR STUDENTS OF SPECIFIED CHARACTERISTICS. (RS)