AN AUTOMATED INSTRUCTIONAL SYSTEM WAS DEVELOPED TO TEACH BEGINNING READING TO HEARING IMPAIRED CHILDREN USING A NON-ORAL METHOD. INSTRUCTION WAS DONE WITH VISUAL PRESENTATION USING 35MM SLIDES ON A REAR PROJECTION SCREEN. THIS TEACHING MACHINE, THE HONEYWELL UNIVERSITY OF MINNESOTA INSTRUCTIONAL DEVICE (HUMID) WAS CONSTRUCTED WITH AN AUTOMATIC DATA RECORDER AND PRINTOUT DEVICE. THE FIRST OF THREE STUDIES DISCOVERED THAT INSTRUMENTATION AND PROGRAMING NEEDED TO BE IMPROVED. TWO PRELIMINARY STUDIES WERE CONDUCTED. THE THIRD STUDY INVOLVED ONE EXPERIMENTAL GROUP OF 10 FIRST-GRADERS AND ONE EXPERIMENTAL GROUP OF TEN 9 AND 10 YEAR OLDS ENROLLED IN CLASSES FOR THE DEAF AND HARD OF HEARING. CONTROL GROUPS WERE MATCHED FOR IQ, AGE, SEX, HEARING AND LANGUAGE PERFORMANCE. OVER A PERIOD OF 35 DAYS, 34 PROGRAMS WERE TAUGHT TO THE EXPERIMENTAL GROUPS. TESTING DID NOT REVEAL ANY STATISTICALLY SIGNIFICANT DIFFERENCES BETWEEN THE TWO EXPERIMENTAL GROUPS, ALTHOUGH THE OLDER GROUP PERFORMED CONSISTENTLY BETTER THAN THE YOUNGER GROUP. ON THE HUMID POST-TEST, THE FIRST-GRADERS PERFORMED SIGNIFICANTLY BETTER (AT THE .01 LEVEL) THAN THEIR CONTROL GROUP. THE OLDER GROUP ALSO PERFORMED BETTER THAN ITS CONTROL GROUP, ALTHOUGH THE DIFFERENCE WAS NOT STATISTICALLY SIGNIFICANT. ON THE STANFORD ACHIEVEMENT TEST SUBTESTS OF WORD MEANING AND PARAGRAPH MEANING GIVEN TO ALL FOUR GROUPS, ONLY THE YOUNGER GROUP SURPASSED THEIR CONTROLS AT A LEVEL WHICH APPROACHED SIGNIFICANCE. THERE WERE NO STATISTICALLY SIGNIFICANT DIFFERENCES BETWEEN THE OLDER EXPERIMENTAL AND CONTROL GROUPS. ALTHOUGH AN AUTOMATED SYSTEM OF NON-ORAL READING INSTRUCTION CAN BE DEVELOPED SUCCESSFULLY, IT WOULD INVOLVE A TREMENDOUS AMOUNT OF PROGRAMING. THE HUMID STAFF ESTIMATES THAT TO BRING DEAF CHILDREN FROM BEGINNING READING TO FOURTH GRADE READING WOULD REQUIRE MORE THAN 800 PROGRAMS OF 30 TO 40 FRAMES EACH AND MUST ALSO BE ACCOMPANIED BY A SYSTEMATIC PROGRAM OF LANGUAGE DEVELOPMENT. FIVE STUDIES WERE CARRIED OUT WITHIN THIS PROJECT BY GRADUATE STUDENTS. THE MAJOR FINDING WAS THAT DATA ON THE RESPONSE DELAYS HAS LIMITED USEFULNESS WITH THIS METHOD OF TEACHING READING. PROGRAMING TECHNIQUES, CONTENT OF PROGRAMS, AND INFORMATION ON THE TECHNICAL DEVELOPMENT OF HUMID ARE PRESENTED. REFERENCE LIST CITES 34 ITEMS. (MW)