VERBAL CUEING UNDER CONDITIONS COMPARABLE TO THE CLASSROOM WAS STUDIED TO TEST THE HYPOTHESES THAT THE SPECIFICITY OF VERBAL CUES DURING TRAINING IS INVERSELY RELATED TO BREADTH OF TRANSFER OF THE ABILITY TO DISCOVER A PROBLEM SOLUTION IN THE ABSENCE OF CUES. THREE DIFFERENT EXPERIMENTS WERE CARRIED OUT, EACH UTILIZING 50 TO 150 CHILDREN AS SUBJECTS. EACH SUBJECT WAS REQUIRED TO SELECT (IN MULTIPLE-CHOICE FASHION) A PICTURE, FIGURE, OR A WORD WHICH MATCHED A MODEL WITH RESPECT TO A DIMENSION TO BE DISCOVERED. ON CRITERION TESTS, SUBJECTS WERE GIVEN NO CUES. THEY WERE REQUIRED TO FIND, THROUGH THEIR OWN RESOURCES, THE APPROPRIATE BASIS FOR MATCHING. CONCLUSIONS INDICATED THAT--(1) THE CLASS CUE GROUP WAS ABLE TO PERFORM WELL ON THE POST-TESTS, (2) SOME EVIDENCE REGARDING THE MOTIVATIONAL RESULTS OF LEARNING BY DISCOVERY WAS UNCOVERED, (3) THE INVESTIGATOR HAD AN ACUTE PROBLEM IN THAT THE STUDY OF CERTAIN VARIABLES WERE AT THE EXPENSE OF OTHERS, (4) INDICATIONS WERE THAT WHETHER OR NOT DISCOVERY LEARNING IS EFFECTIVE PROBABLY DEPENDS UPON THE LEARNER'S HISTORY OF ASSOCIATIONS TO PROBLEMS AND THE CUES AS WELL AS UPON HIS ACQUIRED METHODS OF DISCOVERING, (5) THE EFFECTIVENESS OF A CUE MAY BE PREDICTED ON THE BASIS OF ITS PREVIOUS ASSOCIATIONS AND ON THE TYPE OF TEST USED TO MEASURE ITS EFFECT, (6) CHILDREN CAN BE TAUGHT TO DISCOVER SOLUTIONS TO CERTAIN TYPES OF PROBLEMS, AND (7) THE PROCEDURE OF DEFINING ABSTRACT CONCEPTS SUCH AS "DISCOVERY" IN OPERATIONAL TERMS OF INDEPENDENT AND DEPENDENT VARIABLES ALLOWS ONE TO ANALYZE THE FUNCTIONAL RELATIONS BETWEEN THEM. (HB)