THE PURPOSE OF THIS RESEARCH WAS TO STUDY THE EFFECTS OF DIFFERENT REWARDS FOR INTENTIONAL LEARNING ON INCIDENTAL LEARNING ACQUIRED BY PRESCHOOL EDUCABLE MENTAL RETARDATES IN A GAME SITUATION. AN ADULT EXPERIMENTER TAUGHT EACH RETARDATE TO PLAY MOTOR, SOCIAL AND PROBLEM-SOVLING GAMES (INTENTIONAL LEARNING.) A SECOND ADULT SERVED AS A PLAYER AND EXHIBITED MANNERISMS (INCIDENTAL LEARNING) WHILE PLAYING. TO EMPHASIZE THE RULES, THIS PLAYER MADE ERRORS AND WAS CORRECTED BY THE EXPERIMENTER. THE PLAYERS EITHER WERE GIVEN TANGIBLE, SYMBOLIC, OR SOCIAL REWARDS, OR WERE NOT REWARDED. THEY RECEIVED ONLY ONE TYPE OF REWARD IN ANY ONE GAME. IT HYPOTHESIZED THAT THE HIGHEST INTENTIONAL AND LOWEST INCIDENTAL SCORES WOULD RESULT FROM TANGIBLE REWARDS. THE BEST INTENTIONAL-INCIDENTAL SCORE COMBINATION WOULD RESULT FROM SOCIAL AND SYMBOLIC REWARDS. THE HIGHEST INCIDENTAL SCORES WOULD OCCUR IN THE NO-REWARD CONDITION. THE FOLLOWING RESULTS WERE OBTAINED. IN EXPERIMENT ONE INTENTIONAL LEARNING DID NOT VARY AS A FUNCTION OF TYPE OF REWARD. SUCCESS IN THE GAME APPEARED TO BE A MORE POWERFUL REINFORCER THAN THE REWARDS OFFERED BY THE EXPERIMENTER (THE RETARDATE TYPICALLY EXPERIENCES SOCIAL PLAY DEPRIVATION AND FAILURE IN GAME SITUATIONS). IN EXPERIMENT TWO ALL RETARDATES WERE ACCUSTOMED TO SUCCESS IN GAME SITUATIONS. THE HIGHEST INTENTIONAL SCORES RESULTED FROM TANGIBLE REWARDS. THE HIGHEST INCIDENTAL SCORES OCCURRED IN THE NO-REWARD CONDITION. THE INCIDENTAL SCORES IN THE TANGIBLE AND SOCIAL REWARD CONDITIONS DID NOT DIFFER. BOTH TANGIBLE AND SOCIAL REWARDS WERE ASSOCIATED WITH A SATISFACTORY INTENTIONAL-INCIDENTAL SCORE COMBINATION. DESCRIPTIONS OF THE GAMES ARE PRESENTED IN THE APPENDIX. A REFERENCE LIST INCLUDES 32 ITEMS. (AUTHOR)