SS WERE PRESENTED A LIST OF 40 CRITICAL (CR) WORDS WITH INSTRUCTIONS TO REMEMBER THE WORDS BUT NOT THE ORDER. A RECOGNITION TEST FOR EACH CONSISTED OF A HIGH ASSOCIATE (A) WORD, A FORMALLY SIMILAR (F) WORD, AND A NEUTRAL (N) WORD. ALL SS PRONOUNCED THE WORDS ON THE LEARNING TRIAL, AND 1 GROUP (F) WAS ALSO REQUIRED TO PRONOUNCE AN ADJACENT WORD WHICH WAS EITHER AN A, F, OR N WORD, OR NUMBER. GROUP F AND 1 OTHER GROUP HAD AN IMMEDIATE RECOGNITION TEST. 2 OTHER GROUPS HAD RECOGNITION TESTS 24 HR. AFTER LEARNING, 1 OF THESE HAD AN IMMEDIATE FREE RECALL TEST. FOR GROUP F THE NATURE OF THE ERRORS WAS DETERMINED BY THE NATURE OF THE ADJACENT ITEM, BUT CORRECT RECOGNITION DID NOT DIFFER AS A FUNCTION OF THE TYPE OF ADJACENT ITEM. HOWEVER, THE CONFIDENCE RATINGS OF CORRECTNESS INDICATED COMPENSATORY FACTORS IN THE CASES OF A AND F WORDS. WITHOUT FORCED ADJACENT ITEMS ON THE LEARNING TRIAL, F WORDS HAD NO GREATER INFLUENCE THAN N WORDS, THE A WORDS PROVIDING THE MAJOR SOURCE OF ERROR. FREE RECALL SLIGHTLY ENHANCED 24-HR RECOGNITION FOR THE WORDS RECALLED. 24% FORGETTING WAS OBSERVED OVER 24 HR. WHEN MEASURED BY CORRECT RESPONSES, 18% BY CONFIDENCE RATINGS OF CORRECT RESPONSES. (PsycINFO Database Record (c) 2006 APA, all rights reserved)