MEDIATION AND PSEUDOMEDIATION: A REPLY TO HORTON.

D. L. HORTON (SEE 42:4) HAS ATTEMPTED TO REJECT B. EARHARD AND G. MANDLER'S (SEE 40:3) INTERPRETATION OF DATA FROM EXPERIMENTS WHERE "MEDIATION" HAS BEEN PRODUCED BY STANDARD PAIRED-ASSOCIATE PROCEDURES. IT IS ARGUED THAT (1) HORTON'S REVIEW OF UNLEARNING DATA IS INADEQUATE AND THAT EVIDENCE OF UNLEARNING IS PRESENT IN STUDIES HE CITES AS PROVIDING NO EVIDENCE OF UNLEARNING; (2) HIS EVALUATION OF THE INTERLIST INTERFERENCE INTERPRETATION OF DIFFERENCES BETWEEN EACH OF THE MEDIATED FACILITATION, MEDIATED INTERFERENCE, AND PSEUDOMEDIATION PARADIGMS, AND THE STANDARD CONTROL PARADIGM IS UNSATISFACTORY; AND (3) HIS CONTENTION THAT THE MOST ADEQUATE ACCOUNT OF AVAILABLE DATA IS PROVIDED BY MEDIATION THEORY IS ACHIEVED ONLY AT THE EXPENSE OF PASSING OVER A VERY SUBSTANTIAL BODY OF DATA INCOMPATIBLE WITH A MEDIATION INTERPRETATION OF PAIRED-ASSOCIATE PARADIGMS. (FRENCH SUMMARY) (30 REF.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

MEDIATION OR PSEUDOMEDIATION: A REPLY TO EARHARD AND MANDLER.

B. EARHARD AND G. MANDLER (SEE 40:3) HAVE ARGUED THAT OBSERVED DIFFERENCES IN MEDIATION EXPERIMENTS, WHEN MEDIATION IS DEMONSTRATED THROUGH LABORATORY ESTABLISHED ASSOCIATIONS, CANNOT BE ATTRIBUTED UNEQUIVOCALLY TO MEDIATION MECHANISMS. THEY INTERPRET THESE FINDINGS AS BASED ON INTERLIST INTERFERENCE AND UNLEARNING EFFECTS. AN ATTEMPT IS MADE TO EVALUATE THIS POSITION BY EXAMINING LITERATURE DEALING WITH INTERFERENCE AND UNLEARNING, MEDIATION, AND PSEUDOMEDIATION. IT IS CONCLUDED THAT THE FINDINGS CANNOT BE ADEQUATELY EXPLAINED IN THE MANNER PROPOSED BY EARHARD AND MANDLER AND THAT MEDIATION THEORY OFFERS THE MOST REASONABLE ACCOUNT OF THE EXPERIMENTAL FACTS. (FRENCH SUMMARY) (42 REF.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

CONCEPT-SHIFT AND DISCRIMINATION-REVERSAL LEARNING IN HUMANS.

REVIEWS CURRENT STUDIES UNDER 6 HEADINGS: (1) PARTIAL REINFORCEMENT DURING EXTRADIMENSIONAL SHIFT, (2) NUMBER OF RESPONSE CHOICES, (3) AGE * TYPE OF SHIFT, (4) INTELLIGENCE, (5) DEGREE OF ORIGINAL TRAINING, AND (6) VERBAL, PERCEPTUAL, AND ATTENTIONAL FACTORS. IN ADDITION, SEVERAL THEORETICAL ACCOUNTS OF THE SHIFT PROCESS ARE CONSIDERED AND AN EFFORT IS MADE TO EVALUATE EACH OF THEM IN TERMS OF THE EVIDENCE PRESENTED. (5 P. REF.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

MEDIATION EFFECTS AS A FUNCTION OF MEDIATOR RECALL AND MEANINGFULNESS.

STUDIED MEDIATION EFFECTS AS A FUNCTION OF MEDIATOR RECALL AND MEANINGFULNESS. USING THE A-B, A-C, B-C (A-B, D-C, B-C CONTROL) PARADIGM, B AND C ITEMS WERE PRESENTED AT A 0/6, 2/6, 4/6, OR 6/6 CRITERION OF STAGE 3 ACQUISITION, AND S RECALLED THE ITEMS ASSOCIATED WITH B AND C IN STAGES 1 AND 2. THE A AND D ITEMS WERE PRESENTED AFTER STAGE 2 IN ORDER TO ASSESS DIFFERENCES IN FORWARD AND BACKWARD RECALL AT THE 0/6 CRITERION. HIGH- AND LOW-M VERBAL MATERIAL CONDITIONS WERE ORTHOGONAL TO THE RECALL CONDITIONS. THE MAJOR RESULTS WERE: (1) HIGH-M MATERIALS APPARENTLY YIELD SUPERIOR BACKWARD LEARNING AND LESS BACKWARD UNLEARNING IN STAGES 1 AND 2, THUS MAKING THE MEDIATOR MORE AVAILABLE FOR HIGH-M MATERIAL AT THE ONSET OF STAGE 3; (2) STAGE 3 ACQUISITION PRODUCES A GENERAL DECREASE IN MEDIATOR RECALL; AND (3) THERE WAS A GENERAL LACK OF RELATIONSHIP BETWEEN MEDIATOR RECALL IN STAGE 3 AND B-C PERFORMANCE. THE FINDINGS WERE INTERPRETED IN TERMS OF THE RELATION OF ASSOCIATIVE STRENGTH OF B-A AND C-A AFTER STAGE 1 AND 2 ACQUISITION AND ITS RELATION TO MEDIATOR USAGE. (FRENCH SUMMARY) (19 REF.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

EXPERIMENTAL STRESS AND LEARNING IN THE AGED.

SERIAL ROTE VERBAL LEARNING UNDER FAST (4 SEC.) AND SLOW (10 SEC.) PACING WAS ADMINISTERED TO 6 GROUPS OF AGED MEN OVER 60 (MEAN AGE = 67.8, N = 54) UNDER 3 LEVELS OF EXPERIMENTAL STRESS. MAXIMAL EFFECTS OF INDUCED STRESS INCLUDING RESPONSE INHIBITION WERE PREDICTED FOR SLOWER PACING. RESULTS CONFIRMED THE HYPOTHESIS AND SUGGEST THAT THE INCREASED LEARNING AND RESPONSIVITY OF AGED SS WITH LONGER EXPOSURE DURATION MAY BE A FUNCTION OF ANXIETY REDUCTION. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

POSITIONAL AND SEQUENTIAL PROBES IN SERIAL LEARNING.

24 COLLEGE MALES PARTICIPATED IN 2 EXPERIMENTS WHICH USED A SHORT-TERM MEMORY TASK AND 3 TYPES OF PROBE TESTS. IN EXP. I, DESIGNED TO DISCOVER IF POSITIONAL CUES ARE MORE EFFECTIVE THAN SEQUENTIAL CUES IN SERIAL LIST LEARNING, EACH LIST WAS SEEN ONCE AND FOLLOWED BY A SINGLE PROBE TEST FOR A SINGLE ITEM. SERIAL POSITION CURVES SHOWED THAT THE POSITIONAL PROBE LED TO SOMEWHAT HIGHER RECALL OF ITEMS OCCUPYING THE END POSITIONS OF A LIST THAN THE SEQUENTIAL PROBE. EXP. II WAS UNDERTAKEN TO MAXIMIZE THE EFFECTIVENESS OF THE POSITION PROBE. 5 TRIALS WERE GIVEN BEFORE A RECALL TEST. SERIAL POSITION CURVES FOR THE 3 TYPES OF PROBES SHOWED THAT THE DIFFERENCES BETWEEN THEM WERE NOT SIGNIFICANT. THE OVER-ALL RESULTS WERE INTERPRETED AS INDICATING THAT POSITIONAL CUES AND SEQUENTIAL CUES MAY BE EQUALLY EFFECTIVE IN SERIAL LEARNING. (FRENCH SUMMARY) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

ESCAPE AND APPROACH RESPONSES IN AVOIDANCE LEARNING.

IN EXP. I, 18 MALE ALBINO SPRAGUE-DAWLEY 85-100 DAY OLD RATS WERE ORIGINALLY TRAINED TO RUN FROM A DISTINCTIVE SHOCK CHAMBER (SIDE A) TO A DISTINCTIVE SAFE CHAMBER (SIDE B) AND CONFINED ON SIDE B FOR EITHER 30 OR 150 SEC. AFTER REACHING CRITERION, ALL GROUPS WERE GIVEN REVERSAL TRAINING (SIDE B TO SIDE A) WITH 30-SEC CONFINEMENT. THE GROUP WITH 150-SEC CONFINEMENT IN ORIGINAL LEARNING LEARNED FASTER BUT REVERSED SLOWER THAN THE 2 30-SEC CONTROL GROUPS. IN EXP. II, 52 MALE SS WERE SHOCKED OR NOT SHOCKED IN AN EXTERNAL SHOCK BOX, AND THEN TRANSPORTED BY HAND TO 1 SIDE OF A 1-WAY BOX FOR EITHER 30 OR 150 SEC. OF CONFINEMENT. AFTER 8 TRIALS, ALL GROUPS WERE GIVEN AVOIDANCE TRAINING EITHER TOWARD OR AWAY FROM THEIR PRETRAINING CONFINEMENT CHAMBER. THE GROUP THAT WAS BOTH SHOCKED IN THE EXTERNAL SHOCK BOX AND CONFINED FOR 150 SEC. LEARNED FASTER WHEN RUN TOWARD AND SLOWER WHEN RUN AWAY FROM THE PRETRAINING CONFINEMENT SIDE THAN ALL OTHER GROUPS (CONTROLS). (FRENCH SUMMARY) (17 REF.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

CONJUNCTIVE AND DISJUNCTIVE CONCEPT LEARNING IN HUMANS AND SQUIRREL MONKEYS.

40 HUMANS AND 16 SQUIRREL MONKEYS LEARNED A CONJUNCTIVE AND A DISJUNCTIVE CONCEPT IN A CHOICE PROCEDURE IN WHICH A POSITIVE AND A NEGATIVE INSTANCE WERE PRESENTED ON EACH TRIAL. THERE WERE 2 RELEVANT AND 2 IRRELEVANT DIMENSIONS. HUMANS FOUND DISJUNCTIVE CONCEPTS MORE DIFFICULT; THIS TREND WAS SLIGHTLY REVERSED FOR THE MONKEYS. MONKEYS WERE MORE INFLUENCED BY THE NATURE OF THE RELEVANT DIMENSIONS THAN WERE HUMANS. VINCENT LEARNING CURVES SHOWED IMPROVEMENT PRIOR TO CRITERION IN ALL CASES, IN AGREEMENT WITH THE TRABASSO-BOWER MODEL FOR 2-CUE CONCEPT LEARNING. RESULTS ARE DISCUSSED IN TERMS OF HYPOTHESIS TESTING AND DISCRIMINATION LEARNING MODELS OF CONCEPTUAL BEHAVIOR. (FRENCH SUMMARY) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

VERBAL LEARNING, MODELING, AND JUVENILE DELINQUENCY.

DEALS WITH EXPOSURE TO OBSERVATIONAL LEARNING OPPORTUNITIES AS A POSSIBLE THERAPEUTIC TECHNIQUE WITH JUVENILE DELINQUENTS. INVESTIGATION OF THE RELATIONSHIP BETWEEN MODELING AND THE BEHAVIOR OF JUVENILE DELINQUENTS GREW OUT OF EXPERIMENTAL STUDIES OF PERSONALITY. AVAILABLE EVIDENCE SUGGESTS THAT OBSERVATIONAL OPPORTUNITIES ARE POTENTIALLY POWERFUL BEHAVIOR MODIFICATION INFLUENCES. THESE INFLUENCES ARE MOST NOTICEABLE IN THE AREAS OF PERSONAL ATTITUDES AND SOCIAL BEHAVIOR. THE OUTLINE OF THE 1-YR STUDY DESIGNED TO SHOW THE EFFICACY OF MODELING AS A BEHAVIOR MODIFICATION TECHNIQUE WITH DELINQUENTS IS DESCRIBED. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

LEARNING OF TELEPHONE NUMBERS WITH AND WITHOUT LETTER-EXCHANGE PREFIXES.

36 COLLEGE STUDENTS, RANDOMLY ASSIGNED TO 4 GROUPS, PRACTICED AND WERE TESTED FOR RECALL ON 8- OR 10-UNIT LISTS OF TELEPHONE NUMBERS WITH LETTER EXCHANGES OR WITH DIGITS SUBSTITUTED FOR THE LETTERS. IN TERMS OF MEAN SCORES AND FINAL ACHIEVEMENT LEVELS, GROUPS PRACTICING LISTS OF TELEPHONE NUMBERS WITH LETTERED EXCHANGES PERFORMED BETTER THAN THOSE WHO PRACTICED CORRESPONDING LISTS COMPOSED ENTIRELY OF DIGITS. ITEMS IN THE TRADITIONAL TELEPHONE IDENTIFICATION SYSTEM USING PAIRS OF PREFIXED LETTER EXCHANGES WERE MOST READILY LEARNED. EXPLANATIONS IN TERMS OF SYMBOL ISOLATION, INTERFERENCE, AND FAMILIARITY EFFECTS ARE INDICATED, AND IMPLICATIONS FOR TELEPHONE IDENTIFICATION ARE SPECIFIED. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

DISCRIMINATED LEARNING AND REVERSAL IN THE WHEEL-TURN AVOIDANCE SITUATION.

2 STIMULI, ONLY 1 OF WHICH WAS PAIRED WITH SHOCK, WERE USED IN 2 EXPERIMENTS INVESTIGATING STIMULUS DISCRIMINATION AND REVERSAL EFFECTS IN AVOIDANCE LEARNING. THE FINDINGS INCLUDED: (1) NO DIFFERENCES IN LEARNING ATTRIBUTABLE TO THE STIMULUS (BUZZER OR TONE) THAT WAS PAIRED WITH SHOCK, (2) DISCRIMINATED RESPONDING SUCH THAT THERE WAS CONTINUED RESPONDING TO THE SHOCK-PAIRED STIMULUS AND REDUCED RESPONDING TO THE UNPAIRED STIMULUS, AND (3) DISCRIMINATED RESPONDING AFTER PAIRED AND UNPAIRED STIMULI REVERSALS BUT WITH GRADUALLY DETERIORATING DISCRIMINATION WITH SUCCESSIVE REVERSALS. (FRENCH SUMMARY) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

LEARNING POTENTIAL AND RIGIDITY IN THE ADOLESCENT MENTALLY RETARDED.

PERFORMANCE OF 40 EDUCABLE MENTALLY RETARDED (EMR) ADOLESCENTS AFTER COACHING ON A NONVERBAL REASONING TASK VARIES CONSIDERABLY. SS WHOSE SCORES IMPROVED MARKEDLY ON THE REASONING TASK (GAINERS) CANNOT BE DISTINGUISHED FROM THOSE WHO DID NOT BENEFIT FROM THE COACHING EXPERIENCE (NONGAINERS), WHEN TRADITIONAL IQ AND SCHOOL-HISTORY CRITERIA ARE EXAMINED. GAINER EMRS DEMONSTRATED LESS RIGIDITY THAN NONGAINERS ON A SIMPLE REPETITIVE TASK, TENDING TO SATIATE MORE FREQUENTLY. GAINERS ALSO TENDED TO SHIFT CONCEPTS MORE READILY THAN NONGAINERS AND MA CONTROLS. CA CONTROLS WERE LEAST RIGID ON BOTH TASKS. GAINERS' PERFORMANCE PARALLELED THIS GROUP. A HISTORY OF SOCIAL DEPRIVATION WAS ASSOCIATED WITH PROLONGED PERFORMANCE ON THE SIMPLE REPETITIVE TASKS BUT NOT ON THE CONCEPT-SHIFT PROBLEM. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

PAIRED-ASSOCIATE LEARNING AS RELATED TO "PRONUNCIABILITY" AND WORD FREQUENCY.

EXAMINED THE UNDERWOOD AND SCHULZ HYPOTHESIS THAT PRONUNCIABILITY RATHER THAN FREQUENCY IS THE MAJOR VARIABLE IN LEARNING BY USING WORDS WHICH DIFFERED IN OBJECTIVE FREQUENCY BUT WHICH WERE PRONOUNCED IN THE SAME WAY. 4 PAIRED-ASSOCIATE LISTS WERE CONSTRUCTED WHICH USED THE VARYING COMBINATIONS OF HIGH- AND LOW-FREQUENCY STIMULUS AND RESPONSE WORDS (HH, HL, LH, LL). RESULTS, CONTRARY TO UNDERWOOD AND SCHULZ, REVEALED THAT LEARNING WAS MOST RAPID FOR THE HH COMBINATION, FOLLOWED BY LH, HL, AND LL. (FRENCH SUMMARY) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

OUTER-DIRECTEDNESS IN THE DISCRIMINATION LEARNING OF NORMAL AND MENTALLY RETARDED CHILDREN.

HIGH- AND LOW-DISTRACTIBLE RETARDATES OF MIXED ETIOLOGY AND NORMALS OF A COMPARABLE MA LEARNED A 3-CHOICE SIZE DISCRIMINATION WITH AND WITHOUT AN ADDITIONAL CUE. FOR 1/2 OF THE SS IN THE CUE CONDITIONS THE CUE ALWAYS INDICATED THE CORRECT STIMULUS, AND FOR THE OTHER 1/2 THE CUE INDICATED AN INCORRECT STIMULUS. FOR 1/2 OF THE SS IN EACH OF THE CUE CONDITIONS THE CUE WAS E'S FINGER, FOR THE OTHER 1/2 THE CUE WAS A LIGHT. THE HYPOTHESIS THAT RETARDATES ARE MORE OUTER-DIRECTED IN PROBLEM SOLVING THAN NORMALS GENERATED THE PREDICTION THAT THE LEARNING OF THE RETARDATES WOULD BE MORE ENHANCED IN THE POSITIVE CONDITION AND MORE DEBILITATED IN THE NEGATIVE CONDITIONS. CONFIRMATION OF THIS HYPOTHESIS WAS FOUND IN THE NEGATIVE CONDITIONS. NO DIFFERENCES ASSOCIATED WITH TYPE OF CUE OR LEVEL OF DISTRACTIBILITY WERE FOUND. FAMILIAL RETARDATES WERE FOUND TO BE LESS OUTER-DIRECTED IN THEIR PROBLEM SOLVING THAN WERE NONFAMILIAL RETARDATES. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

ANXIETY (DRIVE) AND VERBAL LEARNING: IMPLICATIONS FOR RESEARCH AND SOME METHODOLOGICAL CONSIDERATIONS.

EXAMINED THE METHODOLOGICAL ADEQUACY OF THE WORK CONSIDERED AS EXPERIMENTAL TESTS OF THE TAYLOR-SPENCE DRIVE THEORY. PERTINENT RESEARCH INVOLVING VERBAL LEARNING PARADIGMS IS REVIEWED IN THE CONTEXT OF RECENT METHODOLOGICAL AND THEORETICAL DEVELOPMENTS IN VERBAL LEARNING. SOURCES OF POTENTIAL CONFOUNDING REFINEMENTS IN METHODOLOGY PERMITTING MORE ADEQUATE TESTS OF THE TAYLOR-SPENCE THEORY ARE PRESENTED. (2 P. REF.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

EXPECTANCY OF SUCCESS AND THE PROBABILITY LEARNING OF MIDDLE-CLASS, LOWER-CLASS, AND RETARDED CHILDREN.

TESTED THE HYPOTHESIS THAT RETARDED AND NORMAL CHILDREN OF THE SAME MA DIFFER IN PERFORMANCE ON A PARTIALLY REINFORCED 3-CHOICE LEARNING TASK BECAUSE THEY HAVE DIFFERING EXPECTANCIES OF SUCCESS. THE PERFORMANCE OF 60 NORMAL LOWER-CLASS (LOW EXPECTANCY) AND 60 NORMAL MIDDLE-CLASS (HIGH EXPECTANCY) SS WAS COMPARED WITH THAT OF 60 NONINSTITUTIONALIZED RETARDATES. 1/3 OF THE SS IN EACH GROUP WERE ASSIGNED TO A SUCCESS, FAILURE, OR CONTROL CONDITION WHICH THEY EXPERIENCED PRIOR TO PERFORMING ON THE CRITERION TASK. 1/2 OF THE SS IN EACH GROUP WERE PENALIZED FOR MAKING ERRORS. NORMAL LOWER-CLASS SS SHOWED THE MOST MAXIMIZING (CORRECT RESPONSES) AND THE LEAST LEFT, MIDDLE, RIGHT PATTERNING OF RESPONSES, WHILE NORMAL MIDDLE-CLASS SS SHOWED THE LEAST MAXIMIZING AND MOST PATTERNING. RETARDED SS FELL BETWEEN THESE 2 GROUPS ON BOTH MEASURES. (29 REF.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

论自然资源与钢铁工业

论述钢铁工业所必需的自然资源如煤，石油，天然气及钢铁生产的原材料－铁矿石在世界的分布，供求情况及对钢铁工业发展的影响，介绍了一些国家利用能源和原材料和新工艺和新方法，以利用我国开发，利用和节约资源，降低钢铁产品的成本，提高钢铁产品在国际市场上的竞争力，促进钢铁工业乃至国民经济的发展。     

SOME RECENT CRITICISMS OF BEHAVIORISM AND LEARNING THEORY: WITH SPECIAL REFERENCE TO BREGER AND MCGAUGH AND TO CHOMSKY.

SOME RECENT CRITICISMS OF BEHAVIORISM, REINFORCEMENT THEORY, AND LEARNING THEORY WERE ELABORATED BY L. BREGER AND J. L. MCGAUGH (SEE 39:4). EXAMINATION OF CRITICISMS BY BREGER AND MCGAUGH, CHOMSKY, AND OTHERS, REVEALS THAT MANY OF THEM CONTAIN ERRORS OF FACT AND INTERPRETATION, OR STEM FROM A CONFUSION OF OBSERVATION AND INFERENCE. PROVIDING PSEUDOEXPLANATIONS BY MERE NAMING IS ANOTHER PHENOMENON FREQUENTLY FOUND IN MENTALISTIC REFORMULATIONS OF BEHAVIOR THEORY. FINALLY, THE CRITICISMS ARE SHOWN TO OFFER MERELY ALTERNATIVE AND LESS PRECISE WAYS OF SPEAKING ABOUT BEHAVIORAL PHENOMENA RATHER THAN A DEMONSTRABLY BETTER SET OF RESEARCH TOOLS AND THERAPEUTIC TECHNIQUES. (2 P. REF.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

论自然资源与钢铁生产发展的关系

论述了煤，石油，天然气和铁矿石等钢铁工业所必需的自然资源在世界的分布，供求情况及对钢铁工业发展的影响，介绍了世界一些主要国家利用能源和原材料的新工艺和新方法，以利于我国开发，利用和节约资源，降低钢铁产品的成本，提高钢铁产品在国际市场的竞争力，促进钢铁工业的发展。     

THE INFLUENCE OF CONTEXTUAL CUES UPON THE LEARNING AND RETENTION OF PAIRED ASSOCIATES.

SS LEARNED A LIST OF NONSENSE SYLLABLE PAIRED ASSOCIATES UNDER 4 CONDITIONS: COLOR CONTEXT, SHAPE CONTEXT, COLOR PLUS SHAPE CONTEXT, AND NO CONTEXT. A RETENTION TEST WAS ADMINISTERED 24 HR. LATER, WITH SS RELEARNING THE LIST UNDER THE SAME OR CHANGED CONDITIONS. RESULTS INDICATE THAT ITEM-SPECIFIC CONTEXTUAL CUES SURROUNDING THE STIMULUS ITEMS ENHANCED LEARNING. WHEN VARIOUS CONTEXT CONDITIONS WERE PRESENTED IN THE RETENTION TEST, RECALL OF THE RESPONSE ITEMS WAS GREATEST WHEN THE RETENTION CONTEXT CONTAINED THE SAME ITEM-SPECIFIC CUES AS THE ORIGINAL CONTEXT. REMOVING CUES PRESENT DURING LEARNING CAUSED A DECREMENT IN RECALL. ADDING OR CHANGING CUES HAD NO EFFECT UPON RECALL, AS LONG AS THE CONTEXT PRESENT DURING LEARNING WAS ALSO PRESENT DURING THE RETENTION TEST. THE EASE WITH WHICH THE PAIRED ASSOCIATES WERE RELEARNED WAS INFLUENCED BY THE RELEARNING CONTEXT, BUT WAS RELATIVELY INDEPENDENT OF THE CONTEXT CONDITIONS UNDER WHICH ORIGINAL LEARNING TOOK PLACE. THESE RESULT S WERE INTERPRETED AS SUPPORTING A MODIFIED STIMULUS SELECTION HYPOTHESIS. (FRENCH ABSTRACT) (PsycINFO Database Record (c) 2010 APA, all rights reserved)