The effect of subject-generated and experimenter-supplied code words on the phonetic mnemonic system

• Published in: Applied cognitive psychology Vol. 5; no. 2; pp. 135 - 148
• Main Authors: PATTON, G. W. R, D'AGARO, W. R, GAUDETTE, M. D
• Format: Journal Article
• Language: English
• Published: Chichester Wiley 01.03.1991
• Subjects: Biological and medical sciences; Code, Coding; Fundamental and applied biological sciences. Psychology; Learning. Memory; Memory; Psychology. Psychoanalysis. Psychiatry; Psychology. Psychophysiology; Undergraduate School; Verbal Task; Human; Acquisition process; Memory; Phonetics; Strategy; Cognition; Recall; Mnemonic learning; Code
• ISSN: 0888-4080; 1099-0720
• DOI: 10.1002/acp.2350050207

Three experiments tested the hypothesis that recall by Ss who use a phonetic mnemonic system supplied by the experimenter is superior to that of subject-generated codes, even when adequate time to produce complete lists of code words for numbers is available. In experiment 1, undergraduates (N = 36) were randomly assigned to two training (experimenter-supplied or subject-created) groups & a control group. Ss learned to use a system based on digit-symbol correspondence for the digits 0 to 9 or constructed their own code words to represent 20 numbers. Ss were tested for recall immediately after training, one, two, or three days later, & seven days after the second test. A 3 X 3 factorial analysis of variance revealed that recall scores increased from the first to second test, & declined after a 6-day delay, thus showing a significant difference between training groups. Higher mean recall scores were observed for experimenter-supplied than for the subject-generated or control groups. The mean recall for the control groups was higher than for the self-generated group. Further analyses show that self-generated lists were incomplete or incorrect as code words. In experiment 2, undergraduates (N = 12) generated code words with the guidance of the experimenter. Data were analyzed with the data from experiment 1. Results indicate no difference between the two self-generated groups, & the same pattern of rankings in recall as the previous analysis. In experiment 3, the self-generated code group was given training until they were fluent in the use of the mnemonic system. Ss (N = 28) were assigned to training groups & tested immediately & seven days after the initial test. Analysis shows better recall in experimenter-supplied groups in both immediate & delayed conditions. These findings support the hypothesis that experimenter-generated codes lead to superior recall. 2 Tables, 22 References. M. Lemons