Abstract
Two experiments, with 54 and 60 college Ss, respectively, had paired-associate words varied either in or out of a sentence context in training; had stimulus, response or both S-R similarity varied in 3 lists of paired-associate words or sentences between training and transfer; and had the meaning relationships of words within the 3 lists of paired-associate words or sentences between training and transfer varied in six ways: unrelated, synonyms, antonyms, homonyms, associates, and identical. Both experiments were designed as 2 (context) by 3 (lists) by 6 (meaning) with repeated measures for meaning. In Exp. I, the paired-associate words for the transfer task were not in a sentence context and in Exp. II the same paired-associate words were in a sentence context. The results indicated that there was reliably greater transfer for words that had homonym, associate or identical than unrelated, synonym, or antonym meaning relationships between training and transfer. This was true whether the paired-associate words were in or out of context in training or in transfer. Only when the training and transfer tasks were in context was there more transfer of learning due to context than no context. Contrary to the theory of transfer, there was no difference due to S, R, or S-R similarity being varied between training and transfer.
Get full access to this article
View all access options for this article.
