Unconscious influences on direct-test performance

Abstract

From intense interest in implicit memory there have evolved various methods for separating the respective influence of implicit (unconscious) and explicit (conscious) processes on performance of various tasks. Two experiments are reported, utilizing a levels-of-processing (LOP) approach to manipulate encoding level and comparable indirect (word-stem completion) and direct (cued-completion) retention tests. Confidence ratings of recollection were taken for each direct test response. The aim of these experiments was to explore the role that guessing plays in direct-test performance (Experiments 1 and 2) and to contrast this with performance in a comparable indirect test (Experiment 2). Analysis of correctly guessed responses showed that direct-test performance was reliably influenced by unconscious processes, but differently as a function of LOP. Guessing stems of nonsemantically processed words was found to enhance performance, whereas guessing stems of semantically processed words had no effect on performance. Results are discussed in terms of the similarity between guessing in a direct test and engaging in an indirect test, and subjects' unwitting resourcefulness at being able to retrieve words they cannot explicitly remember.