Academic journal article Memory & Cognition

Strategic Behavior without Awareness? Effects of Implicit Learning in the Eriksen Flanker Paradigm

Academic journal article Memory & Cognition

Strategic Behavior without Awareness? Effects of Implicit Learning in the Eriksen Flanker Paradigm

Article excerpt

This experiment investigated whether subjects' selection and use of strategies in detecting a target letter in a flanker task requires intention. Subjects' expectancies for compatible and incompatible trials (trials on which the response to the flanker stimulus was consistent or inconsistent with the target response) were manipulated by presenting cues that signaled the occurrence of these types of trials. Three groups of subjects received explicit, partially explicit, or implicit instructions about the meaning of the cues. By the end of the experiment, all the groups were able to select and use strategies based on the cues to improve their performance. However, this strategy selection developed slowly with practice in the latter two groups, whereas it was present from the outset in the first group. In addition, forced choice tests performed after the experiment showed that the subjects in the implicit condition could not intentionally indicate which stimuli were most likely to follow a given cue. Thus, the data suggest that the selection of strategies occurred outside the subjects' awareness, and without their intention.

All cognitive tasks probably involve at least some minimal strategy selection. When people are required repeatedly to choose among different possibilities and to analyze the costs and benefits of a given situation, they learn rules and develop strategies to perform the task. Although in many cases the selection of a particular strategy is a deliberate process, in other instances it may also involve mechanisms that are not under conscious awareness. The present study investigates conditions in which both types of strategy selection may occur.

Numerous studies have demonstrated that even under time pressure, subjects are able to extract relevant information from the environment and use strategies to direct their task performance (e.g., Coles, Gratton, Bashore, Eriksen, & Donchin, 1985; Gratton, Coles, Sirevaag, Eriksen, & Donchin, 1988; Logan & Zbrodoff, 1979, 1982; Szymanski & MacLeod, 1996). Gratton, Coles, and Donchin (1992), using the Eriksen flanker task (Eriksen & Eriksen, 1974), showed that strategy selection is adaptive and depends on expectations about stimulus conditions. Specifically, when subjects were led to expect that compatible-noise conditions (i.e., conditions in which target and noise letters were associated with the same response) would occur, they used a strategy leading to fast responses, but with high susceptibility to distractor conditions. When subjects were led to expect that incompatiblenoise conditions (i.e., conditions in which target and noise letters were associated with conflicting responses) would occur, they used a strategy leading to slow responses, but also to low susceptibility to distractor conditions.

Gratton et al. (1992) did not investigate the extent to which these forms of strategy adjustments require that the subjects have explicit awareness of the difference in noise expectancy conditions (i.e., compatible vs. incompatible trials). In fact, evidence from event-related brain potential (ERP) studies (e.g., mismatch negativity; see Näätänen, 1982) and investigations of implicit-learning processes (e.g., Carlson & Flowers, 1996; Miller, 1987; Stadler, Warren, & Lesch, 2000) suggest that subjects are able to compute probabilities of events even in the apparent absence of awareness. If expectancies can be computed at a subconscious level, can strategic choices made on the basis of expectancy manipulations also occur without explicit awareness? The present study examined this possibility.

As in Gratton et al. (1992), the subjects in the present study were presented with a visual display (the imperative stimulus) composed of five letters (HHHHH, SSHSS, SSSSS, and HHSHH). The middle letter was designated as the target, and the other letters (labeled noise letters) were used as distractors. The subjects were instructed to respond to one of the possible target letters (H or S) with one hand and to the other letter with the other hand. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.