The Influence of Working Memory Load on Expectancy-Based Strategic Processes in the Stroop-Priming Task

Abstract

The present study investigated whether a differential availability of cognitive control resources as a result of varying working memory (WM) load could affect the capacity for expectancy-based strategic actions. Participants performed a Stroop-priming task in which a prime word (GREEN or RED) was followed by a colored target (red vs. green) that participants had to identify. The prime was incongruent or congruent with the target color on 80 and 20% of the trials, respectively, and participants were informed about the differential proportion of congruent vs. incongruent trials. This task was interleaved with a WM task, such that the prime word was preceded by a sequence of either a same digit repeated five times (low load) or five different random digits (high load), which should be retained by participants. After two, three, or four Stroop trials, they had to decide whether or not a probe digit was a part of the memory set. The key finding was a significant interaction between prime-target congruency and WM load: Whereas a strategy-dependent (reversed Stroop) effect was found under low WM load, a standard Stroop interference effect was observed under high WM load. These findings demonstrate that the availability of WM is crucial for implementing expectancy-based strategic actions.

Expectancy-Based Strategic Processes Are Influenced by Spatial Working Memory Load and Individual Differences in Working Memory Capacity

“…The results by Ortells et al (2017) replicate and extent those obtained by other recent studies, in showing that limiting the availability of cognitive (WM) resources with a WM task demanding a high load, can induce a less efficient strategic processing of goal-relevant information (e.g., Heyman et al, 2014 ; Hutchison et al, 2014 ).…”

“… Ortells et al (2017) found that the implementation of expectancy-based attention strategies in that version of the Stroop task critically depended on the availability of WM resources, as there was a reliable congruency by WM load interaction. Thus, when the WM task demanded a low load, participants were able to strategically process the prime to anticipate the target color, as their responses were reliably faster on incongruent than on congruent trials.…”

“…Although much less investigated, some recent studies have reported evidence that an efficient implementation of controlled facilitatory strategies like expectancy generation also relies on the availability of cognitive control resources, such as WM (e.g., Heyman et al, 2014 ; Hutchison et al, 2014 ; Ortells et al, 2017 ).…”

“…In a recent study, Ortells et al (2017) used the combined WM/selective attention paradigm originally developed by De Fockert et al (2001) in a Stroop-priming task which allows measuring of qualitatively different behavioral effects resulting from strategic vs. non-strategic processing. In this task, participants are required to identify the color (e.g., red) of a target patch which is preceded by either an incongruent (e.g., GREEN) or a congruent (RED) prime word, on 80% and 20% of the trials, respectively.…”

“…As participants foreknowledged that the incongruent prime-target pairs were much more frequent than the congruent ones, and there are only two possible colors, a useful strategy would be to prepare to respond to the opposite target color to that of the prime. By implementing that strategy, participants perform much better on incongruent than on congruent trials, thus showing a reversed Stroop effect (e.g., Merikle and Joordens, 1997 ; Ortells et al, 2017 ; see also Logan et al, 1984 ). This Stroop task was combined with a verbal WM task of either high or low load.…”