Stroop proactive control and task conflict are modulated by concurrent working memory load

Abstract

Performance on the Stroop task reflects two types of conflict-informational (between the incongruent word and font color) and task (between the contextually relevant color-naming task and the irrelevant, but automatic, word-reading task). According to the dual mechanisms of control theory (DMC; Braver, 2012), variability in Stroop performance can result from variability in the deployment of a proactive task-demand control mechanism. Previous research has shown that when proactive control (PC) is diminished, both increased Stroop interference and a reversed Stroop facilitation (RF) are observed. Although the current DMC model accounts for the former effect, it does not predict the observed RF, which is considered to be behavioral evidence for task conflict in the Stroop task. Here we expanded the DMC model to account for Stroop RF. Assuming that a concurrent working memory (WM) task reduces PC, we predicted both increased interference and an RF. Nineteen participants performed a standard Stroop task combined with a concurrent n-back task, which was aimed at reducing available WM resources, and thus overloading PC. Although the results indicated common Stroop interference and facilitation in the low-load condition (zero-back), in the high-load condition (two-back), both increased Stroop interference and RF were observed, consistent with the model's prediction. These findings indicate that PC is modulated by concurrent WM load and serves as a common control mechanism for both informational and task Stroop conflicts.

Adjustment of control in the numerical Stroop task

“…Most studies examined top-down mechanisms by examining the ability to inhibit reading of a single colored word, or examining several stimuli in which there was only one conflict that stemmed from the combination of all stimuli (i.e., Appelbaum et al, 2014; Bugg et al, 2011; Cohen et al, 1990; Entel et al, 2015; Gratton et al, 1992; Hutchison, 2011; Kalanthroff et al, 2015; Lindsay & Jacoby, 1994; Logan & Zbrodoff, 1979; Steinhauser & Hubner, 2009; Tzelgov, Henik et al, 1992). In our study, we used the numerical Stroop task in which participants were asked to compare the magnitude of numbers or their physical sizes.…”

“…Control adjustment has been studied mostly by employing a color-word Stroop task (e.g., Bugg et al, 2011; Entel, Tzelgov, Bereby-Meyer, & Shahar, 2015; Hutchison, 2011; Kalanthroff, Avnit, Henik, Davelaar, & Usher, 2015). This version of the Stroop task is limited due to its simplicity and the asymmetry between the different dimensions of the stimulus.…”

Priming can affect naming colours using the study-test procedure. Revealing the role of task conflict

“…Figure 1: Proactive-control/task-conflict (PC-TC) model. Adapted from Kalanthroff et al (2015). Task conflict is represented as the inhibitory connection between the task demand layer and the response layer.…”

“…In the Proactive-control/task-conflict (PC-TC) model (Kalanthroff, Avnit, Henik, Davelaar & Usher, 2015) additional features are added to implement the effects of task conflict (see figure 1). Task conflict is implemented as an inhibitory connection between the task demand layer and the response layer.…”

“…To illustrate the workings of the model with respect to task conflict Kalanthroff et al (2015) describe how when proactive control (PC) to colour naming is high the task demand colour naming unit is given additional activation top-down and thus the influence of bottomup connections into the task demand units is negligible. However, when PC to colour naming is low this allows the bottom-up connections from the word input units to the word reading units to increase competition in the task demand layer and thus inhibit units in the response layer.…”