No attentional control in working memory: A systematic re-analysis of latent-variable studies

Attentional control - also called executive functions or cognitive control - refers to our ability to maintain and implement a goal and goal-relevant information in the face of distraction. There is growing evidence that attentional control cannot be established as a psychometric construct with the measures used so far. This asks for other ways of measuring attentional control. Early research has used structural equation modeling to extract attentional control from working-memory tasks (i.e., tasks used to measure the temporary maintenance and manipulation of information) and short-term memory tasks (i.e., tasks used to measure the temporary maintenance of information). Attentional control was first modeled as the residual variance of working memory after controlling for short-term memory. Second, it was modeled as the common factor across working-memory and short-term memory measures, with higher loadings for working-memory measures. After a systematic search, we found three datasets in which both models could be systematically tested. Moreover, we determined the robustness of the results by bootstrapping 5000 correlation matrices from the original correlation matrix and by estimating both models using these simulated correlation matrices. Across the datasets, the models could be successfully estimated, but they were not fully and exactly replicated. Moreover, all these model estimations were not robustly observed when the 5000 simulated correlation matrices were used. Therefore, attentional control could not be reliably extracted from working-memory tasks at the latent-variable level. These findings further strengthen the case against a psychometric construct of attentional control, increasing the challenges in the conceptualization of this construct.