The many faces of learning-guided cognitive control.

Abstract

Cognitive control refers to processes that enable adaptive, goaldirected behavior. Once ascribed to smart agents that willfully biased behavior in a top-down fashion (Norman & Shallice, 1986), an emerging "learning perspective" embodies the view that associative learning and memory processes are central to control (for recent reviews, see Abrahamse et al., 2016;Braem & Egner, 2018;Chiu & Egner, 2019;Egner, 2014). The guiding question of this special issue is how people learn to adapt control in a context-sensitive manner ("control learning"). Broadly speaking, the hypothesis probed by the articles herein is that this occurs via learning about regularities in the (task) environment, which in turn guides the engagement of control. This can take place in the form of incremental learning of the contextual likelihood of control demands (e.g., the accumulating realization that the current block of trials seems to be of high difficulty), and/or by associating specific stimuli or "events" with specific control demands, which can subsequently be retrieved in response to those stimuli/events. However, depending on context, adaptive control and the processes of learning and memory can also be at odds with one another. The studies in this special issue tackle three key themes surrounding learning-control interactions. Editor's Note. This is an introduction to the special issue "The Contribution of Learning and Memory Processes to Cognitive Control." Please see theTable of Contents here:

Between-task transfer of item-specific control is replicable and extends to novel conditions.

“…Of interest in the present study is learning-guided cognitive control (Abrahamse et al, 2016; Braem & Egner, 2018; Bugg & Egner, 2021; Chiu & Egner, 2019; Egner, 2014) and more specifically, item-specific control. Item-specific control refers to the adjustment of cognitive control settings (i.e., relatively focused vs. relatively relaxed) based on learned associations between predictive cues (i.e., stimulus features) and the probability of encountering conflict.…”

Cognitive Flexibility and Stability at the Task-Set Level: A Dual-Dimension Framework

“…This example illustrates that control regulation can be primed or cued by the contextual settings or the broader environment in which cognitive control is demanded and exercised. Empirical evidence supporting this contextual regulation of cognitive control has been extensively documented in recent literature (Abrahamse et al, 2016;Braem et al, 2019;Braem & Egner, 2018;Bugg & Egner, 2021;Chiu & Egner, 2019;Egner, 2014). In recent years following this recognition, a distinct construct has been coined to differentiate contextual regulation of cognitive control from cognitive control per se.…”

Flexibility by association? No evidence for an influence of cue–transition associations on voluntary task switching.

“…These theories make a clear distinction between automatic and controlled processes, and cognitive control has the purpose of counteracting automatic processes. In contrast, recent theories argue that cognitive control can be triggered by low-level learning mechanisms in a bottom-up fashion (Abrahamse et al, 2016;Braem & Egner, 2018;Bugg & Egner, 2021). When a certain control state is repeatedly activated in a particular context (referring to aspects of the stimulus and/or of the response, or contextual, task-irrelevant features), reciprocal associations are formed.…”