On the adaptive cognitive control of attention and action

Abstract

Cognition coordinates attention and action to perform tasks through the activity of goal representations. The Dual Modes of Control Theory (Braver, J. R. Gray, & Burgess, 2007) offers an explanation of cognitive control as an interactive network that operates in two distinct modes: (1) proactive control and (2) reactive control. The modes are defined by different temporal patterns of activation in the control network. Cognitive control emerges from the interaction of several brain regions that perform four functions: (1) contextual chunking, (2) competition monitoring, (3) concentrated activation, and (4) consequence accounting.

The experimental evidence presented here provides some support for both hypotheses. The Context Maintenance Hypothesis is supported by the effect of long-term activities on measures of cognitive control mode. The Rational Analysis Hypothesis is supported by evidence that statistical properties of task environments, such as the predictiveness of a contextual cue, effect measures of cognitive control mode. A manipulation of the short-term task environment failed to provide evidence that might have further differentiated the hypotheses.

An understanding of variation in cognitive control is essential to explain phenomena such as multitasking, task-switching, interruptions, action errors, cognitive workload, action selection, attention-deficits, and schizophrenia. Whereas individual difference explanations emphasize the stability of performance constraints, a contextual adaptation account emphasizes the adaptability of task performance. This research will help to understand how both internal and external factors determine cognitive control performance.

All behavior involves a mix of the two cognitive control modes. This thesis attempts to investigate factors that affect the mixture of the control modes. Experimental evidence is organized in terms of two competing hypotheses for the source of the factors that determine cognitive control mode mixture. The Context Maintenance Hypothesis proposes that proactive control is the default mode of control and stable individual differences in the capacity determine the exact mixture of control modes. The Rational Analysis Hypothesis proposes that reactive control is the default mode of control and adaptive, environmental differences in experience, and statistical properties of task environments determine the exact mixture of control modes.