Think fast! The role of automaticity in the cognitive control of action

传统上，对人类认知的研究主要集中在执行控制和绩效监控上——这些过程需要记忆、注意力和精神努力。然而，我们的日常生活由大量的行为组成，这些行为在很少有意识的指导下自动进行，长期以来，人们一直对“自动性”在认知控制和行动中的作用感兴趣。高级脑功能的双过程理论假设了两个独立的认知系统——一个是有意识的和反射的，另一个是自动的和反射的——它们指导我们的决定和行动。这两种类型的控制可能是由不同的皮层网络调节的，它们必须在快速反应和选择适当反应的竞争需求下无缝地相互作用。同样，错误检测通常是一个快速且高度自动化的过程，然而，有意识地意识到我们的错误对于学习和强化适当的行为是至关重要的。涉及对意外事件的快速运动反应的实验室任务可以揭示认知控制的许多方面，但迄今为止还没有精确地揭示这两种控制类型之间的转换是如何以及何时发生的。在之前的研究中，我们已经表明，在需要快速反应和抑制的任务中，自动过程和控制过程是并行激活的，并且与有意识控制减少相关的反应并不总是与表现下降相关。此外，一些研究表明，任务表现的改善与反应变异性的减少有关，而不是反应减慢。这使我们产生了一个普遍的假设，即性能的提高（例如，更好的抑制控制）与受控和自动处理之间的最佳转换密切相关，而不是增加执行（自上而下）控制。

The study of human cognition has traditionally focused on executive control and performance monitoring – processes that require memory, attention and mental effort. However, our everyday life consists of a wide repertoire of behaviors that proceed automatically with little conscious guidance, and there has been a long-standing interest in the role of “automaticity” in cognitive control and action. Dual-process theories of higher brain function posit two separate cognitive systems – one conscious and reflective, the other automatic and reflexive – which guide our decisions and actions. These two types of control are likely regulated by different cortical networks, which must seamlessly interact under competing demands of rapid responding and selection of appropriate responses. Similarly, error detection is often a rapid and highly automatic process, yet conscious awareness of our errors is critical in learning and reinforcement of appropriate behaviors. Laboratory tasks involving rapid motor responses to unexpected events can reveal many aspects of cognitive control, but to date have not precisely revealed how and when transitions between these two types of control occur. In previous research, we have shown that automatic and controlled processes were activated in parallel during tasks requiring rapid responding and inhibition, and that responses associated with decreased conscious control were not always associated with decreased performance. Furthermore, several studies have shown that improvements in task performance are associated with decreased response variability, rather than response slowing. This has led us to the general hypothesis that improved performance (e.g., better inhibitory control) is closely related to optimal shifting between controlled and automatic processing, rather than increasing executive (top-down) control. In the proposed research, we will use MEG functional imaging of brain activity during rapid motor response tasks in order to reveal the temporal organization of neural activity that indexes periods of automatic responding, and intrusions of controlled processes during rapid response selection and error processing. We will use a novel combination of kinematic measures and MEG to measure brain activity during goal-directed (reaching) tasks to separate cortical activity related to response selection, initiation, and inhibitory control, and identify specific time points at which controlled processing overrides automatic responses. We will also examine the relationship between periods of automatic responding or off-task mental states and activation of brain networks associated with internalized thought (e.g., "default-mode") during both successful response switching and failed inhibition. This will be achieved using established methods such as self-report probe trials, in addition to unique measures of pupil diameter as an index of fluctuations in attentional control. These studies will allow us to determine whether, in addition to attentional lapses leading to performance errors, such brain states can also reflect periods of accurate performance with minimal executive control, and further, whether there are differences in brain responses signaling early (unconscious) error detection and post-error adjustments during periods of varying automatic control. In a final phase of the project, we plan to adapt our response inhibition tasks to studies in children, with the goal of studying the maturation of controlled and automatic processing with development. This will enhance our understanding of both normal cognitive development and deficits of cognitive control prevalent in many developmental disorders, and provide novel insights into the role of automaticity in human behavior across the lifespan.