Hierarchical Control of Cognitive Processes

许多研究表明，从事高要求任务（如开车、木工、打字）的新手对每一步都很有意识，一个接一个地注意细节。另一方面，熟练的执行者只意识到高层次的目标，把细节留给较低层次的过程（层次控制）。这些研究没有回答一个重要的问题：这种等级控制是如何在实践中出现的？专家们如何开发更低层次的控制流程来处理细节？我们提出的研究重点是熟练的打字，其中联想学习使一个单词中的所有击键同时在运动系统中被激活，因此运动系统必须决定何时进行每次击键。新手打字一次激活一个按键，使用高级程序来控制序列顺序。当打字员获得技能时，单词与它们的每一个组成键都联系在一起，所以一个单词的出现会立即激活所有的键，迫使运动系统控制序列顺序。这将创建产生击键的较低级控制过程，这与产生要输入的单词的较高级控制过程不同，从而产生分层控制。拟议的研究将在四个项目中检验这一命题，在这些项目中，研究人员将向打字员提供新型键盘和任务，并在许多次会议中检查学习情况。他们将根据平行激活和对任务细节的有意识意识来衡量层级控制的出现。该项目将对基础研究和应用研究产生重要影响。打字现在已经成为现代生活中无处不在的活动，随着21世纪劳动力的增加，理解这项技能背后的科学将对教育产生广泛的影响。拟议的研究还可能为评价其他需要专门知识的领域的成功培训提供新的标准，超越对速度和准确性的简单测试，对分层控制的出现进行更复杂的测试。更一般地说，这些项目将促进认知控制研究和运动控制研究之间的新联系，因为这两者都是打字所需要的。

Many studies have shown that novice performers in demanding tasks (e.g., driving, carpentry, typing) are conscious of each step, attending to the details one by one. On the other hand, skilled performers are only conscious of high-level goals, leaving the details to lower-level processes (hierarchical control). These studies leave an important question unanswered: How does this hierarchical control emerge with practice? How do experts develop lower-level control processes that take care of the details? The proposed research focuses on skilled typewriting, where associative learning causes all the keystrokes in a word to be activated in the motor system at the same time, so the motor system must decide when to make each keystroke. Novice typists activate one keystroke at a time, using higher-level processes to control serial order. As typists acquire skill, words are associated with each of their constituent keystrokes, so the presentation of a word activates all of its keystrokes at once, forcing the motor system to control serial order. This creates a lower-level control process that produces keystrokes, which is distinct from the higher-level control process that produces words to be typed, resulting in hierarchical control. The proposed research will test this proposition in four projects in which the investigators will present typists with novel types of keyboards and tasks, and examine learning over many sessions. They will measure the emergence of hierarchical control in terms of parallel activation and the conscious awareness of the details of the task. The project will have an important impact on basic and applied research. Typing has now become a ubiquitous activity in modern life, and understanding the science behind the skill will have a broad impact on education as our workforce ramps up for the 21st century. The proposed research may also provide new criteria for evaluating successful training in other areas requiring expertise, going beyond simple tests of speed and accuracy to more sophisticated tests for the emergence of hierarchical control. More generally, the projects will promote new connections between research on cognitive control and research on motor control, since both are required in typewriting.