Rules through Recursion: How Interactions between the Frontal Cortex and Basal Ganglia May Build Abstract, Complex Rules from Concrete, Simple Ones
Rules through Recursion: How Interactions between the Frontal Cortex and Basal Ganglia May Build Abstract, Complex Rules from Concrete, Simple Ones
复制标题
通过递归的规则：额叶皮层和基底神经节之间的相互作用如何从具体、简单的规则构建抽象、复杂的规则
DOI：
10.1093/acprof:oso/9780195314274.003.0022
复制
发表时间：
2007
影响因子：
15
通讯作者：
T. J. Buschman
中科院分区：
文献类型：
--
作者：
E. Miller;T. J. Buschman
（2006）注意和记忆相关的反应，神经元在外侧顶内区在空间和形状延迟匹配样本任务。大脑已经进化到处理两个相互矛盾的要求它必须对熟悉的情况做出快速反应，同时能够适应新的情况并为未来做好计划。以反射性或习惯性的方式对直接环境做出快速反应相对简单：熟悉的刺激激活了建立良好的神经通路，产生了刻板的神经元。这就是所谓的“自下而上”或“刺激驱动”的处理。这些行为可以快速自动地执行，因为它们是“混凝土”;它们依赖于特定的刺激-反应关系，相同的提示总是会产生相同的反应。神经科学的一个公理是，这种反射反应是通过反复激活神经通路形成的，这加强了它们的连接。然后，它们可以简单地触发-以自动的方式发射，几乎没有变化，因此几乎不需要内部监督。相比之下，真正复杂的、目标导向的行为需要不同的操作模式。必须解决新的情况，目标方向要求有能力对熟悉的环境采取行动，而不仅仅是做出反应。在复杂的情况下导航以实现长期计划的目标不能依赖于不协调的反应。它们必须从你自己内部“自上而下”地精心安排。通过获取和建立世界如何运作的知识，我们可以预测什么结果是可取的，并确定什么战略将有助于实现这些目标。然而，简单地记录和重放以前的经验是不够的。相关的关系需要从虚假的巧合中整理出来，聪明的动物得到了他们经验拼图的“大画面”：他们在广泛的经验中找到了共同的结构，形成了“抽象”规则--可以很容易地适应新情况的普遍原则。这些抽象的规则是总体原则和一般概念，是高层次思维的基础。提供419
(2006) Attention and memory-related responses of neurons in the lateral intraparietal area during spatial and shape-delayed match-to-sample tasks. The brain has evolved to deal with two competing requirements-it must respond quickly to familiar situations while being able to adapt to novel ones and plan for the future. Quickly responding to the immediate environment in a reflexive, or habitual, fashion is relatively straightforward: Familiar stimuli activate well-established neural pathways that produce stereotyped behav iors. This is so-called "bottom-up," or "stimulus-driven," processing. These behaviors can be executed quickly and automatically because they are "con crete"; they rely on specific stimulus-response relationships, and the same cue always elicits the same response. It is an axiom of neuroscience that such re flexive reactions are formed by repeated activation of neural pathways, which strengthens their connections. Then, they can be simply triggered-fired off in an automatic fashion, with little variation and, hence, little need for internal oversight. In contrast, truly sophisticated, goal-directed behavior requires a different m ode of operation. Novel situations must be resolved, and goal direction re quires the ability to act on, not just react to, a familiar environment. Navigat ing complex si tuations to achieve long-planned goals cannot rely on uncoor dinated reactions. They must be orchestrated "top-down" from within oneself. By acquiring and building on knowledge of how the world works, we can predict what outcomes are desirable and determine what strategies will aid in attaining them. However, simply recording and replaying previous experi ences does not suffice. Relevant relationships need to be sorted out from spu rious coincidences, and smart animals get the "big picture" of the jigsaw puzzle of their experiences: They find the common structure across a wide range of experiences to form "abstract" rules-generalized principles that can be readily adapted to novel situations. These abstract rules are the overarching principles and general concepts that are th e basis for high-level thought. They provide 419
登录
查看更多内容
影响因子：
--
作者：
R. Wong;R. Traub;R. Miles
通讯作者：
R. Wong;R. Traub;R. Miles
DOI：
10.1073/pnas.86.22.9015
发表时间：
1989-11-01
影响因子：
11.1
作者：
GOLDMANRAKIC, PS;LERANTH, C;GEFFARD, M
通讯作者：
GEFFARD, M
影响因子：
2.5
作者：
BARBAS, H;DEOLMOS, J
通讯作者：
DEOLMOS, J
影响因子：
2.5
作者：
Asaad, WF;Rainer, G;Miller, EK
通讯作者：
Miller, EK
影响因子：
56.9
作者：
Jog, MS;Kubota, Y;Graybiel, AM
通讯作者：
Graybiel, AM