Marcus G F
Department of Psychology, New York University, NY 10003, USA.
Cogn Psychol. 1998 Dec;37(3):243-82. doi: 10.1006/cogp.1998.0694.
Humans routinely generalize universal relationships to unfamiliar instances. If we are told "if glork then frum," and "glork," we can infer "frum"; any name that serves as the subject of a sentence can appear as the object of a sentence. These universals are pervasive in language and reasoning. One account of how they are generalized holds that humans possess mechanisms that manipulate symbols and variables; an alternative account holds that symbol-manipulation can be eliminated from scientific theories in favor of descriptions couched in terms of networks of interconnected nodes. Can these "eliminative" connectionist models offer a genuine alternative? This article shows that eliminative connectionist models cannot account for how we extend universals to arbitrary items. The argument runs as follows. First, if these models, as currently conceived, were to extend universals to arbitrary instances, they would have to generalize outside the space of training examples. Next, it is shown that the class of eliminative connectionist models that is currently popular cannot learn to extend universals outside the training space. This limitation might be avoided through the use of an architecture that implements symbol manipulation.
人类经常将普遍关系推广到不熟悉的事例上。如果我们被告知“如果是glork，那么就是frum”，以及“是glork”，我们就能推断出“是frum”；任何作为句子主语的名称都可以作为句子的宾语出现。这些普遍性在语言和推理中无处不在。关于它们如何被推广的一种解释认为，人类拥有操纵符号和变量的机制；另一种解释则认为，在科学理论中，符号操纵可以被摒弃，而倾向于用相互连接的节点网络来进行描述。这些“消除式”联结主义模型能提供一种真正的替代方案吗？本文表明，消除式联结主义模型无法解释我们如何将普遍性推广到任意事物上。论证过程如下。首先，如果按照目前的设想，这些模型要将普遍性推广到任意事例上，它们就必须在训练示例的空间之外进行泛化。接下来，文章表明，目前流行的消除式联结主义模型类别无法学会在训练空间之外推广普遍性。通过使用一种实现符号操纵的架构，这种限制或许可以避免。
