论文部分内容阅读
通过质疑基于模型推理的认知论题,尝试对在自然化认识论纲领下的认知-历史分析方法进行规范性研究。类比建模是基于模型(Model-based)推理的主要形式之一,类比建模的基本机制主要包括两个部分:一是对模型来源的泛化抽象,二是基于目标域的特征对模型来源进行限制或修正。这两步反复操作,最终构造出适用于目标对象域的模型。模型与对象域的“适切性”(fitness)则是对以上机制恰当性的基本评价标准,类型层级理论对相似性和差异性的分析,为测度“适切性”提供了一条可操作的方法。基于类型层级理论,并结合贝叶斯方法可以解释类比建模何以能够提高模型的可信度,以及类比的创造性与科学合理性之间的关系。这一工作对基于模型推理的科学认知论题的提出了一种可能的反驳。
By questioning cognitive issues based on model reasoning, we try to conduct a normative study of the cognitive-historical analysis methods under the epistemology of naturalization. Analogical modeling is one of the main forms based on model-based reasoning. The basic mechanism of analogical modeling consists of two parts: one is the generalization and abstraction of the source of the model, the other is based on the characteristics of the target domain, Limit or amend. This two-step operation is repeated to finally construct a model suitable for the target object domain. The “fitness” of the model and the object domain is the basic evaluation criterion for the appropriateness of the above mechanism. The analysis of the similarity and difference of the type hierarchy theory provides a measure of “fitness” A workable method. Based on the theory of type hierarchy and Bayesian method, we can explain why analogical modeling can improve the credibility of the model and the relationship between analogy creativity and scientific rationality. This work proposes a possible refutation of scientific cognitive topics based on model reasoning.