AN IMPLICIT SCALING MODEL OF COMPARATIVE JUDGMENT

我们的许多一般世界知识都是关于属性和属性的 对象的数量。通常，这些属性构成 基础连续体是“可分级的”(例如，冷、热等)。这个 早期提案的主要目的是记录类似的效果 关于范畴判断(例如，“鲸鱼个头大”)和关系的背景 (例如，“鲸鱼比跳蚤大”)。这次会议的重点是 目前的建议是开发这些财产使用的正式模式 并将这一解释扩展到语义的影响 对这些判断进行组织。 这项建议假定对可分级物业的评估是尊重的。 到与当前上下文相关联的规范。在这种概念化中， 当前的上下文被描述为“参考类”，规范是 被描述为位于最极端值处的“参考点” 与引用类关联。根据建议的模型 比较判断，相对幅度是通过比较 每个备选方案到公共参考点的距离。这些 机制被扩展到分类判断，通过假设这些 判断是通过比较物体和物体之间的距离来做出的 指向标准的参考点。建议进行更多实验，以 检验概念类别在以下方面发挥重要作用的假设 建立参考类。 理解语义记忆中的这些基本问题是至关重要的 以确定特定情况下记忆损伤的确切性质 各种形式的痴呆症。例如，阿尔茨海默病患者似乎有 在正常老年人中不存在的语义记忆功能受损(Nebe， (1989年)。目前，尚不清楚这些损伤是否源于 主要是因为丢失了语义信息，降低了可访问性 这些信息，或对其使用的后续限制。一项决议 在没有足够的语义理论的情况下，这些问题是不可能的 记忆。

Much of our general world knowledge is about the properties and attributes of objects. Frequently, these properties constitute a location on an underlying continuum that is "gradable" (e.g., cold, hot, etc.). The primary aim of an earlier proposal was to document similar effects of context on categorical judgments (e.g., "whales are large") and relational judgments (e.g., "whales are larger than fleas"). The focus of the current proposal is to develop a formal model of these uses of property knowledge and to extend this account to the effects of semantic organization on these judgments. This proposal assumes that gradable properties are evaluated with respect to norms associated with the current context. In this conceptualization, the current context is described as a "reference class" and the norms are described as "reference points" located at the most extreme values associated with the reference class. According to the proposed model of comparative judgment, relative magnitude is determined by comparing the distances of each alternative to a common reference point. These mechanisms are extended to categorical judgments by assuming that these judgments are made by comparing the distance between an object and a reference point to a criterion. Additional experiments are proposed to test the hypothesis that conceptual categories play an important role in establishing reference classes. An understanding of these fundamental issues in semantic memory is crucial to a determination of the exact nature of memory impairments in certain forms of dementia. For example, Alzheimer patients appear to have impaired semantic memory function that is absent in the normal old (Nebes, 1989). Currently, it is not known whether these impairments stem primarily from a loss of semantic information, reduced accessibility of this information, or subsequent limitations on its use. A resolution of these issues is unlikely in the absence of adequate theories of semantic memory.