EXPLORING THE SAC MODEL OF MEMORY

探索记忆的 SAC 模型

• 批准号: 6128007
• 负责人: LYNNE Marie REDER
• 金额: $ 14.91万
• 依托单位: CARNEGIE-MELLON UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 1996
• 资助国家: 美国
• 起止时间: 1996-08-01 至 2004-04-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/6128007
• 关键词: behavioral /social science research tag; clinical research; cognition; computational neuroscience; experience; human subject; memory; model design /development; neural information processing; perceptual distortions; psychological models; psychological tests; semantics

The goal of this proposal is to help achieve a unified model of implicit and explicit memory phenomena and memory distortions that span higher level conceptual and lower level perceptual effects. SAC, which stands for Source Of Activation Confusion, will be tested and challenged in several ways. First, novel predictions (i.e., not necessarily expected by other accounts) are made that could disconfirm SAC, and second, each data set will be modeled at a detailed level using SAC. Although SAC has already achieved some success in explaining and fitting data in some domains, there remain many issues to explore that were not addressed during the previous funding period. The goal is to test predictions at higher and lower levels of description. These include testing hypotheses concerning how perceptual features are integrated with conceptual features representations, and how representations change with experience, i.e., how and when higher level structures are built, and whether SAC's memory representation and processes work in an analogous fashion for elaborations (conceptual processing) and for perceptual features (data driven processing). In sum, the goals are to further test and attempt to disconfirm SAC by (a) challenging the model to account for the wide range of cross-over dissociations in the implicit/explicit memory domain, as well as memory confabulations; (b) test the model in domains where purported results already challenge SAC's account; (c) develop a more detailed representation and a more completely specified process model of encoding, matching and retrieval; and (d) shed light on aspects of cognitive functioning that will constrain theory building (within SAC and other cognitive theories). This theoretical advance will be important in helping to understand human memory, both its strength and fallibilities.

这个建议的目标是帮助实现一个统一的模型，内隐和外显记忆现象和记忆扭曲，跨越更高层次的概念和更低层次的知觉效果。 SAC，代表激活混淆的来源，将在几个方面进行测试和挑战。 首先，新的预测（即，不一定是其他账户预期的），可能会否定SAC，其次，每个数据集将使用SAC在详细级别上建模。 虽然SAC在解释和拟合某些领域的数据方面已经取得了一些成功，但仍有许多问题需要探索，而这些问题在上一个资助期内没有得到解决。 我们的目标是在更高和更低的描述水平上测试预测。这些包括测试关于感知特征如何与概念特征表征相结合的假设，以及表征如何随经验而变化，即，如何以及何时建立更高层次的结构，以及SAC的记忆表示和处理是否以类似的方式为精心制作（概念处理）和感知特征（数据驱动处理）工作。 总之，我们的目标是进一步测试和试图否定SAC，方法是：（a）挑战模型，以解释内隐/外显记忆领域中广泛的交叉解离，以及记忆虚构;（B）在声称的结果已经挑战SAC解释的领域中测试模型;（c）发展更详细的表示方法和更完整的编码、配对和检索程序模式;（d）阐明认知功能的某些方面，这些方面将限制理论的建立（在SAC和其他认知理论中）。 这一理论上的进步将有助于理解人类记忆的力量和错误。