Automatic Process Model Generation for Cognitive Modeling

用于认知建模的自动过程模型生成

• 批准号: 529624975
• 负责人: Professor Dr. Marco Ragni
• 金额: --
• 依托单位: Professur für Prädiktive Verhaltensanalyse
• 依托单位国家: 德国
• 项目类别: Research Grants
• 资助国家: 德国
• 项目状态: 未结题
• 来源: https://gepris.dfg.de/gepris/projekt/529624975?language=en
• 关键词: Automatic; Process; Model; Generation; Cognitive

The goal of this project is to achieve a paradigm shift in the cognitive modeling of human reasoning processes, which has so far focused on cognitive theories for aggregated responses. By identifying and evaluating subcomponents and atomic processes of existing theories and formalizing the space of cognitive operations in which specific parameters for interindividual differences are integrated, we aim to generate cognitive process models in an automated manner that can achieve high predictive performance and adaptively adjust to the individual, thus allowing for individualized cognitive modeling of human inference, and thus represents a solution to the problems identified in the state of the art and identified potential for improvement in current inference research. Since the focus of component recombination is on predictive power, and the method is thus free of additional theoretical assumptions, it can be guaranteed that the cognitive processes are selected solely on the basis of their actual added value for the process model.

该项目的目标是实现人类推理过程的认知建模的范式转变，迄今为止，该模型一直专注于聚合反应的认知理论。通过识别和评估现有理论的子组件和原子过程，并将其中整合了个体间差异的特定参数的认知操作空间形式化，我们的目标是以自动化的方式生成认知过程模型，该模型可以实现高预测性能并自适应地调整个体，从而允许对人类推理进行个性化的认知建模，并因此代表了对现有技术中所识别的问题的解决方案以及当前推理研究中所识别的改进潜力。由于成分重组的焦点在于预测能力，因此该方法没有额外的理论假设，因此可以保证仅基于认知过程对过程模型的实际附加值来选择认知过程。