Making Instance-based Learning Theory Usable, Transparent, and Understandable: Instance-based Learning Tool

摘要

Although cognitive models of human behavior enjoy a rich history in cognitive psychology, they lack awidespread impact, partly due to the complexities of the modeling process including the need to knowsoftware programming. We will demonstrate a modeling tool, called IBLTool, which represents InstancebasedLearning Theory (IBLT). IBLT is a theory of decisions from experience in interactive, dynamicenvironments. IBLTool addresses the complexity and programming challenges in cognitive modeling andhelps in the development of cognitive models for a particular task. The IBLTool makes IBLT usable,transparent, and understandable for the cognitive modeling community. The tool uses graphical userinterfaces to represent the modeling process and knowledge representations from the IBLT in a step-bystepmanner. To derive predictions of human behavior in the IBLTool for a decision task, a taskimplementation, independent from the IBLTool connects and interacts with the tool interactively as ahuman would. Data about human behavior is collected as part of the task, while the IBLTool collects dataabout the execution of the model and its parameters. We will introduce IBLT and provide a step-by-stepdemonstration of building cognitive models with the IBLTool.