Reimplementing a Diagrammatic Reasoning Model in Herbal

Introduction This paper builds upon a study of how people find faults in a simple device and a corresponding cognitive model (Ritter & Bibby, 2008). This existing model, Diag, was implemented in Soar 6 and is based on the idea that learning consists of procedural, declarative, and episodic learning. Diag was developed to analyze human behavior while solving a simple diagrammatic problem (Ritter & Bibby, 2008), a task with similarities to many important real world problem solving tasks. Because Diag predicted astonishing results and is implemented in a version of Soar that is no longer supported, an implementation in an up-to-date cognitive architecture is necessary to make the model available again and more flexible to future changes. We maintained Diag’s basic structure while reimplementing it in a high-level behavior representation language, Herbal, that generates Soar models and can generate different variants more quickly that in Soar directly. Herbal compiles into Soar 9, which allows not only that the model can be used again for further research with current Soar models but it is also made accessible to more researchers. This newly implemented model, called Diag-H, was validated by comparing its predictions to the existing data. It could be shown that Diag-H creates almost the same results as Diag but also incorporates the advantages of Herbal.