Dual-process theories of reasoning as a framework for investigating and supporting student reasoning in physics

For over 30 years, research-based materials developed by the physics education research community have helped transform introductory physics instruction. Many of these materials focus on the development of student conceptual understanding and place considerable emphasis on qualitative inferential reasoning. An emerging body of research, however, suggests that poor student performance on certain physics tasks – even after research-based instruction – may stem more from the nature of human reasoning itself than from specific conceptual difficulties. Dual-process theories of reasoning (DPToR), from cognitive science, model human reasoning and decision-making using two processes: fast, automatic, and intuitive process 1 and slow, effortful, and analytical process 2. While DPToR have many similarities to theoretical frameworks more commonly used in PER, they also offer unique affordances. Analysis of student reasoning patterns through the lens of DPToR indicates that students may struggle to engage analytical processing productively when responding to certain physics questions. This phenomenon is particularly pronounced on questions containing perceptually compelling features that may distract students, at least initially, from pursuing more productive reasoning pathways. Work to date has shown that DPToR may account for, in a mechanistic fashion, observed patterns of student reasoning and may play a generative role in both research design and the development of instructional interventions. As part of a multi-institutional effort to leverage DPToR to examine and support student reasoning in physics, novel methodologies to probe student reasoning have been developed (including, for example, reasoning chain construction tasks) and DPToR-aligned instructional interventions aimed at promoting productive and consistent reasoning have been designed and tested. Examples from this ongoing work are used to illustrate the value of dual-process theories for PER.