Each teacher can experience it every day: students come to science courses with prior knowledge which can act both as building blocks and as obstacles for new learning. It is widely acknowledged that preconceptions are present at both pre-university and university level, in particular in general physics education. These preconceptions may constitute important obstacles to learning since, successfully used in past experiences and contexts, they are considered as a priori ’always true’ by their owners and are then really difficult to overcome.
As engineering teachers at university level, our practices in electricity, electromagnetism and electronics have offered many opportunities (questions in class, lab sessions, exam marking, etc) to realise that our specific context was not immune to this phenomenon. Despite our intuitive efforts and questionings about our teaching approach and material, we have been each year facing repetitive unexpected ‘mistakes’ from students in the context of electricity courses dedicated to second-year engineering students at Université Libre de Bruxelles (ULB). Frustrated that we were not able to get the messages across effectively to students and motivated by the scarcity of published works in our specific niche, we decided to investigate areas in science education related to the teaching situations and issues we used to face.
This thesis was the opportunity to explore the field (model-based learning, conceptual change, epistemological obstacle and pre/misconception) with the aim to improve our understanding, practices and teaching material. One intuitive ’to-be-tested’ idea acted as a starting point: switching the focus from the models themselves, i.e. the substances and subjects we use to learn and teach, to another central concept around which this whole piece of research is gravitating: what we call the domains of validity associated with those models, i.e. the range of situations in which they can be appropriately used and applied regarding the related context and desired outcomes.
By embracing a two-component view of knowledge (considered as the association of a model and a DoV), we propose a new theoretical framework: the Domain of Validity Framework (or DoV framework). This framework explains the obstacle to learning as an overgeneralised DoV. It is specifically designed for developing teaching strategies and for use in the field, with the aim of helping teachers to trigger the overcoming of students’ preconceptions. The instructional techniques derived from this practice-oriented framework confront students with a paradoxical situation so that the student realises the limits of the original DoV and subsequently both searches for a new model and reduces the domain of validity of the original model. This instructional model also emphasises the importance of teaching not just models, but their domains of validity and, then, also means being explicit about the two components of knowledge.
A series of studies integrated to a mixed methods research design has been built to assess the usefulness and effectiveness of our ideas and new framework to help teachers both diagnosing students’ learning obstacles and conceiving teaching strategies, methods and tools to help students to overcome such obstacles. Those studies include analyses of past examinations (summative assessments) and lab tests (formative assessments), the conception and impact analysis of new exercises and lab sessions with pre/post-tests research design, a qualitative inquiry based on student’s interviews, a case study based on the history of Maxwell’s discoveries and an autonomous educational web app aiming to help students overcome their preconceptions in electricity and in basic mechanics.
Wherever we tested it, the implementation of the DoV framework through studies have shown interesting results. Investigating the implications of the concept of Domain of Validity (or DoV) regarding the literature, we have demonstrated the integrative power of our theory in relation to other scientific constructs related to prior knowledge, firstly by resolving apparent oppositions between these constructs, and secondly by redefining (or at least linking with our model) known terms using a small set of precisely defined terms. We have shown that engineering students at university level make mistakes in electricity partly on account of preconceptions they experience in that field, but also highlighted that their preconceptions are mostly different than those provided by the literature. Characterised by its ability to help teachers develop new techniques, the DoV framework has also proven to be a useful and ready-to-use tool for teachers to diagnose difficult-to-overcome students’ learning barriers, to build effective teaching strategies and methods as well as to reconsider the chronological sequence of the contents to be taught.
As experiences and results have been gained, the framework has continued to evolve through iterations and exchanges between the theoretical and on-the-field levels, progressively integrating incremental enhancements opening new doors, new understanding and new applications. It also unveiled some unexpected, interesting and surprising concerns and questions we tried to address, such as the transposition of the DoV framework from a conceptual to a methodological level or the seemingly high interconnectedness existing between our ease to overcome a learning obstacle and our ability to diversify and switch between different modes of representation we use to describe phenomena or situations.
Although we claim that our theory has high integrative power and applicability, it has its own domain of validity like any other model. It does not address all the issues related to prior knowledge and conceptual change. While we have given an example from and tested the theory in our field of electrical engineering, further research is needed to demonstrate its broad applicability across fields of science, the effectiveness of different teaching strategies based on the theory, the relationship with other theories, and the socio-cultural, emotional and affective dimensions of overcoming DoV-based preconceptions.