A Non-Linear Electronic Textbook Format to Facilitate Deep Learning

Many educators are calling for an increased emphasis on the interactions between the concepts covered in an educational unit because of its benefits for learning. There is evidence that at present, standard linear textbooks do not adequately support this ``deep'' approach to curriculum design, and we think this is partially due to the linear format. In this thesis, we propose a non-linear electronic textbook format that organizes content in a way that accords with cognitive science theories of how concepts are represented in the mind. Structuring content in this way makes the relation between ideas more explicit, and reduces extraneous cognitive load, which we believe will aid learning. We implemented a functional prototype of this design, and we designed a study to measure it in terms of pedagogical efficacy (objective and self-reported), usability, and overall likeability, in comparison with a linear control.