Studying Memory Decay and Spacing within Knowledge Tracing

Knowledge Tracing estimates a students knowledge on a set of skills and predicts whether the student will answer correctly if given a question linked to subsets of such skills. We conduct an indepth analysis on capturing cognitive science principles such as memory decay and spacing and measure their effects within knowledge tracing. To do this, we propose a new algorithm called MemDec which incorporates memory decay theory into knowledge estimation. This model is further expanded to consider the spacing effect, another pivotal cognitive science concept. We explore different methods of modeling the rate and weight of decay, with and without the spacing effect, and analyze the role they play in predicting student performance within real-world data. Variations of the model are compared between each other as well as against other existing algorithms.