Using Prior Data to Inform Initial Performance Predictions of Individual Students


The predictive performance equation (PPE) is a mathematical model of learning and retention that uses regularities seen in human learning to predict future performance. Previous research (Collins, Gluck, Walsh Krusmark & Gunzelmann, 2016) found that prior data could be used to inform PPE’s free parameters when generating predictions of a group’s aggregate performance, allowing for more accurate initial performance predictions. Here we investigate an extension of this methodology to predict performance of individuals, rather than aggregate samples. This paper documents the results of that investigation, which is on the critical path to the use of this cognitive technology in education and training.

Back to Table of Contents