An empirical comparison of forgetting models

This paper presents a comparative study of several forgetting models. We compare the performance of these forgetting models in measuring the amount of forgetting following a break in an environment where learning and forgetting occur intermittently. Comparisons are made of the performance of each model with respect to empirical data from a procedural inspection task. We identify models that perform better for predicting forgetting and, hence, performance following breaks in intermittent tasks. Results are presented indicating the efficiency, stability, and balance of each model with respect to predictive performance. Accurate predictions of performance are important for management in determining labor costing, scheduling, and setting time standards.