Close
This site uses cookies

By using this site, you consent to our use of cookies. You can view our terms and conditions for more information.

Return to Session

Detecting Learning Phases to Improve Performance Prediction

Authors
Mr. Michael Collins
Cognitive Models and Agents Branch ~ Cognitive Models and Agents Branch
Dr. Caitlin Tenison
Educational Testing Service ~ Cognitive and Technology Sciences
Kevin Gluck
Air Force Research Laboratory ~ 711 HPW/RHAC
Prof. John Anderson
Carnegie Mellon University, United States of America
Abstract

Models of learning and retention make predictions of human performance based on the interaction of cognitive mechanisms with temporal features such as the number of repetitions, time since last presentation, and item spacing. These features have been shown to consistently influence performance across a variety of domains. Typically omitted from these accounts are the changes in cognitive process and key mechanisms used by people while acquiring a skill. Here we integrate a model of skill acquisition (Tenison & Anderson, 2016) with the Predictive Performance Equation (PPE; Walsh, Gluck, Gunzelmann, Jastrzembski, & Krusmark, 2019) using Bayesian change detection (Lee, 2019). Our results show this allows for both better representation of an individual’s performance during training and improved out-of-sample prediction.

Discussion
New
From the live Q&A chat Last updated 3 months ago

To not lose the reference, here is the post from Frank Ritter in the live Q&A: you might like: Delaney, P. F., Reder, L. M., Staszewski, J. J., & Ritter, F. E. (1998). The strategy specific nature of improvement: The power law applies by strategy within task. Psychological Science, 9(1), 1-8. 
It shows something similar to your strategy shif...

Dr. Leslie Blaha 2 comments
Calibrating the TAPPED model Last updated 3 months ago

Thanks for the very informative presentation! I was quite inspired by this project. I have one question regarding the calibration of your TAPPED model. If I understood correctly, you took the correlation (r) and root mean squared deviation (RMSD) as a way to "calibrate" your model (as shown on Table 2), and then retained the parameter values that l...

Mr. Pierre Gianferrara 2 comments
order of mechanisms and boredom Last updated 3 months ago

Very cool stuff! I was wondering whether your TAPPED model also allows for participants jumping back (e.g., from automatic to associative). I can imagine that people do shuttle back and forth in optimizing these stages of learning, especially in complex situations. I was also wondering whether the model would benefit from the addition of a "boredo...

Marieke Van Vugt 2 comments