Application of Machine Learning Models for Tracking Participant Skills in Cognitive Training
A key need in cognitive training interventions is to personalize task difficulty to each user and to adapt this difficulty to continually apply appropriate challenges as users improve their skill to perform the tasks. Here we examine how Bayesian filtering approaches, such as hidden Markov models an...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2020-07-01
|
| Series: | Frontiers in Psychology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/article/10.3389/fpsyg.2020.01532/full |
| _version_ | 1811294434054111232 |
|---|---|
| author | Sanjana Sandeep Sanjana Sandeep Christian R. Shelton Anja Pahor Anja Pahor Susanne M. Jaeggi Aaron R. Seitz Aaron R. Seitz |
| author_facet | Sanjana Sandeep Sanjana Sandeep Christian R. Shelton Anja Pahor Anja Pahor Susanne M. Jaeggi Aaron R. Seitz Aaron R. Seitz |
| author_sort | Sanjana Sandeep |
| collection | DOAJ |
| description | A key need in cognitive training interventions is to personalize task difficulty to each user and to adapt this difficulty to continually apply appropriate challenges as users improve their skill to perform the tasks. Here we examine how Bayesian filtering approaches, such as hidden Markov models and Kalman filters, and deep-learning approaches, such as the long short-term memory (LSTM) model, may be useful methods to estimate user skill level and predict appropriate task challenges. A possible advantage of these models over commonly used adaptive methods, such as staircases or blockwise adjustment methods that are based only upon recent performance, is that Bayesian filtering and deep learning approaches can model the trajectory of user performance across multiple sessions and incorporate data from multiple users to optimize local estimates. As a proof of concept, we fit data from two large cohorts of undergraduate students performing WM training using an N-back task. Results show that all three models predict appropriate challenges for different users. However, the hidden Markov models were most accurate in predicting participants' performances as a function of provided challenges, and thus, they placed participants at appropriate future challenges. These data provide good support for the potential of machine learning approaches as appropriate methods to personalize task performance to users in tasks that require adaptively determined challenges. |
| first_indexed | 2024-04-13T05:17:22Z |
| format | Article |
| id | doaj.art-54890a82d23441dd9aa3cbc3ced20ce0 |
| institution | Directory Open Access Journal |
| issn | 1664-1078 |
| language | English |
| last_indexed | 2024-04-13T05:17:22Z |
| publishDate | 2020-07-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Psychology |
| spelling | doaj.art-54890a82d23441dd9aa3cbc3ced20ce02022-12-22T03:00:51ZengFrontiers Media S.A.Frontiers in Psychology1664-10782020-07-011110.3389/fpsyg.2020.01532514349Application of Machine Learning Models for Tracking Participant Skills in Cognitive TrainingSanjana Sandeep0Sanjana Sandeep1Christian R. Shelton2Anja Pahor3Anja Pahor4Susanne M. Jaeggi5Aaron R. Seitz6Aaron R. Seitz7Department of Computer Science, University of California, Riverside, Riverside, CA, United StatesBrain Game Center, University of California, Riverside, Riverside, CA, United StatesDepartment of Computer Science, University of California, Riverside, Riverside, CA, United StatesBrain Game Center, University of California, Riverside, Riverside, CA, United StatesSchool of Education, University of California, Irvine, Irvine, CA, United StatesSchool of Education, University of California, Irvine, Irvine, CA, United StatesBrain Game Center, University of California, Riverside, Riverside, CA, United StatesDepartment of Psychology, University of California, Riverside, Riverside, CA, United StatesA key need in cognitive training interventions is to personalize task difficulty to each user and to adapt this difficulty to continually apply appropriate challenges as users improve their skill to perform the tasks. Here we examine how Bayesian filtering approaches, such as hidden Markov models and Kalman filters, and deep-learning approaches, such as the long short-term memory (LSTM) model, may be useful methods to estimate user skill level and predict appropriate task challenges. A possible advantage of these models over commonly used adaptive methods, such as staircases or blockwise adjustment methods that are based only upon recent performance, is that Bayesian filtering and deep learning approaches can model the trajectory of user performance across multiple sessions and incorporate data from multiple users to optimize local estimates. As a proof of concept, we fit data from two large cohorts of undergraduate students performing WM training using an N-back task. Results show that all three models predict appropriate challenges for different users. However, the hidden Markov models were most accurate in predicting participants' performances as a function of provided challenges, and thus, they placed participants at appropriate future challenges. These data provide good support for the potential of machine learning approaches as appropriate methods to personalize task performance to users in tasks that require adaptively determined challenges.https://www.frontiersin.org/article/10.3389/fpsyg.2020.01532/fullcognitive memory traininghidden Markov modelBayesian filteringvideo gamesn-back trainingdeep-learning |
| spellingShingle | Sanjana Sandeep Sanjana Sandeep Christian R. Shelton Anja Pahor Anja Pahor Susanne M. Jaeggi Aaron R. Seitz Aaron R. Seitz Application of Machine Learning Models for Tracking Participant Skills in Cognitive Training Frontiers in Psychology cognitive memory training hidden Markov model Bayesian filtering video games n-back training deep-learning |
| title | Application of Machine Learning Models for Tracking Participant Skills in Cognitive Training |
| title_full | Application of Machine Learning Models for Tracking Participant Skills in Cognitive Training |
| title_fullStr | Application of Machine Learning Models for Tracking Participant Skills in Cognitive Training |
| title_full_unstemmed | Application of Machine Learning Models for Tracking Participant Skills in Cognitive Training |
| title_short | Application of Machine Learning Models for Tracking Participant Skills in Cognitive Training |
| title_sort | application of machine learning models for tracking participant skills in cognitive training |
| topic | cognitive memory training hidden Markov model Bayesian filtering video games n-back training deep-learning |
| url | https://www.frontiersin.org/article/10.3389/fpsyg.2020.01532/full |
| work_keys_str_mv | AT sanjanasandeep applicationofmachinelearningmodelsfortrackingparticipantskillsincognitivetraining AT sanjanasandeep applicationofmachinelearningmodelsfortrackingparticipantskillsincognitivetraining AT christianrshelton applicationofmachinelearningmodelsfortrackingparticipantskillsincognitivetraining AT anjapahor applicationofmachinelearningmodelsfortrackingparticipantskillsincognitivetraining AT anjapahor applicationofmachinelearningmodelsfortrackingparticipantskillsincognitivetraining AT susannemjaeggi applicationofmachinelearningmodelsfortrackingparticipantskillsincognitivetraining AT aaronrseitz applicationofmachinelearningmodelsfortrackingparticipantskillsincognitivetraining AT aaronrseitz applicationofmachinelearningmodelsfortrackingparticipantskillsincognitivetraining |