Unicorn, Hare, or Tortoise? Using Machine Learning to Predict Working Memory Training Performance
People differ considerably in the extent to which they benefit from working memory (WM) training. Although there is increasing research focusing on individual differences associated with WM training outcomes, we still lack an understanding of which specific individual differences, and in what combin...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Ubiquity Press
2023-09-01
|
| Series: | Journal of Cognition |
| Subjects: | |
| Online Access: | https://account.journalofcognition.org/index.php/up-j-jc/article/view/319 |
| _version_ | 1797656015052734464 |
|---|---|
| author | Yi Feng Anja Pahor Aaron R. Seitz Dennis L. Barbour Susanne M. Jaeggi |
| author_facet | Yi Feng Anja Pahor Aaron R. Seitz Dennis L. Barbour Susanne M. Jaeggi |
| author_sort | Yi Feng |
| collection | DOAJ |
| description | People differ considerably in the extent to which they benefit from working memory (WM) training. Although there is increasing research focusing on individual differences associated with WM training outcomes, we still lack an understanding of which specific individual differences, and in what combination, contribute to inter-individual variations in training trajectories. In the current study, 568 undergraduates completed one of several N-back intervention variants over the course of two weeks. Participants’ training trajectories were clustered into three distinct training patterns (high performers, intermediate performers, and low performers). We applied machine-learning algorithms to train a binary tree model to predict individuals’ training patterns relying on several individual difference variables that have been identified as relevant in previous literature. These individual difference variables included pre-existing cognitive abilities, personality characteristics, motivational factors, video game experience, health status, bilingualism, and socioeconomic status. We found that our classification model showed good predictive power in distinguishing between high performers and relatively lower performers. Furthermore, we found that openness and pre-existing WM capacity to be the two most important factors in distinguishing between high and low performers. However, among low performers, openness and video game background were the most significant predictors of their learning persistence. In conclusion, it is possible to predict individual training performance using participant characteristics before training, which could inform the development of personalized interventions. |
| first_indexed | 2024-03-11T17:23:02Z |
| format | Article |
| id | doaj.art-e2b403c3f1c8465a913b77bf80f087f8 |
| institution | Directory Open Access Journal |
| issn | 2514-4820 |
| language | English |
| last_indexed | 2024-03-11T17:23:02Z |
| publishDate | 2023-09-01 |
| publisher | Ubiquity Press |
| record_format | Article |
| series | Journal of Cognition |
| spelling | doaj.art-e2b403c3f1c8465a913b77bf80f087f82023-10-19T08:10:28ZengUbiquity PressJournal of Cognition2514-48202023-09-0161535310.5334/joc.319318Unicorn, Hare, or Tortoise? Using Machine Learning to Predict Working Memory Training PerformanceYi Feng0https://orcid.org/0000-0002-2485-9720Anja Pahor1https://orcid.org/0000-0002-9396-4620Aaron R. Seitz2https://orcid.org/0000-0003-4936-9303Dennis L. Barbour3https://orcid.org/0000-0003-0851-0665Susanne M. Jaeggi4https://orcid.org/0000-0002-6165-2526University of California, Irvine, School of Education, School of Social Sciences (Department of Cognitive Sciences), Irvine, CaliforniaUniversity of California, Riverside, Department of Psychology, Riverside, California; Northeastern University, Department of Psychology, Boston, Massachusetts, US; University of Maribor, Department of Psychology, MariborUniversity of California, Riverside, Department of Psychology, Riverside, California; Northeastern University, Department of Psychology, Boston, MassachusettsWashington University in St. Louis, Department of Biomedical Engineering, St. Louis, MissouriUniversity of California, Irvine, School of Education, School of Social Sciences (Department of Cognitive Sciences), Irvine, California; Northeastern University, Department of Psychology, Boston, MassachusettsPeople differ considerably in the extent to which they benefit from working memory (WM) training. Although there is increasing research focusing on individual differences associated with WM training outcomes, we still lack an understanding of which specific individual differences, and in what combination, contribute to inter-individual variations in training trajectories. In the current study, 568 undergraduates completed one of several N-back intervention variants over the course of two weeks. Participants’ training trajectories were clustered into three distinct training patterns (high performers, intermediate performers, and low performers). We applied machine-learning algorithms to train a binary tree model to predict individuals’ training patterns relying on several individual difference variables that have been identified as relevant in previous literature. These individual difference variables included pre-existing cognitive abilities, personality characteristics, motivational factors, video game experience, health status, bilingualism, and socioeconomic status. We found that our classification model showed good predictive power in distinguishing between high performers and relatively lower performers. Furthermore, we found that openness and pre-existing WM capacity to be the two most important factors in distinguishing between high and low performers. However, among low performers, openness and video game background were the most significant predictors of their learning persistence. In conclusion, it is possible to predict individual training performance using participant characteristics before training, which could inform the development of personalized interventions.https://account.journalofcognition.org/index.php/up-j-jc/article/view/319machine learningindividual differencesworking memory |
| spellingShingle | Yi Feng Anja Pahor Aaron R. Seitz Dennis L. Barbour Susanne M. Jaeggi Unicorn, Hare, or Tortoise? Using Machine Learning to Predict Working Memory Training Performance Journal of Cognition machine learning individual differences working memory |
| title | Unicorn, Hare, or Tortoise? Using Machine Learning to Predict Working Memory Training Performance |
| title_full | Unicorn, Hare, or Tortoise? Using Machine Learning to Predict Working Memory Training Performance |
| title_fullStr | Unicorn, Hare, or Tortoise? Using Machine Learning to Predict Working Memory Training Performance |
| title_full_unstemmed | Unicorn, Hare, or Tortoise? Using Machine Learning to Predict Working Memory Training Performance |
| title_short | Unicorn, Hare, or Tortoise? Using Machine Learning to Predict Working Memory Training Performance |
| title_sort | unicorn hare or tortoise using machine learning to predict working memory training performance |
| topic | machine learning individual differences working memory |
| url | https://account.journalofcognition.org/index.php/up-j-jc/article/view/319 |
| work_keys_str_mv | AT yifeng unicornhareortortoiseusingmachinelearningtopredictworkingmemorytrainingperformance AT anjapahor unicornhareortortoiseusingmachinelearningtopredictworkingmemorytrainingperformance AT aaronrseitz unicornhareortortoiseusingmachinelearningtopredictworkingmemorytrainingperformance AT dennislbarbour unicornhareortortoiseusingmachinelearningtopredictworkingmemorytrainingperformance AT susannemjaeggi unicornhareortortoiseusingmachinelearningtopredictworkingmemorytrainingperformance |