Application of Machine Learning Models for Tracking Participant Skills in Cognitive Training
Sanjana Sandeep, Christian R Shelton, Anja Pahor, Susanne M Jaeggi, Aaron R Seitz, Sanjana Sandeep, Christian R Shelton, Anja Pahor, Susanne M Jaeggi, Aaron R Seitz
Abstract
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.
Keywords: Bayesian filtering; cognitive memory training; deep-learning; hidden Markov model; n-back training; video games.
Copyright © 2020 Sandeep, Shelton, Pahor, Jaeggi and Seitz.
Figures
References
- Au J., Buschkuehl M., Duncan G. J., Jaeggi S. M. (2016). There is no convincing evidence that working memory training is not effective: a reply to melby-lervåg and hulme (2015). Psychon. Bull. Rev. 23, 331–337. 10.3758/s13423-015-0967-4
- Baddeley A. (2012). Working memory: theories, models, and controversies. Annu. Rev. Psychol. 63, 1–29. 10.1146/annurev-psych-120710-100422
- Barton M., Lord F. (1981). An upper asymptote for the three-parameter logistic item-response model. ETS Res. Rep. Ser. 1981, i-8. 10.1002/j.2333-8504.1981.tb01255.x
- Baum L. E., Petrie T., Soules G., Weiss N. R. (1970). A maximization technique occurring in the statistical analysis of probabilistic functions of Markov chains. Ann. Math. Stat. 41, 164–171. 10.1214/aoms/1177697196
- Bengio Y., Frasconi P. (1994). An input output hmm architecture, in Proceedings of the 7th International Conference on Neural Information Processing Systems, NIPS'94 (Cambridge, MA: MIT Press; ), 427–434.
- Breiman L. (1996). Bagging predictors. Mach. Learn. 24, 123–140. 10.1007/BF00058655
- Chan K., Lenard C., Mills T. (2012). An introduction to Markov chains, in The Forty-Nineth Annual Conference of the Mathematical Associatoin of Victoria (Bundoora, VIC: ).
- Dempster A. P., Laird N. M., Rubin D. B. (1977). Maximum likelihood from incomplete data via the em algorithm. J. R. Stat. Soc. Ser. B 39, 1–38. 10.1111/j.2517-6161.1977.tb01600.x
- Deveau J., Jaeggi S. M., Zordan V., Phung C., Seitz A. R. (2015). How to build better memory training games. Front. Syst. Neurosci. 8:243. 10.3389/fnsys.2014.00243
- Green C., Bavelier D., Kramer A., Vinogradov S., Ansorge U., Ball K., et al. (2018). Improving methodological standards in behavioral interventions for cognitive enhancement. J. Cogn. Enhance. 3, 2–29. 10.1007/s41465-018-0115-y
- Hambleton R. K., Swaminathan H. (1984). Item Response Theory: Principles and Applications. New York, NY: Springer. 10.1007/978-94-017-1988-9
- Hochreiter S., Schmidhuber J. (1997). Long short-term memory. Neural Comput. 9, 1735–1780. 10.1162/neco.1997.9.8.1735
- Julier S. J., Uhlmann J. K. (1997). New extension of the Kalman filter to nonlinear systems, in SPIE 3068: Signal Processing, Sensor Fusion, and Target Recognition VI (San Diego, CA: ). 10.1117/12.280797
- Karbach J., Unger K. (2014). Executive control training from middle childhood to adolescence. Front. Psychol. 5:390. 10.3389/fpsyg.2014.00390
- Kriegeskorte N. (2015). Deep neural networks: a new framework for modeling biological vision and brain information processing. Annu. Rev. Vis. Sci. 1, 417–446. 10.1146/annurev-vision-082114-035447
- Martin A., Quinn K. (2002). Dynamic ideal point estimation via Markov chain Monte Carlo for the U.S. supreme court, 1953–1999. Polit. Anal. 10:134 10.1093/pan/10.2.134
- Melby-Lervåg M., Hulme C. (2013). Is working memory training effective? A meta-analytic review. Dev. Psychol. 49, 270–291. 10.1037/a0028228
- Melby-Lervåg M., Redick T. S., Hulme C. (2016). Working memory training does not improve performance on measures of intelligence or other measures of “far transfer”: evidence from a meta-analytic review. Perspect. Psychol. Sci. 11, 512–534. 10.1177/1745691616635612
- Mohammed S., Flores L., Deveau J., Hoffing R., Phung C., Parlett C., et al. . (2017). The benefits and challenges of implementing motivational features to boost cognitive training outcome. J. Cogn. Enhance. 1, 491–507. 10.1007/s41465-017-0047-y
- Pedullà L., Brichetto G., Tacchino A., Vassallo C., Zaratin P., Battaglia M., et al. . (2016). Adaptive vs. non-adaptive cognitive training by means of a personalized app: a randomized trial in people with multiple sclerosis. J. Neuroeng. Rehabil. 13:88. 10.1186/s12984-016-0193-y
- Pergher V., Shalchy M. A., Pahor A., Hulle M. M. V., Jaeggi S. M., Seitz A. R. (2019). Divergent research methods limit understanding of working memory training. J. Cogn. Enhance. 4, 100–120. 10.1007/s41465-019-00134-7
- Rabiner L. R. (1990). A tutorial on hidden markov models and selected applications in speech recognition, in Readings in Speech Recognition, eds Waibel A., Lee K.F. (San Francisco, CA: Morgan Kaufmann Publishers Inc.), 267–296. 10.1016/B978-0-08-051584-7.50027-9
- Rasch G. (1980). Probabilistic Models for Some Intelligence and Attainment Tests. Chicago, IL: University of Chicago Press.
- Rebane J. (2018). Seq 2 Seq RNNs and ARIMA models for cryptocurrency prediction: a comparative study, in Proceedings of the 24th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (London, UK: ).
- Redick T. S. (2019). The hype cycle of working memory training. Curr. Direct. Psychol. Sci. 28, 423–429. 10.1177/0963721419848668
- Rutledge K., van den Bos W., McClure S. M., Schweitzer J. (2012). Training cognition in ADHD: Current findings, borrowed concepts, and future directions. Neurotherapeutics 9, 542–558. 10.1007/s13311-012-0134-9
- Särkkä S. (2013). Bayesian Filtering and Smoothing. New York, NY: Cambridge University Press.
- Seitz A. R. (2018). A new framework of design and continuous evaluation to improve brain training. J. Cogn. Enhance. 2, 78–87. 10.1007/s41465-017-0058-8
- Simons D., Boot W., Charness N., Gathercole S., Chabris C., Hambrick D., et al. . (2016). Do brain-training programs work? Psychol. Sci. Publ. Interest 17, 103–186. 10.1177/1529100616661983
- Soveri A., Antfolk J., Karlsson L., Salo B., Laine M. (2017). Working memory training revisited: a multi-level meta-analysis of n-back training studies. Psychon. Bull. Rev. 24, 1077–1096. 10.3758/s13423-016-1217-0
- Soveri A., Antfolk J., Karlsson L., Salo B., Laine M. (2018). Working memory training revisited: a multi-level meta-analysis of n-back training studies. Psychon. Bull. Rev. 10.31234/
- Stepankova Georgi H., Lukavsky J., Buschkuehl M., Kopecek M., Ripova D., Jaeggi S. (2013). The malleability of working memory and visuospatial skills: a randomized controlled study in older adults. Dev. Psychol. 50, 1049–1059. 10.1037/a0034913
- Valsiner J., Molenaar P., Lyra M., Chaudhary N. (2009). Dynamic Process Methodology in the Social and Developmental Sciences. Dordrecht: Springer; 10.1007/978-0-387-95922-1
- Wang X., Berger J., Burdick D. (2013). Bayesian analysis of dynamic item response models in educational testing. Ann. Appl. Stat. 7, 126–153. 10.1214/12-AOAS608
- Wenliang L., Seitz A. (2018). Deep neural networks for modeling visual perceptual learning. J. Neurosci. 38, 1620–1627. 10.1523/JNEUROSCI.1620-17.2018
Source: PubMed