Computational psychometrics
Computational psychometrics is an interdisciplinary field fusing theory-based psychometrics, learning and cognitive sciences, and data-driven AI-based computational models as applied to large-scale/high-dimensional learning, assessment,[1] biometric, or psychological data. Computational psychometrics is frequently concerned with providing actionable and meaningful feedback to individuals based on measurement and analysis of individual differences as they pertain to specific areas of enquiry.
The relatively recent availability of large-scale psychometric data in accessible formats, alongside the rapid increase in CPU processing power, widespread accessibility and application of cluster and cloud computing, and the development of increasingly sensitive instruments for collecting biometric information has allowed big-data analytical and computational methods to expand the scale and scope of traditional psychometric areas of enquiry and modeling.[citation needed]
Pursuing a computational approach to psychometrics often involves scientists working in multidisciplinary teams with expertise in artificial intelligence, machine learning, deep learning and neural network modeling, natural language processing, mathematics and statistics, developmental and cognitive psychology, computer science, data science, learning sciences, virtual and augmented reality, and traditional psychometrics. [citation needed]
Another important subfield of computational science and, specifically, AI is what has been called psychometric artificial intelligence (PAI). PAI involves the use of psychometrically developed evaluations, such as intelligence tests and thinking style tests, to be solved algorithmically by an artificial agent. The goal of PAI is to put to the test the design and processing mechanisms proposed by AI researchers in order to get knowledge from both artificial and natural cognitive systems.[2][3]
Application
[edit]Computational psychometrics incorporates both theoretical and applied components ranging from item response theory, classical test theory, and Bayesian approaches to modeling[4] knowledge acquisition and discovery of network psychometric models.[5] Computational psychometrics studies the computational basis of learning and measurement of traits, such as skills, knowledge, abilities, attitudes, and personality traits via mathematical modeling, intelligent learning and assessment virtual systems,[6] and computer simulation of large-scale, complex data which traditional psychometric approaches are ill-equipped to handle. Recent investigations into these hard to measure constructs include work on collaborative problem solving,[7][8][9][10] teamwork, and decision making, among others.
Computational psychometrics is also related to the study of social complexity. Concepts such as complex systems and emergence have been considered in the study of team assembly and performance. In psychological and medical research it is focused on computational models based on technology enhanced-experimental results. Active areas of enquiry include cognitive, emotional, behavioral, diagnostic, and mental health issues. A computational psychometrics approach in this capacity frequently makes use of emerging capabilities such as biometric and multimodal sensors, virtual and augmented reality, as well as affective and wearable computing technologies.[11]
References
[edit]- ^ von Davier, Alina A. (2017). "Computational Psychometrics in Support of Collaborative Educational Assessments". Journal of Educational Measurement. 54 (1): 3–11. doi:10.1111/jedm.12129.
- ^ Bringsjord, Selmer (September 2011). "Psychometric artificial intelligence". Journal of Experimental & Theoretical Artificial Intelligence. 23 (3): 271–277. doi:10.1080/0952813x.2010.502314. S2CID 19947368.
- ^ Braynen, Alec (2022). Towards More Task-Generalized and Explainable AI through Psychometrics (Thesis).
- ^ Polyak, Stephen T.; von Davier, Alina A.; Peterschmidt, Kurt (2017). "Computational Psychometrics for the Measurement of Collaborative Problem Solving Skills". Frontiers in Psychology. 8: 20–29. doi:10.3389/fpsyg.2017.02029. PMC 5712874. PMID 29238314.
- ^ Marsman, M.; Borsboom, D.; Kruis, J.; Epskamp, S.; van Bork, R.; Waldorp, L.J.; van der Maas, H.L.J.; Maris, G. (2018). "An Introduction to Network Psychometrics: Relating Ising Network Models to Item Response Theory Models". Multivariate Behavioral Research. 53 (1): 15–35. doi:10.1080/00273171.2017.1379379. hdl:11245.1/92b56bb7-f929-4361-919f-0dc02b5eb032. PMID 29111774.
- ^ Greiff, Samuel; Gasevic, Dragan; von Davier, Alina A. (2017). Using process data for assessment in Intelligent Tutoring Systems. A psychometrician's, cognitive psychologist's, and computer scientist's perspective. Army Research Laboratory. pp. 171–179. hdl:10993/32037.
- ^ Innovative Assessment of Collaboration. Methodology of Educational Measurement and Assessment. 2017. doi:10.1007/978-3-319-33261-1. ISBN 978-3-319-33259-8.[page needed]
- ^ von Davier, Alina A.; Hao, Jiangang; Kyllonen, Patrick (2017). "Interdisciplinary research agenda in support of assessment of collaborative problem solving: lessons learned from developing a Collaborative Science Assessment Prototype". Computers in Human Behavior. 76 (November): 631–640. doi:10.1016/j.chb.2017.04.059.
- ^ Flor, Michael; Yoon, Su-Youn; Hao, Jiangang; Liu, Lei; von Davier, Alina (2016). "Automated classification of collaborative problem solving interactions in simulated science tasks". Proceedings of the 11th Workshop on Innovative Use of NLP for Building Educational Applications: 31–41. doi:10.18653/v1/W16-0504. S2CID 390510.
- ^ Flor, Michael; Yoon, Su-Youn; Hao, Jiangang; Liu, Lei; von Davier, Alina A. (June 2016). "Automated classification of collaborative problem solving interactions in simulated science tasks". Proceedings of the 11th Workshop on Innovative Use of NLP for Building Educational Applications. W16-0504. San Diego, CA: Association for Computational Linguistics: 31–41. doi:10.18653/v1/W16-0504. S2CID 390510.
- ^ Cipresso, Pietro; Matic, Aleksandar; Giakoumis, Dimitris; Ostrovsky, Yuri (2015). "Advances in Computational Psychometrics". Computational and Mathematical Methods in Medicine. 2015: 418683. doi:10.1155/2015/418683. PMC 4539436. PMID 26346251.