Computer testing is currently widely used in medicine, psychology and education with the purpose of diagnostics, evaluations of competence level and applicability to perform various functions, including education quality control. The quality of testing and validity of its results largely depend on the testing technologies which have been the scope of scientific research in the recent decades. The article describes a new approach to assessment of human cognitive abilities. The problem-solving process employs an adaptive testing model based on the resources of educable Markov models. The testee is offered a specific test where every task includes manipulations with a technical system in real-time and real-space scale. The assessment of cognitive abilities relies on the information characterizing internal dynamics of solving the test tasks. The described approach is free from such typical drawbacks of computer testing as increase of probability of randomly chosen correct answers, loss of information on logics behind the problem-solving and the process of solving separate tasks
Keywords: cognitive abilities, tests of cognitive abilities, finite state automation, computer testing, diagnostics
Kuravsky L.S., Baranov S.N. The concept of multifactor Markov networks and
its application to forecasting and diagnostics of technical systems// Proc.
Condition Monitoring. Cambridge, 2005.
Kuravsky L.S., Baranov S.N., Yuryev G.A. Synthesis and identification of
hidden Markov models based on a novel statistical technique in condition
monitoring//Proc. 7th International Conference on Condition Monitoring
&Machinery Failure Prevention Technologies. Stratford-upon-Avon, 2010.