Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Response Surface Methodology01:16

Response Surface Methodology

Response Surface Methodology (RSM) is a collection of statistical and mathematical techniques used to develop, improve, and optimize processes. It is particularly valuable when many input variables or factors potentially influence a response variable.
The process of RSM involves several key steps:
Survival Tree01:19

Survival Tree

Survival trees are a non-parametric method used in survival analysis to model the relationship between a set of covariates and the time until an event of interest occurs, often referred to as the "time-to-event" or "survival time." This method is particularly useful when dealing with censored data, where the event has not occurred for some individuals by the end of the study period, or when the exact time of the event is unknown.
 Building a Survival Tree
Constructing a survival tree begins...
Methods of Medium Optimization01:28

Methods of Medium Optimization

Optimizing growth media enhances microbial proliferation and maximizes product yield. Statistical experimental design methodologies provide structured and reproducible approaches, offering progressively higher levels of robustness and efficiency.The One-Factor-at-a-Time (OFAT) MethodThe One-Factor-at-a-Time (OFAT) method involves adjusting a single variable while keeping all others constant. However, it cannot detect interactions between variables, often leading to suboptimal outcomes when...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Dependent latent class partial credit models for careless and insufficient effort responding in online survey data.

Behavior research methods·2026
Same author

Planned missingness in intensive longitudinal studies: Extensions and comparisons of multiform designs.

Behavior research methods·2026
Same author

Bayesian evaluation for latent variable models: A tutorial on computing information criteria and bayes factors with the r package bleval.

Psychological methods·2026
Same author

Three-level vector autoregressive models.

Psychological methods·2026
Same author

Multidimensional School Climate and Mental Health Among Chinese Vocational High School Students: The Role of Personal Growth Initiative.

Behavioral sciences (Basel, Switzerland)·2026
Same author

A transition evaluation model with probability-based effectiveness indicators-a new measurement model for problem-solving process data.

Behavior research methods·2026

Related Experiment Video

Updated: Jun 20, 2026

Operant Procedures for Assessing Behavioral Flexibility in Rats
08:30

Operant Procedures for Assessing Behavioral Flexibility in Rats

Published on: February 15, 2015

20.8K

An adaptive testing item selection strategy via a deep reinforcement learning approach.

Pujue Wang1,2,3, Hongyun Liu4,5, Mingqi Xu6

  • 1Beijing Key Laboratory of Learning and Cognition, School of Psychology, Capital Normal University, No. 23 Bai Dui Zi Jia, Beijing, 100048, China.

Behavior Research Methods
|September 13, 2024
PubMed
Summary

A new deep Q-network (DQN) strategy enhances computerized adaptive testing (CAT) by optimizing item selection. This reinforcement learning approach shows improved accuracy over traditional methods in simulated and real-world assessments.

Keywords:
Computerized adaptive testingDeep Q-networkDeep learningItem selection strategyReinforcement learning

More Related Videos

The Double-H Maze: A Robust Behavioral Test for Learning and Memory in Rodents
09:01

The Double-H Maze: A Robust Behavioral Test for Learning and Memory in Rodents

Published on: July 8, 2015

12.5K
New Variations for Strategy Set-shifting in the Rat
09:45

New Variations for Strategy Set-shifting in the Rat

Published on: January 23, 2017

8.2K

Related Experiment Videos

Last Updated: Jun 20, 2026

Operant Procedures for Assessing Behavioral Flexibility in Rats
08:30

Operant Procedures for Assessing Behavioral Flexibility in Rats

Published on: February 15, 2015

20.8K
The Double-H Maze: A Robust Behavioral Test for Learning and Memory in Rodents
09:01

The Double-H Maze: A Robust Behavioral Test for Learning and Memory in Rodents

Published on: July 8, 2015

12.5K
New Variations for Strategy Set-shifting in the Rat
09:45

New Variations for Strategy Set-shifting in the Rat

Published on: January 23, 2017

8.2K

Area of Science:

  • Educational Measurement and Psychometrics
  • Artificial Intelligence in Education
  • Computerized Adaptive Testing (CAT)

Background:

  • Computerized adaptive testing (CAT) traditionally optimizes item selection based on immediate test information.
  • Recent advances in reinforcement learning (RL) and deep neural networks (DNNs) offer potential for more sophisticated item selection strategies.
  • Existing CAT methods may not fully leverage information from the entire item pool for optimal examinee assessment.

Purpose of the Study:

  • To reformulate CAT within a reinforcement learning framework.
  • To propose and evaluate a novel item selection strategy using the deep Q-network (DQN) method.
  • To compare the performance of the DQN-based strategy against traditional CAT item selection methods.

Main Methods:

  • Developed a DQN-based algorithm for item selection in CAT.
  • Conducted simulated studies using various item banks and examinee response distributions.
  • Validated the DQN strategy using empirical data from real-world item banks and examinee responses.
  • Compared DQN performance against five traditional item selection strategies: maximum Fisher information, Fisher information weighted by likelihood, Kullback‒Leibler information weighted by likelihood, maximum posterior weighted information, and maximum expected information.
  • Investigated the impact of sample size and trait level distribution during training on DQN performance.

Main Results:

  • The DQN-based item selection strategy demonstrated lower Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) compared to traditional methods.
  • Superior performance of DQN was observed across most conditions in both simulated and real data scenarios.
  • The study provides insights into monitoring the training process for optimal Q-network convergence.

Conclusions:

  • The deep Q-network (DQN) approach represents a significant advancement in computerized adaptive testing item selection.
  • DQN-based strategies offer improved accuracy and efficiency in estimating examinee trait levels.
  • The study provides practical guidance and code for implementing DQN-based CAT strategies.