Shaoyang Guo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Shaoyang Guo has a total of 29 co-authors. Here are the top co-authors by number of shared publications:
- Chanjin Zheng
East China Normal University
6 shared publications - Xiaoyu Li
School of Education and Intelligent Education Research Center, Yangzhou University, Yangzhou 225009, China
4 shared publications Yanlei Chen
School of Educational Science, Liaocheng University, Liaocheng, China
3 shared publicationsChanjin Zheng
Lab of Artificial Intelligence for Education East China Normal University Shanghai China
2 shared publications- Qian Sun
Department of Psychology, Suzhou University of Science and Technology, Suzhou 215009, China
2 shared publications - Wei Xia
Shanghai Key Laboratory of Green Chemistry and Chemical Processes, School of Chemistry and Molecular Engineering, East China Normal University, Shanghai 200062, China
2 shared publications
Latest Publications
- Open Access
- 1 September 2026
- Published by MDPI AG in Journal of Intelligence
- Vol. 14 (9) , 197
A Theory-Driven Game-Based Assessment of Collaborative Problem-Solving
Collaborative problem-solving (CPS) is an essential 21st-century competency, yet its valid and ecologically sound assessment remains challenging due to inconsistent frameworks, poorly aligned task design, and over-reliance on behavior frequency counting in scoring. This study developed and provided empirical support for a theory-driven game-based CPS assessment embedded in a real-time strategy game, using a customized Collaborative Siege Task with asymmetric roles and ill-structured problems, to elicit more realistic collaborative interactions. Based on a validated three-facet CPS framework from Chinese cultural context, the task systematically triggers social and cognitive processes of collaboration across five progressive stages. By sampling from the eastern provinces of China, a pilot study with 36 participants confirmed task feasibility and indicator effectiveness, while a formal study with 298 valid participants examined psychometric properties using item response theory and confirmatory factor analysis. Findings supported the hierarchical higher-order factor structure, satisfactory item difficulty, discrimination, reliability, and criterion-related validity against peer assessment, reasoning ability, and personality. The assessment showed no sex, major, or role bias. This study provides a reliable gamified assessment tool for CPS measurement with demonstrated potential for scalability pending further development of automated scoring procedures, offering implications for large-scale evaluation and competency-oriented instruction.
- 1 September 2026
- Published by MDPI AG in Journal of Intelligence
- Vol. 14 (9) , 197
Open AccessA Theory-Driven Game-Based Assessment of Collaborative Problem-Solving
Collaborative problem-solving (CPS) is an essential 21st-century competency, yet its valid and ecologically sound assessment remains challenging due to inconsistent frameworks, poorly aligned task design, and over-reliance on behavior frequency counting in scoring. This study developed and provided empirical support for a theory-driven game-based CPS assessment embedded in a real-time strategy game, using a customized Collaborative Siege Task with asymmetric roles and ill-structured problems, to elicit more realistic collaborative interactions. Based on a validated three-facet CPS framework from Chinese cultural context, the task systematically triggers social and cognitive processes of collaboration across five progressive stages. By sampling from the eastern provinces of China, a pilot study with 36 participants confirmed task feasibility and indicator effectiveness, while a formal study with 298 valid participants examined psychometric properties using item response theory and confirmatory factor analysis. Findings supported the hierarchical higher-order factor structure, satisfactory item difficulty, discrimination, reliability, and criterion-related validity against peer assessment, reasoning ability, and personality. The assessment showed no sex, major, or role bias. This study provides a reliable gamified assessment tool for CPS measurement with demonstrated potential for scalability pending further development of automated scoring procedures, offering implications for large-scale evaluation and competency-oriented instruction.
- Open Access
- 7 July 2026
- Published by MDPI AG in Behavioral Sciences
- Vol. 16 (7) , 1140
A Deep Learning-Based Latent Trait Model for Forced-Choice Personality Assessment
In the era of intelligent assessment, psychometric tests are becoming increasingly important for personnel selection, career development, and mental health assessment. Forced-choice tests are common in personality assessments because they require participants to select from closely related options, lowering the risk of response distortion. However, traditional latent trait models for forced-choice tests suffer from severe computational bottlenecks in high-dimensional settings. Furthermore, existing deep learning-based cognitive diagnosis models are primarily designed for independent items in educational scenarios (predicated on absolute scoring), making them structurally maladapted to the ipsative data (relative preference comparisons) generated by forced-choice tests. To address these challenges, this study presents a deep learning-based Forced-Choice Neural Latent Trait (FCNLT) Model that overcomes the limitations of traditional models and is applicable to the three most common item block types found in forced-choice tests. To account for the unidimensionality of items, participants’ latent trait levels and item characteristics are represented as interpretable latent embeddings. FCNLT mines these features through nonlinear mapping and introduces a weighted BPR-based ranking loss to natively align with the relative-scoring nature of forced-choice data. Additionally, the monotonicity assumption is utilized to improve the interpretability of the trait estimates. The FCNLT’s effectiveness is validated by experiments on real-world and simulated datasets that show its accuracy, interpretability, and robustness.
- 7 July 2026
- Published by MDPI AG in Behavioral Sciences
- Vol. 16 (7) , 1140
Open AccessA Deep Learning-Based Latent Trait Model for Forced-Choice Personality Assessment
In the era of intelligent assessment, psychometric tests are becoming increasingly important for personnel selection, career development, and mental health assessment. Forced-choice tests are common in personality assessments because they require participants to select from closely related options, lowering the risk of response distortion. However, traditional latent trait models for forced-choice tests suffer from severe computational bottlenecks in high-dimensional settings. Furthermore, existing deep learning-based cognitive diagnosis models are primarily designed for independent items in educational scenarios (predicated on absolute scoring), making them structurally maladapted to the ipsative data (relative preference comparisons) generated by forced-choice tests. To address these challenges, this study presents a deep learning-based Forced-Choice Neural Latent Trait (FCNLT) Model that overcomes the limitations of traditional models and is applicable to the three most common item block types found in forced-choice tests. To account for the unidimensionality of items, participants’ latent trait levels and item characteristics are represented as interpretable latent embeddings. FCNLT mines these features through nonlinear mapping and introduces a weighted BPR-based ranking loss to natively align with the relative-scoring nature of forced-choice data. Additionally, the monotonicity assumption is utilized to improve the interpretability of the trait estimates. The FCNLT’s effectiveness is validated by experiments on real-world and simulated datasets that show its accuracy, interpretability, and robustness.
- Open Access
- 31 March 2026
- Published by Frontiers Media SA in Frontiers in Psychology
- Vol. 17, 1746754
The No-U-Turn sampler and its mixture-modeling revision for the 4-parameter normal ogive model
Introduction: The No-U-Turn Sampler (NUTS), widely applied in psychometrics via the Stan platform, lacks algorithm-level systematic introduction for item response theory (IRT) models and tailored optimizations for specific models. This study systematically explicates the NUTS algorithm for the 4-parameter normal ogive (4PNO) model and proposes a Mixture-Modeling NUT sampler (MMNUTS) to enhance its estimation. Methods: A simulation study that considers both educational and psychological scenarios with an empirical example from a behavioral scale was designed to examine the feasibility of the proposed MMNUTS. A recoverability comparison is performed among the NUT sampler in Stan, MMNUTS, and the Gibbs-within-Gibbs sampler in the R package fourPNO. Results: All samplers achieved satisfactory convergence, with estimation accuracy improved at larger sample sizes. MMNUTS showed significant superiority in running time and upper asymptote parameter recovery in the psychological scenario with less informative priors and 5,000 samples. The 4PNO model yielded optimal fit in the empirical analysis, with MMNUTS demonstrating faster convergence and higher computational efficiency than the other two samplers. Discussion: The proposed MMNUTS maintains robust estimation accuracy while improving computational efficiency for the 4PNO model, providing a foundational reference for developing tailored sampling algorithms for complex psychometric models. The MATLAB codes for MMNUTS are presented in the appendix.
- 31 March 2026
- Published by Frontiers Media SA in Frontiers in Psychology
- Vol. 17, 1746754
Open AccessThe No-U-Turn sampler and its mixture-modeling revision for the 4-parameter normal ogive model
Introduction: The No-U-Turn Sampler (NUTS), widely applied in psychometrics via the Stan platform, lacks algorithm-level systematic introduction for item response theory (IRT) models and tailored optimizations for specific models. This study systematically explicates the NUTS algorithm for the 4-parameter normal ogive (4PNO) model and proposes a Mixture-Modeling NUT sampler (MMNUTS) to enhance its estimation. Methods: A simulation study that considers both educational and psychological scenarios with an empirical example from a behavioral scale was designed to examine the feasibility of the proposed MMNUTS. A recoverability comparison is performed among the NUT sampler in Stan, MMNUTS, and the Gibbs-within-Gibbs sampler in the R package fourPNO. Results: All samplers achieved satisfactory convergence, with estimation accuracy improved at larger sample sizes. MMNUTS showed significant superiority in running time and upper asymptote parameter recovery in the psychological scenario with less informative priors and 5,000 samples. The 4PNO model yielded optimal fit in the empirical analysis, with MMNUTS demonstrating faster convergence and higher computational efficiency than the other two samplers. Discussion: The proposed MMNUTS maintains robust estimation accuracy while improving computational efficiency for the 4PNO model, providing a foundational reference for developing tailored sampling algorithms for complex psychometric models. The MATLAB codes for MMNUTS are presented in the appendix.
- Open Access
- 18 August 2025
- Published by MDPI AG in Behavioral Sciences
- Vol. 15 (8) , 1120
The Impact of Cyber-Ostracism on Bystanders’ Helping Behavior Among Undergraduates: The Moderating Role of Rejection Sensitivity
This study examined how undergraduate bystanders respond to cyber-ostracism events and the moderating role of rejection sensitivity in shaping helping behaviors using two experiments. In Experiment 1 (N = 276), we first measured participants’ rejection sensitivity, then manipulated cyber-ostracism using a social media interaction scenario, and finally, measured helping behavior towards the (non-)ostracism target using a questionnaire. Experiment 2 (N = 258) sought to replicate and extend the findings of Experiment 1 using a methodologically refined design, in which we employed a modified Cyberball paradigm to manipulate cyber-ostracism and measured bystanders’ helping behavior through their resource allocation decisions (i.e., token sharing). The results revealed that witnessing cyber-ostracism significantly promoted bystanders’ helping behavior. This facilitative effect was more pronounced among bystanders with high rejection sensitivity. These findings shed light on the psychological mechanisms underlying bystanders’ helping responses in the context of cyber-ostracism and provide a new perspective for understanding interpersonal interactions in digital environments.
- 18 August 2025
- Published by MDPI AG in Behavioral Sciences
- Vol. 15 (8) , 1120
Open AccessThe Impact of Cyber-Ostracism on Bystanders’ Helping Behavior Among Undergraduates: The Moderating Role of Rejection Sensitivity
This study examined how undergraduate bystanders respond to cyber-ostracism events and the moderating role of rejection sensitivity in shaping helping behaviors using two experiments. In Experiment 1 (N = 276), we first measured participants’ rejection sensitivity, then manipulated cyber-ostracism using a social media interaction scenario, and finally, measured helping behavior towards the (non-)ostracism target using a questionnaire. Experiment 2 (N = 258) sought to replicate and extend the findings of Experiment 1 using a methodologically refined design, in which we employed a modified Cyberball paradigm to manipulate cyber-ostracism and measured bystanders’ helping behavior through their resource allocation decisions (i.e., token sharing). The results revealed that witnessing cyber-ostracism significantly promoted bystanders’ helping behavior. This facilitative effect was more pronounced among bystanders with high rejection sensitivity. These findings shed light on the psychological mechanisms underlying bystanders’ helping responses in the context of cyber-ostracism and provide a new perspective for understanding interpersonal interactions in digital environments.
- 1 October 2024
- Published by Wiley in British Journal of Mathematical and Statistical Psychology
- Vol. 78 (1) , 167-189
Applying support vector machines to a diagnostic classification model for polytomous attributes in small‐sample contexts
Over several years, the evaluation of polytomous attributes in small‐sample settings has posed a challenge to the application of cognitive diagnosis models. To enhance classification precision, the support vector machine (SVM) was introduced for estimating polytomous attribution, given its proven feasibility for dichotomous cases. Two simulation studies and an empirical study assessed the impact of various factors on SVM classification performance, including training sample size, attribute structures, guessing/slipping levels, number of attributes, number of attribute levels, and number of items. The results indicated that SVM outperformed the pG‐DINA model in classification accuracy under dependent attribute structures and small sample sizes. SVM performance improved with an increased number of items but declined with higher guessing/slipping levels, more attributes, and more attribute levels. Empirical data further validated the application and advantages of SVMs.
- 1 October 2024
- Published by Wiley in British Journal of Mathematical and Statistical Psychology
- Vol. 78 (1) , 167-189
Applying support vector machines to a diagnostic classification model for polytomous attributes in small‐sample contexts
Over several years, the evaluation of polytomous attributes in small‐sample settings has posed a challenge to the application of cognitive diagnosis models. To enhance classification precision, the support vector machine (SVM) was introduced for estimating polytomous attribution, given its proven feasibility for dichotomous cases. Two simulation studies and an empirical study assessed the impact of various factors on SVM classification performance, including training sample size, attribute structures, guessing/slipping levels, number of attributes, number of attribute levels, and number of items. The results indicated that SVM outperformed the pG‐DINA model in classification accuracy under dependent attribute structures and small sample sizes. SVM performance improved with an increased number of items but declined with higher guessing/slipping levels, more attributes, and more attribute levels. Empirical data further validated the application and advantages of SVMs.
- 30 September 2024
- Published by Elsevier in Information Processing & Management
- Vol. 62 (1)
An interpretable polytomous cognitive diagnosis framework for predicting examinee performance
- 30 September 2024
- Published by Elsevier in Information Processing & Management
- Vol. 62 (1)
An interpretable polytomous cognitive diagnosis framework for predicting examinee performance
- Open Access
- 19 January 2024
- Published by Springer Nature in Humanities and Social Sciences Communications
- Vol. 11 (1) , 1-10
The relationship of real-life interpersonal difficulties and Chinese adolescents’ online deviant behavior: a U-shaped mediated moderating effect of online morality
- 19 January 2024
- Published by Springer Nature in Humanities and Social Sciences Communications
- Vol. 11 (1) , 1-10
Open AccessThe relationship of real-life interpersonal difficulties and Chinese adolescents’ online deviant behavior: a U-shaped mediated moderating effect of online morality
- Open Access
- 1 September 2023
- Published by National Council on Measurement in Education (NCME) in Chinese/english Journal of Educational Measurement and Evaluation
- Vol. 4 (3) , 4
小花狮:基于大语言模型的中文作文智能辅导系统
长期以来,中小学语文教师一直面临着作文评价负担重、效率低、提质难的问题。文章介绍了一种以大语言模型为基础,能够实现利用深层语言特征评估作文质量,给出可解释批改意见的小花狮作文智能辅导系统(简称小花狮)。通过对小花狮的整体功能设计、评估框架结构和评分算法原理进行论述,本文提供了中文语境下中小学作文智能化评估的理论、技术与工程问题的一个解决方案。小规模的入校实验表明,小花狮在语言检错、修辞手法和动作神态等方面的评估效果较好,能够基本满足中小学语文教学实践的需要。未来小花狮将基于GPT模型研发“教学评”一致性评估算法和个性化的精准评语生成算法,进一步提高主题分析、篇章结构和真情实感方面的评估效果,开展系统的人工智能社会实验。
- 1 September 2023
- Published by National Council on Measurement in Education (NCME) in Chinese/english Journal of Educational Measurement and Evaluation
- Vol. 4 (3) , 4
Open Access小花狮:基于大语言模型的中文作文智能辅导系统
长期以来,中小学语文教师一直面临着作文评价负担重、效率低、提质难的问题。文章介绍了一种以大语言模型为基础,能够实现利用深层语言特征评估作文质量,给出可解释批改意见的小花狮作文智能辅导系统(简称小花狮)。通过对小花狮的整体功能设计、评估框架结构和评分算法原理进行论述,本文提供了中文语境下中小学作文智能化评估的理论、技术与工程问题的一个解决方案。小规模的入校实验表明,小花狮在语言检错、修辞手法和动作神态等方面的评估效果较好,能够基本满足中小学语文教学实践的需要。未来小花狮将基于GPT模型研发“教学评”一致性评估算法和个性化的精准评语生成算法,进一步提高主题分析、篇章结构和真情实感方面的评估效果,开展系统的人工智能社会实验。
- Open Access
- 1 September 2023
- Published by National Council on Measurement in Education (NCME) in Chinese/english Journal of Educational Measurement and Evaluation
- Vol. 4 (3) , 3
ELion: An Intelligent Chinese Composition Tutoring System Based on Large Language Models
For a long time, Chinese language teachers in primary and secondary schools have been confronting challenges of heavy workload, low efficiency, and difficulty in improving the quality of composition evaluations. This article introduces “ELion”, an intelligent Chinese composition tutoring system based on large language models. The system utilizes deep linguistic features to evaluate the quality of compositions and provide interpretable feedback. By discussing the overall design, evaluation framework structure, and scoring algorithm principles of ELion, this paper addresses the theoretical, technical, and engineering issues of intelligent evaluation of Chinese compositions in the educational context. Small-scale experiments conducted in schools demonstrate that ELion performs well in language error detection, rhetorical techniques, and the expression of actions and emotions. It can basically meet the needs of Chinese language teaching in primary and secondary schools. In the future, ELion will further develop algorithms for ”instruction-learning-evaluation” alignment assessment, and personalized precise feedback generation, based on the GPT model. This will improve the evaluation effectiveness in topic analysis, text structure, and genuine emotional expression. Additionally, systematic field experiments for the system will be conducted to explore the application of artificial intelligence in education.
- 1 September 2023
- Published by National Council on Measurement in Education (NCME) in Chinese/english Journal of Educational Measurement and Evaluation
- Vol. 4 (3) , 3
Open AccessELion: An Intelligent Chinese Composition Tutoring System Based on Large Language Models
For a long time, Chinese language teachers in primary and secondary schools have been confronting challenges of heavy workload, low efficiency, and difficulty in improving the quality of composition evaluations. This article introduces “ELion”, an intelligent Chinese composition tutoring system based on large language models. The system utilizes deep linguistic features to evaluate the quality of compositions and provide interpretable feedback. By discussing the overall design, evaluation framework structure, and scoring algorithm principles of ELion, this paper addresses the theoretical, technical, and engineering issues of intelligent evaluation of Chinese compositions in the educational context. Small-scale experiments conducted in schools demonstrate that ELion performs well in language error detection, rhetorical techniques, and the expression of actions and emotions. It can basically meet the needs of Chinese language teaching in primary and secondary schools. In the future, ELion will further develop algorithms for ”instruction-learning-evaluation” alignment assessment, and personalized precise feedback generation, based on the GPT model. This will improve the evaluation effectiveness in topic analysis, text structure, and genuine emotional expression. Additionally, systematic field experiments for the system will be conducted to explore the application of artificial intelligence in education.
- 2 February 2023
- Published by Wiley in British Journal of Mathematical and Statistical Psychology
- Vol. 76 (3) , 585-604
Mixture‐modelling‐based Bayesian MH‐RM algorithm for the multidimensional 4PLM
Several recent works have tackled the estimation issue for the unidimensional four‐parameter logistic model (4PLM). Despite these efforts, the issue remains a challenge for the multidimensional 4PLM (M4PLM). Fu et al. (2021) proposed a Gibbs sampler for the M4PLM, but it is time‐consuming. In this paper, a mixture‐modelling‐based Bayesian MH‐RM (MM‐MH‐RM) algorithm is proposed for the M4PLM to obtain the maximum a posteriori (MAP) estimates. In a comparison of the MM‐MH‐RM algorithm to the original MH‐RM algorithm, two simulation studies and an empirical example demonstrated that the MM‐MH‐RM algorithm possessed the benefits of the mixture‐modelling approach and could produce more robust estimates with guaranteed convergence rates and fast computation. The MATLAB codes for the MM‐MH‐RM algorithm are available in the online appendix.
- 2 February 2023
- Published by Wiley in British Journal of Mathematical and Statistical Psychology
- Vol. 76 (3) , 585-604
Mixture‐modelling‐based Bayesian MH‐RM algorithm for the multidimensional 4PLM
Several recent works have tackled the estimation issue for the unidimensional four‐parameter logistic model (4PLM). Despite these efforts, the issue remains a challenge for the multidimensional 4PLM (M4PLM). Fu et al. (2021) proposed a Gibbs sampler for the M4PLM, but it is time‐consuming. In this paper, a mixture‐modelling‐based Bayesian MH‐RM (MM‐MH‐RM) algorithm is proposed for the M4PLM to obtain the maximum a posteriori (MAP) estimates. In a comparison of the MM‐MH‐RM algorithm to the original MH‐RM algorithm, two simulation studies and an empirical example demonstrated that the MM‐MH‐RM algorithm possessed the benefits of the mixture‐modelling approach and could produce more robust estimates with guaranteed convergence rates and fast computation. The MATLAB codes for the MM‐MH‐RM algorithm are available in the online appendix.