MTDRA-Net: graph-temporal neural network for dynamic evaluation of college students’ responsibility
- 1Changchun Finance College
- 2School of Life Sciences, Changchun Sci-Tech University
- 3Jilin No.8 Middle School
Abstract
Responsibility awareness is a core psychological trait in higher education, yet traditional scale assessments suffer from low stability and limited structural interpretability. This study proposes the Multi-Task Dynamic Responsibility Awareness Network (MTDRA-Net), integrating graph attention and temporal convolution networks. Using longitudinal data from 244 undergraduates, the model captures asymmetric inter-dimensional interactions and heterogeneous temporal dynamics across cognition, emotion, volition, and behavior. Results show MTDRA-Net achieves an intra-class correlation coefficient of 0.90, outperforming traditional scales by 12.5%. Confirmatory factor analysis confirms superior structural validity (CFI = 0.955, RMSEA = 0.05). This model was designed to provide a more stable and structurally consistent assessment tool for evaluating college students’ sense of responsibility while using the same measurement instrument. It also provided technical support for teachers and administrators in making continuous student development decisions.
Related
- Association between sleep quality trajectories and depression among college students: computer-simulated network analysis
- Design, implementation, and evaluation of a comprehensive educational program for microbiology laboratory personnel and graduate students in teaching hospitals
- Virtual Operating Theatre Simulation: Development and Evaluation of a Web-Based Learning Tool for Medical Students
- Cross-cultural adaptation and psychometric evaluation of the Chinese version of the Digital and Technological Self-Efficacy Scale among nursing students: a three-phase validation study.
- Spatial and Temporal Evaluation of Violent and Total Crime Following Community-Led Urban Greening
BibTeX
@article{pan2026mtdranet,
title = {{MTDRA-Net: graph-temporal neural network for dynamic evaluation of college students’ responsibility}},
author = {Jia Pan and Chang Liu and Luoyu Zhang},
journal = {Scientific Reports},
year = {2026},
month = jul,
doi = {10.1038/s41598-026-63028-7},
url = {https://doi.org/10.1038/s41598-026-63028-7},
note = {CC BY},
}
RIS
TY - JOUR
TI - MTDRA-Net: graph-temporal neural network for dynamic evaluation of college students’ responsibility
AU - Jia Pan
AU - Chang Liu
AU - Luoyu Zhang
PY - 2026
DA - 2026/07/01
JO - Scientific Reports
DO - 10.1038/s41598-026-63028-7
UR - https://doi.org/10.1038/s41598-026-63028-7
AB - Responsibility awareness is a core psychological trait in higher education, yet traditional scale assessments suffer from low stability and limited structural interpretability. This study proposes the Multi-Task Dynamic Responsibility Awareness Network (MTDRA-Net), integrating graph attention and temporal convolution networks. Using longitudinal data from 244 undergraduates, the model captures asymmetric inter-dimensional interactions and heterogeneous temporal dynamics across cognition, emotion, volition, and behavior. Results show MTDRA-Net achieves an intra-class correlation coefficient of 0.90, outperforming traditional scales by 12.5%. Confirmatory factor analysis confirms superior structural validity (CFI = 0.955, RMSEA = 0.05). This model was designed to provide a more stable and structurally consistent assessment tool for evaluating college students’ sense of responsibility while using the same measurement instrument. It also provided technical support for teachers and administrators in making continuous student development decisions.
SN - 2045-2322
ER -