{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,31]],"date-time":"2026-10-31T03:59:52Z","timestamp":1793419192830,"version":"4.5.0"},"reference-count":46,"publisher":"MDPI AG","issue":"3","license":[{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"crossref","award":["62373387"],"award-info":[{"award-number":["62373387"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]},{"award":["62373387"],"award-info":[{"award-number":["62373387"]}],"id":[{"id":"https:\/\/ror.org\/01h0zpd94","id-type":"ROR","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"Shenzhen Fundamental Research Program","doi-asserted-by":"publisher","award":["JCYJ20240813151301003"],"award-info":[{"award-number":["JCYJ20240813151301003"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JEMR"],"abstract":"<jats:p>Under the Active Vision hypothesis, eye movements are not passive responses to visual stimuli but are actively guided by task demands and internal goals. In driving, scanpaths may therefore reflect an ongoing process of information sampling for risk assessment. However, current computational models often isolate scanpath prediction from risk assessment, overlooking their intrinsic cognitive coupling. In this study, we investigate whether driver scanpaths and traffic risk perception can be jointly modeled within a unified framework. We propose a computational approach based on the introduced Adversarial Inverse Reinforcement Learning (AIRL), where gaze behavior is interpreted as a policy that maximizes a latent safety-related reward. By employing a generator to simulate human-like sequences of fixations and saccades, and a discriminator to approximate the internal reward signal, our framework ensures that generated scanpaths synergistically inform downstream risk perception. To facilitate this research, we constructed the BDDA dataset, aggregating over 13,000 spatio-temporal gaze points with explicit risk annotations to study this joint mechanism. Experimental results indicate that simultaneously modeling the \u201cwhere\u201d (scanpath dynamics) and the \u201cwhy\u201d (risk perception) significantly outperforms the compared baseline methods on the proposed BDDA dataset. These findings provide computational evidence for a functional coupling between visual attention and risk perception, supporting the view that eye movements serve as an active mechanism for acquiring task-relevant information in safety-critical environments.<\/jats:p>","DOI":"10.3390\/jemr19030059","type":"journal-article","created":{"date-parts":[[2026,6,3]],"date-time":"2026-06-03T02:37:26Z","timestamp":1780454246000},"page":"59","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":2,"title":["Active Vision in Driving: Joint Modeling of Scanpaths and Risk Perception"],"prefix":"10.3390","volume":"19","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4128-886X","authenticated-orcid":false,"given":"Chao","family":"Gou","sequence":"first","affiliation":[{"name":"School of Intelligent Systems Engineering, Sun Yat-sen University, Shenzhen 518107, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-3505-5948","authenticated-orcid":false,"given":"Yueyao","family":"Lin","sequence":"additional","affiliation":[{"name":"School of Intelligent Systems Engineering, Sun Yat-sen University, Shenzhen 518107, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2470-0293","authenticated-orcid":false,"given":"Yuchen","family":"Zhou","sequence":"additional","affiliation":[{"name":"School of Intelligent Systems Engineering, Sun Yat-sen University, Shenzhen 518107, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-4297-9562","authenticated-orcid":false,"given":"Wenjie","family":"Shi","sequence":"additional","affiliation":[{"name":"School of Intelligent Systems Engineering, Sun Yat-sen University, Shenzhen 518107, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5522-6910","authenticated-orcid":false,"given":"Jincheng","family":"Jiang","sequence":"additional","affiliation":[{"name":"School of Intelligent Systems Engineering, Sun Yat-sen University, Shenzhen 518107, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,6,1]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Ni, J., Shao, Y., Guo, Y., and Gu, Y. (2025). Interpretable Quantification of Scene-Induced Driver Visual Load: Linking Eye-Tracking Behavior to Road Scene Features via SHAP Analysis. J. Eye Mov. Res., 18.","DOI":"10.3390\/jemr18050040"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"125975","DOI":"10.1016\/j.eswa.2024.125975","article-title":"Driver Cognitive Distraction Detection based on eye movement behavior and integration of multi-view space-channel feature","volume":"266","author":"Qiao","year":"2025","journal-title":"Expert Syst. Appl."},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Kasahara, I., Stent, S., and Park, H.S. (2022). Look both ways: Self-supervising driver gaze estimation and road scene saliency. Proceedings of the ECCV, Springer.","DOI":"10.1007\/978-3-031-19778-9_8"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"108117","DOI":"10.1016\/j.engappai.2024.108117","article-title":"A review of driver gaze estimation and application in gaze behavior understanding","volume":"133","author":"Sharma","year":"2024","journal-title":"Eng. Appl. Artif. Intell."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Zhou, Y., Tang, J., Xiao, X., Lin, Y., Liu, L., Guo, Z., Fei, H., Xia, X., and Gou, C. (2025, January 19\u201320). Where, What, Why: Towards Explainable Driver Attention Prediction. Proceedings of the 2025 IEEE\/CVF International Conference on Computer Vision Workshops (ICCVW), Honolulu, HI, USA.","DOI":"10.1109\/ICCV51701.2025.00257"},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Li, M., Zhu, J., Huang, Z., and Gou, C. (2024). Imitating the human visual system for scanpath predicting. Proceedings of the ICASSP 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE.","DOI":"10.1109\/ICASSP48485.2024.10447354"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Jiang, Y., Guo, Z., Rezazadegan Tavakoli, H., Leiva, L.A., and Oulasvirta, A. (2024). EyeFormer: Predicting personalized scanpaths with transformer-guided reinforcement learning. Proceedings of the 37th Annual ACM Symposium on User Interface Software and Technology, Association for Computing Machinery.","DOI":"10.1145\/3654777.3676436"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Wang, B., Tan, D., Kuo, Y.L., Sun, Z., Wolfe, J.M., Cham, T.J., and Zhang, M. (2025, January 11\u201315). Gazing at Rewards: Eye Movements as a Lens into Human and AI Decision-Making in Hybrid Visual Foraging. Proceedings of the Computer Vision and Pattern Recognition Conference, Nashville, TN, USA.","DOI":"10.1109\/CVPR52734.2025.01380"},{"key":"ref_9","first-page":"16923","article-title":"Risk assessment methodologies for autonomous driving: A survey","volume":"23","author":"Chia","year":"2022","journal-title":"IEEE T-ITS"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Qiu, Y., Busso, C., Misu, T., and Akash, K. (2022). Incorporating gaze behavior using joint embedding with scene context for driver takeover detection. Proceedings of the ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE.","DOI":"10.1109\/ICASSP43922.2022.9747779"},{"key":"ref_11","first-page":"20120","article-title":"A Driving Risk Assessment Framework Considering Driver\u2019s Fatigue State and Distraction Behavior","volume":"25","author":"Chen","year":"2024","journal-title":"IEEE T-ITS"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Yang, Z., Mondal, S., Ahn, S., Xue, R., Zelinsky, G., Hoai, M., and Samaras, D. (2024). Unifying top-down and bottom-up scanpath prediction using transformers. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, IEEE.","DOI":"10.1109\/CVPR52733.2024.00166"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"2983","DOI":"10.1109\/TPAMI.2019.2920636","article-title":"Gravitational laws of focus of attention","volume":"42","author":"Zanca","year":"2019","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"152","DOI":"10.1016\/j.visres.2014.12.026","article-title":"Saccadic model of eye movements for free-viewing condition","volume":"116","author":"Liu","year":"2015","journal-title":"Vis. Res."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"\u0130leri, R., Alt\u0131ntop, \u00c7.G., Latifo\u011flu, F., and Demirci, E. (2025). DyslexiaNet: Examining the Viability and Efficacy of Eye Movement-Based Deep Learning for Dyslexia Detection. J. Eye Mov. Res., 18.","DOI":"10.3390\/jemr18050056"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Takinami, S., Morita, Y., Seita, J., and Oshika, T. (2026). An Open-Source Horizontal Strabismus Simulator as an Evaluation Platform for Monocular Gaze Estimation Using Deep Learning Models. J. Eye Mov. Res., 19.","DOI":"10.3390\/jemr19010020"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"2101","DOI":"10.1109\/TPAMI.2019.2956930","article-title":"Visual scanpath prediction using IOR-ROI recurrent mixture density network","volume":"43","author":"Sun","year":"2019","journal-title":"TPAMI"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Chen, X., Jiang, M., and Zhao, Q. (2021). Predicting human scanpaths in visual question answering. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, IEEE.","DOI":"10.1109\/CVPR46437.2021.01073"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"7","DOI":"10.1167\/jov.22.5.7","article-title":"DeepGaze III: Modeling free-viewing human scanpaths with deep learning","volume":"22","author":"Bethge","year":"2022","journal-title":"J. Vis."},{"key":"ref_20","unstructured":"Schwinn, L., Precup, D., Eskofier, B., and Zanca, D. (2022). Behind the machine\u2019s gaze: Neural networks with biologically-inspired constraints exhibit human-like visual attention. arXiv."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Mondal, S., Yang, Z., Ahn, S., Samaras, D., Zelinsky, G., and Hoai, M. (2023). Gazeformer: Scalable, effective and fast prediction of goal-directed human attention. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, IEEE.","DOI":"10.1109\/CVPR52729.2023.00145"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Xue, R., Xu, J., Mondal, S., Le, H., Zelinsky, G., Hoai, M., and Samaras, D. (2025). Few-shot Personalized Scanpath Prediction. Proceedings of the Computer Vision and Pattern Recognition Conference, IEEE.","DOI":"10.1109\/CVPR52734.2025.01260"},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Yang, Z., Huang, L., Chen, Y., Wei, Z., Ahn, S., Zelinsky, G., Samaras, D., and Hoai, M. (2020). Predicting goal-directed human attention using inverse reinforcement learning. Proceedings of the Computer Vision and Pattern Recognition Conference, IEEE.","DOI":"10.1109\/CVPR42600.2020.00027"},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Huang, Z., Zhou, Y., Zhu, J., and Gou, C. (2024). Driver scanpath prediction based on inverse reinforcement learning. Proceedings of the ICASSP 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE.","DOI":"10.1109\/ICASSP48485.2024.10446034"},{"key":"ref_25","unstructured":"Kara, O., Nisar, H., and Rehg, J.M. (2025). DiffEye: Diffusion-Based Continuous Eye-Tracking Data Generation Conditioned on Natural Images. arXiv."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Cartella, G., Cuculo, V., D\u2019Amelio, A., Cornia, M., Boccignone, G., and Cucchiara, R. (2025). Modeling Human Gaze Behavior with Diffusion Models for Unified Scanpath Prediction. Proceedings of the IEEE\/CVF International Conference on Computer Vision, IEEE.","DOI":"10.1109\/ICCV51701.2025.01504"},{"key":"ref_27","unstructured":"Kang, J., Groth, C., Wu, T., Torrens, F., Sangkloy, P., Wetzstein, G., and Sun, Q. (2026). Infinite Gaze Generation for Videos with Autoregressive Diffusion. arXiv."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Beckers, K., Heisel, M., Frese, T., and Hatebur, D. (2013). A structured and model-based hazard analysis and risk assessment method for automotive systems. Proceedings of the 2013 IEEE 24th International Symposium on Software Reliability Engineering (ISSRE), IEEE.","DOI":"10.1109\/ISSRE.2013.6698923"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"188","DOI":"10.1016\/j.trc.2018.02.003","article-title":"Infrastructure-cooperative algorithm for effective intersection collision avoidance","volume":"89","author":"Fu","year":"2018","journal-title":"Transp. Res. Part C Emerg. Technol."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"61","DOI":"10.1016\/j.aap.2019.01.029","article-title":"A new integrated collision risk assessment methodology for autonomous vehicles","volume":"127","author":"Katrakazas","year":"2019","journal-title":"Accid. Anal. Prev."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"9379","DOI":"10.1109\/JIOT.2022.3141044","article-title":"Spatiotemporal scene-graph embedding for autonomous vehicle collision prediction","volume":"9","author":"Malawade","year":"2022","journal-title":"IEEE Internet Things J."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"2249","DOI":"10.1109\/TIV.2023.3257169","article-title":"Gsc: A graph and spatio-temporal continuity based framework for accident anticipation","volume":"9","author":"Wang","year":"2023","journal-title":"IEEE Trans. Intell. Veh."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"110071","DOI":"10.1016\/j.patcog.2023.110071","article-title":"Dynamic attention augmented graph network for video accident anticipation","volume":"147","author":"Song","year":"2024","journal-title":"Pattern Recognit."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"104478","DOI":"10.1016\/j.tre.2025.104478","article-title":"RiskNet: Interaction-aware risk forecasting for autonomous driving in long-tail scenarios","volume":"205","author":"Liu","year":"2026","journal-title":"Transp. Res. Part E Logist. Transp. Rev."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"108041","DOI":"10.1016\/j.aap.2025.108041","article-title":"Collision risk prediction and takeover requirements assessment based on radar-video integrated sensors data: A system framework based on LLM","volume":"218","author":"Liu","year":"2025","journal-title":"Accid. Anal. Prev."},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Huang, Z., Zhou, Y., and Gou, C. (2024). Task-Oriented Scanpath Prediction with Spatial-Temporal Information in Driving Scenarios. Proceedings of the Chinese Conference on Pattern Recognition and Computer Vision (PRCV), Springer.","DOI":"10.1007\/978-981-97-8792-0_13"},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"8268","DOI":"10.1007\/s11227-021-04151-2","article-title":"Driver attention prediction based on convolution and transformers","volume":"78","author":"Gou","year":"2022","journal-title":"J. Supercomput."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"692","DOI":"10.3758\/BRM.42.3.692","article-title":"ScanMatch: A novel method for comparing fixation sequences","volume":"42","author":"Cristino","year":"2010","journal-title":"Behav. Res. Methods"},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"1079","DOI":"10.3758\/s13428-012-0212-2","article-title":"It depends on how you look at it: Scanpath comparison in multiple dimensions with MultiMatch, a vector-based approach","volume":"44","author":"Dewhurst","year":"2012","journal-title":"Behav. Res. Methods"},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Assens, M., Giro-i Nieto, X., McGuinness, K., and O\u2019Connor, N.E. (2018). PathGAN: Visual scanpath prediction with generative adversarial networks. Proceedings of the ECCV Workshops, Springer.","DOI":"10.1007\/978-3-030-11021-5_25"},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Sui, X., Fang, Y., Zhu, H., Wang, S., and Wang, Z. (2023). Scandmm: A deep markov model of scanpath prediction for 360deg images. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, IEEE.","DOI":"10.1109\/CVPR52729.2023.00675"},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Bao, W., Yu, Q., and Kong, Y. (2021). Drive: Deep reinforced accident anticipation with visual explanation. Proceedings of the IEEE\/CVF International Conference on Computer Vision, IEEE.","DOI":"10.1109\/ICCV48922.2021.00752"},{"key":"ref_43","first-page":"9590","article-title":"A dynamic spatial-temporal attention network for early anticipation of traffic accidents","volume":"23","author":"Karim","year":"2022","journal-title":"IEEE T-ITS"},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"744","DOI":"10.1016\/j.ins.2023.03.075","article-title":"THAT-Net: Two-layer hidden state aggregation based two-stream network for traffic accident prediction","volume":"634","author":"Liu","year":"2023","journal-title":"Inf. Sci."},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"1531","DOI":"10.3758\/s13423-022-02117-w","article-title":"Peripheral vision in real-world tasks: A systematic review","volume":"29","author":"Vater","year":"2022","journal-title":"Psychon. Bull. Rev."},{"key":"ref_46","doi-asserted-by":"crossref","unstructured":"Chen, X., Jiang, M., and Zhao, Q. (2024). Beyond average: Individualized visual scanpath prediction. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, IEEE.","DOI":"10.1109\/CVPR52733.2024.02402"}],"container-title":["Journal of Eye Movement Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1995-8692\/19\/3\/59\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,4]],"date-time":"2026-06-04T04:40:33Z","timestamp":1780548033000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1995-8692\/19\/3\/59"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,1]]},"references-count":46,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2026,6]]}},"alternative-id":["jemr19030059"],"URL":"https:\/\/doi.org\/10.3390\/jemr19030059","relation":{},"ISSN":["1995-8692"],"issn-type":[{"value":"1995-8692","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,1]]}}}