Toward Inclusive AI in Mathematics Education: Investigating Usability and Accessibility Challenges for Students with Visual Disabilities

Amjad Ali; Barkat Ullah

Authors

Amjad Ali Department of Computer Science, University of Shangla, KP, Pakistan
Barkat Ullah Department of Computer Science, University of Peshawar, KP, Pakistan

Keywords:

Accessibility in AI, Students with Visual Disabilities, Mathematics Education, Assistive Technology, AI in Education

Abstract

The rapid adoption of large language models (LLMs) in education presents new opportunities for personalized learning; however, their accessibility for visually impaired students in mathematics learning remains insufficiently explored. This study investigates the accessibility, usability, and instructional effectiveness of four widely used LLM platforms, ChatGPT, Gemini, LLaMA, and DeepSeek, when supporting visually impaired learners in solving mathematical problems. A mixed-methods research design was employed, involving 12 students with visual disabilities and 3 instructors. Data were collected through structured surveys, task-based mathematical problem solving, and instructor-based evaluation of explanation quality. Quantitative analysis included descriptive statistics and comparative accuracy testing, while qualitative analysis examined user experiences and accessibility barriers. The results indicate that 66% of participants experienced incompatibility between AI-generated mathematical notation and screen reader technologies, while 83% reported the absence of adaptive verbosity controls for explanations. Usability analysis revealed that 100% of participants encountered difficulties navigating multi-step solutions due to insufficient structural organization. Comparative evaluation showed that ChatGPT and Gemini achieved higher average solution accuracy (~80%), whereas LLaMA and DeepSeek demonstrated lower accuracy (60%). Instructor assessments further indicated that 67% of AI-generated explanations lacked clear stepwise reasoning, requiring additional clarification for learners. Participants emphasized the need for structured, step-by-step explanations, screen reader–compatible mathematical notation, adaptive verbosity controls, and audio-guided navigation to enhance accessibility. The study highlights that while LLMs demonstrate moderate computational capability, significant accessibility and usability limitations remain for visually impaired learners. The findings contribute to the development of accessibility-aware AI frameworks for inclusive STEM education, providing design recommendations for improving LLM-based mathematics learning tools.

References

Simone Grassini, “Shaping the Future of Education: Exploring the Potential and Consequences of AI and ChatGPT in Educational Settings,” Educ. Sci., vol. 13, no. 7, p. 692, 2023, doi: https://doi.org/10.3390/educsci13070692.

“AI in Education Systems: Successful Cases and Perspectives - 1st Editi.” Accessed: Mar. 09, 2026. [Online]. Available: https://www.routledge.com/AI-in-Education-Systems-Successful-Cases-and-Perspectives/Slyusar Kondratenko/p/book/

Kiranpreet Kaur, Shilpa, Ridhi Jindal, “The ChatGPT revolution: Reshaping education through artificial intelligence Available to Purchase,” AIP Conf. Proc, 2024, doi: https://doi.org/10.1063/5.0221430.

Carmen del Rosario Navas Bonilla, Luis Miguel Viñan Carrasco, “The Future of Education: A Systematic Literature Review of Self-Directed Learning with AI,” Futur. Internet, vol. 17, no. 8, p. 366, 2025, doi: https://doi.org/10.3390/fi17080366.

Amjad Ali, Shah Khusro, “Accessible interactive learning of mathematical expressions for school students with visual disabilities,” PeerJ Comput. Sci., 2024, [Online]. Available: https://peerj.com/articles/cs-2599/

O. G. Klingenberg, A. H. Holkesvik, and L. B. Augestad, “Digital learning in mathematics for students with severe visual impairment: A systematic review,” Br. J. Vis. Impair., vol. 38, no. 1, pp. 38–57, Jan. 2020, doi: 10.1177/0264619619876975.

Elena Filipova, “Enhancing Educational Accessibility for Visually Impaired Individuals through Artificial Intelligence: Challenges and Opportunities,” Int. J. Comput. Appl., vol. 186, no. 67, 2025, [Online]. Available: https://www.ijcaonline.org/archives/volume186/number67/enhancing-educational-accessibility-for-visually-impaired-individuals-through-artificial-intelligence-challenges-and-opportunities/

Amjad Ali, Shah Khusro, “AIME-solve: Accessible interactive mathematical expressions solver for optimized learning and visual empowerment,” SoftwareX, vol. 30, p. 102143, 2025, doi: https://doi.org/10.1016/j.softx.2025.102143.

Amjad Ali, Shah Khusro, “SA-MEAS: Sympy-based automated mathematical equations analysis and solver,” SoftwareX, vol. 25, p. 101596, 2024, doi: https://doi.org/10.1016/j.softx.2023.101596.

Chokri Kooli, Rim Chakraoui, “AI-driven assistive technologies in inclusive education: benefits, challenges, and policy recommendations,” Sustain. Futur., vol. 10, p. 101042, 2025, doi: https://doi.org/10.1016/j.sftr.2025.101042.

Evdokia Voultsiou & Lefteris Moussiades, “A systematic review of AI, VR, and LLM applications in special education: Opportunities, challenges, and future directions,” Educ. Inf. Technol., vol. 30, pp. 19141–19181, 2025, [Online]. Available: https://link.springer.com/article/10.1007/s10639-025-13550-4

Inigo Lopez-Gazpio, “Integrating large language models into accessible and inclusive education: access democratization and individualized learning enhancement supported by generative artificial intelligence,” Information, vol. 16, no. 6, p. 473, 2025, doi: https://doi.org/10.3390/info16060473.

Janice Ahn, Rishu Verma, Renze Lou, Di Liu, Rui Zhang, Wenpeng Yin, “Large language models for mathematical reasoning: Progresses and challenges,” arXiv:2402.00157, 2024, [Online]. Available: https://arxiv.org/abs/2402.00157

M. dos S. Soares, C. A. Furukawa, M. I. Cagnin, and D. M. B. Paiva, “Accessible learning objects: a systematic literature review,” Univers. Access Inf. Soc. 2023 234, vol. 23, no. 4, pp. 1931–1945, Aug. 2023, doi: 10.1007/s10209-023-01025-7.

Dalal Al Faia, Khalid Alomar, “Exploring Adaptive, Adaptable, Mixed, and Static Arabic Rule-Based Chatbot Effects on Usability, Learning Success, and Engagement,” Appl. Sci., vol. 15, no. 24, p. 13266, 2025, doi: https://doi.org/10.3390/app152413266.

R. A. Khalil, Z. Safelnasr, N. Yemane, M. Kedir, A. Shafiqurrahman and N. SAEED, “Advanced Learning Technologies for Intelligent Transportation Systems: Prospects and Challenges,” IEEE Open J. Veh. Technol., vol. 5, pp. 397–427, 2024, doi: 10.1109/OJVT.2024.3369691.

Adekunle, Adetunji Noah, “United Nations sustainable development goal 4 : exploring inclusive education of students with developmental disabilities in Lagos State, Nigeria,” Open Collect., 2021, [Online]. Available: https://open.library.ubc.ca/soa/cIRcle/collections/ubctheses/24/items/1.0406184

Jinke Jiang, “Implementing voice assistant for visually impaired using LLMs and Vision Language Models,” Univ. Victoria, 2024, [Online]. Available: https://dspace.library.uvic.ca/items/7b594ca7-d12c-4ed3-8cf2-be432b368240

“Enhancing the Education of Accessibility with Artificial Intelligence.” Accessed: Mar. 09, 2026. [Online]. Available: https://erepo.uef.fi/items/1a74799f-6625-4ba8-ae4c-7309a64682ba

Ibrahim A. Elshaer, Sameer M. AlNajdi, “Measuring the impact of large language models on academic success and quality of life among students with visual disability: An assistive technology perspective,” Bioengineering, vol. 12, no. 10, p. 1056, 2025, doi: https://doi.org/10.3390/bioengineering12101056.

“Using Technology | American Foundation for the Blind.” Accessed: Mar. 09, 2026. [Online]. Available: https://afb.org/blindness-and-low-vision/using-technology

Andrea R. Harkins-Brown, Linda Z. Carling, “Artificial Intelligence in Special Education,” Encyclopedia, vol. 5, no. 1, p. 11, 2025, doi: 10.3390/encyclopedia5010011.

Y. A. Jarbi, “Influence of Assistive Technologies on Visually Impaired Students’ Experiences of Social Inclusion in Higher Education in Qatar,” Jun. 2024, doi: 10.25392/LEICESTER.DATA.26068081.V1.

A. Saha, “Towards Accessible Skilled Work: Understanding Practices and Augmenting Instructional Support among Blind Audio Producers,” Libr. Arch, Jan. 2024, doi: 10.21985/N2-7HGQ-9570.

Nosheen Fayyaz, Shah Khusro, “Accessibility of Tables in PDF Documents Issues, Challenges and Future Directions,” Inf. Technol. Libr., vol. 40, no. 3, 2021, [Online]. Available: https://ital.corejournals.org/index.php/ital/article/view/12325

R. Milallos, T. Oh, and R. L. Peiris, “Exploring the Effectiveness of Electrotactile Feedback for Data Visualization for Blind and Visually Impaired Users,” Lect. Notes Comput. Sci. (including Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinformatics), vol. 14698 LNCS, pp. 413–428, 2024, doi: 10.1007/978-3-031-60884-1_28.

Dominik Spinczyk, Michał Maćkowski, Wojciech Kempa, Katarzyna Rojewska, “Factors influencing the process of learning mathematics among visually impaired and blind people,” Comput. Biol. Med., vol. 104, pp. 1–9, 2019, doi: https://doi.org/10.1016/j.compbiomed.2018.10.025.

J. Matoušek et al., “Speech and web-based technology to enhance education for pupils with visual impairment,” J. Multimodal User Interfaces 2020 142, vol. 14, no. 2, pp. 219–230, Apr. 2020, doi: 10.1007/s12193-020-00323-1.

T. S. Herzberg, M. Tuttle, and L. P. Rosenblum, “Digital Mathematics Experiences and Perspectives of High School and College Students Who Are Blind or Have Low Vision,” J. Vis. Impair. Blind., Jan. 2026, doi: 10.1177/0145482X251404037.

Barbara Mirel, “Interaction Design for Complex Problem Solving Developing Useful and Usable Software,” Interact. Des. Complex Probl. Solving Dev. Useful Usable Softw., 2003, [Online]. Available: https://www.sciencedirect.com/book/monograph/

/interaction-design-for-complex-problem-solving

Taiwo Raphael Feyijimi, John Ogbeleakhu Aliu, “ChatGPT’s Expanding Horizons and Transformative Impact Across Domains: A Critical Review of Capabilities, Challenges, and Future Directions,” Computers, vol. 14, no. 9, p. 366, 2025, doi: https://doi.org/10.3390/computers14090366.

Rafael Martínez-Peláez, Luis J. Mena, “Can Large Language Models Foster Critical Thinking, Teamwork, and Problem-Solving Skills in Higher Education?: A Literature Review,” Systems, vol. 13, no. 11, p. 1013, 2025, doi: https://doi.org/10.3390/systems13111013.

R. Van Campenhout, D. Soto-Karlin, M. Selinger, and B. Jerome, “Learning Engineering in Practice: A Case Study on Developing LLM-Based Educational Tools,” Lect. Notes Comput. Sci., vol. 15813 LNCS, pp. 132–150, 2025, doi: 10.1007/978-3-031-92970-0_10.

Madhav A. Kankhar, C. Namrata Mahender, “A Comprehensive Study of Tactile Education System for Visual Impaired People,” Proc. Int. Conf. Recent Adv. Mod. Sustain. Intell. Technol. Appl., 2025, [Online]. Available: https://www.atlantis-press.com/proceedings/ramsita-25/126011501

Tetiana Yermakova, Iryna Bondarenko, “Inclusive education and interdisciplinary approaches to the environment, pedagogical support, and social development: a systematic review,” Pedagog. Phys. Cult. Sport., vol. 29, no. 5, 2025.

Y. Wang, L. Wang, and K. L. Siau, “Human-Centered Interaction in Virtual Worlds: A New Era of Generative Artificial Intelligence and Metaverse,” Int. J. Hum. Comput. Interact., vol. 41, no. 2, pp. 1459–1501, 2025, doi: 10.1080/10447318.2024.2316376.

“Article - Accommodating Students with Disabilities in Science, Technology, Engineering, and Mathematics (STEM) - US - ADCET.” Accessed: Mar. 09, 2026. [Online]. Available: https://www.adcet.edu.au/resource/8209/accommodating-students-with-disabilities-in-science-technology-engineering-and-mathematics-stem-us

Oyebola Olusola Ayeni, Chika Chioma Unachukwu, “Innovations in STEM education for students with disabilities: A critical examination,” Int. J. Sci. Res. Arch., vol. 11, no. 1, pp. 1797–1809, 2024, doi: 10.30574/ijsra.2024.11.1.0285.

Sahil Sharma, Puneet Mittal, Mukesh Kumar & Vivek Bhardwaj, “The role of large language models in personalized learning: a systematic review of educational impact,” Discov. Sustain., vol. 6, no. 243, 2025, [Online]. Available: https://link.springer.com/article/10.1007/s43621-025-01094-z

Gatot Hery Djatmiko, Obsatar Sinaga, “Digital Transformation and Social Inclusion in Public Services: A Qualitative Analysis of E-Government Adoption for Marginalized Communities in Sustainable Governance,” Sustainability, vol. 17, no. 7, p. 2908, 2025, doi: https://doi.org/10.3390/su17072908.

Shalece Kohnke, Tiffanie Zaugg, “Artificial Intelligence: An Untapped Opportunity for Equity and Access in STEM Education,” Educ Sci, vol. 15, no. 1, p. 68, 2025, doi: https://doi.org/10.3390/educsci15010068.

Amjad Ali, Shah Khusro & Shabbab Ali Algamdi, “Accessible interactive learning of missing-digit arithmetic problems for students with visual disabilities,” Sci. Rep., 2025, [Online]. Available: https://www.nature.com/articles/s41598-025-01473-6

Aray M. Kassenkhan, Mateus Mendes, “An Adaptive Task Difficulty Model for Personalized Reading Comprehension in AI-Based Learning Systems,” Algorithms, vol. 19, no. 2, p. 100, 2026, doi: https://doi.org/10.3390/a19020100.

H. Jiang and H. Bian, “Development and Initial Validation of a Blended Mindfulness-Based Stress Reduction Curriculum for Junior High School Teachers,” Mindfulness 2026, pp. 1–16, Feb. 2026, doi: 10.1007/s12671-025-02756-9.