Exploring AI's role in enhancing research accessibility and equity for researchers with Specific Learning Disabilities

Artificial Intelligence (AI) is transforming the world and re-shaping multiple occupational sectors, such as healthcare, education, and more. One area where AI is having a substantial impact is scientific research, through presenting entirely new opportunities for sourcing ideas and information, and transforming the skillsets required for producing written academic outputs. Within this, an additional advantage of AI is its potential to create a more accessible and equitable research environment, particularly individuals with ‘Specific Learning Difficulties’ (SpLD), an umbrella term referring to challenges in areas such as reading, writing, and numeracy that include dyslexia, dysgraphia, dyspraxia, and dyscalculia. For whom academic research can pose barriers can include reading comprehension and writing skills, and access to scholarly literature, which may lead to less publications and slower dissemination of research outputs. This may impact career progression and funding opportunities. Individuals with learning disabilities also have a higher risk of burnout due to (i) the barriers stated above, and increased effort required to overcome these barriers and (ii) poorer time management skills and high workloads. This proposal is inspired by my personal experiences with overcoming obstacles and institutional biases related to my learning disabilities. Nonetheless, if implemented correctly, AI Large Language Models (LLM) could be implemented across academic institutions to reduce the inequalities in research output formulation affecting those with SpLD, and help to reduce many of the challenges in this area whilst better managing their workload. Despite the availability of AI technologies and increased use of LLMs worldwide, there remains a gap in knowledge around use of AI and systematically implementing these tools within research institutions to ensure accessibility and usability. This proposal aims to bridge that gap by exploring AI’s role in making research more inclusive and by advocating for policy changes to support AI adoption in academic environments. This project employs a comprehensive mixed methods design to investigate how AI can enhance research accessibility for those with specific learning difficulties. A systematic review will synthesise current literature on AI’s application and impact in academic research, establishing a solid foundation for subsequent studies. An online survey of 300 researchers with SpLD will gather quantitative data on the tools used, benefits realised, and challenges encountered across various research stages, with accessible formats to ensure broad participation. Complementing the survey, semi-structured interviews with both researchers with SpLD and institutional leads will explore in depth the experiences of using AI and uncover barriers and facilitators to its adoption in research. Purposive and snowball sampling will ensure a diverse range of views, enhancing the representativeness of the findings. Experimental case studies employing a within-subject design will compare AI-enabled versus traditional methods across tasks that mirror everyday research activities. These studies will measure improvements in efficiency, accuracy, and workload management through both descriptive statistics and reflexive thematic analysis. Together, these methods will yield robust qualitative and quantitative evidence on how AI can transform research practices for individuals with SpLD. The outputs will include a systematic review, several research papers, and a comprehensive report detailing policy recommendations. Dissemination will be achieved through academic publications and presentations to key stakeholders, including the UK Metascience unit and university equity bodies. Ultimately, the project aims to drive policy change and foster a more inclusive research environment, ensuring that AI adoption benefits researchers across all abilities.