My name is Victoria Yaneva and I am a final-year PhD student at the Research Institute of Information and Language Processing (RIILP) at the University of Wolverhampton. The aim of my research is to develop technology that helps people with autism to read and comprehend text better. As many autistic people are known to experience specific reading comprehension difficulties, helping them read better would lead to reducing school dropout, enhancing their academic achievements, employability and social inclusion.
The first step to achieving this is to provide content for developers who want to make their textbooks or websites accessible to people with autism (and thus for every other user) with a tool which is easy to use and at the same time reflects the difficulties autistic readers face. We achieve this through a series of experiments with autistic adults conducted over the course of 3 years, where we measure their comprehension of different texts and record their eye-movements whilst reading. Based on this data we are in the process of developing an automatic accessibility evaluation tool called AUTOR, which can measure the accessibility of texts aimed at people with 3 different levels of autism severity. AUTOR will also be able to pinpoint particular areas of difficulty in the text based on the data collected from eye-tracking.
AUTOR has writing and reading modes. The writing mode could benefit non-experts who want to develop accessible content but are not familiar with the specific reading difficulties autism entails. It could also help experts who know how to write accessible content but need to have a quick and reliable to way to both evaluate it, and illustrate any complex words with images. People with autism could also use the tool in a reading mode, which would help them select material at their preferred reading level and personalise it by including definitions, images or changing the way the text is presented.
To find out more about my work please feel free to contact me at firstname.lastname@example.org.