Dyslexia can be defined simply as “ the difficulty in acquiring reading, spelling and writing skills.” ⁽¹⁾  The video ⁽²⁾ below provides a short introduction to dyslexia. 

As noted by Athanasaki et al ⁽³⁾ “ Poor reading skills have an impact on the vocabulary development of dyslexic readers and to their exposure to relevant background knowledge. This leads to repeated failure in the classroom and has a negative effect on their motivation and self esteem.”

Assistive technology has been available for some time which allows dyslexic children to learn the same material as their peers, in the inclusive classroom setting.  http://www.readingrockets.org/article/assistive-technology-kids-learning-disabilities-overview contains a fairly comprehensive list of the various types of assistive technologies available. 

Moving beyond assistive technology, and by way of an example of how technology can be used to afford both a differentiated and adaptive learning experience I’d like to focus on the AGENT-DYSL project.

The AGENT-DYSL software development project “brings together speech and image recognition as well as semantic technologies to build a truly adaptive reading support system for children with dyslexia.” ⁽⁴⁾. It offers personalised learning opportunities via “customized presentation of reading material, based on individual profiles built up through ‘observation’ of each child... These individual profiles are used in deciding how to optimise the text presentation for each child and situation.

In addition to being able to 'listen to children reading', the AGENT-DYSL system is also able to 'see' the children while they read….and dynamically adapt the document presentation accordingly (a tired child or a child under emotional stress is more likely to have reduced reading performance).” ⁽⁵⁾

In more detail the software works as follows:

1. A learner profile is developed via the learner reading some text aloud and this being recorded by both a microphone and camera. This data is then analysed via speech recognition and face analysis services.

2. The system then produces word based confidence scores and error profiles to help diagnose the current reading status/confidence of the learner. In addition the face analysis service uses head pose as well as eye gaze to proximate learners state e.g attentive, frustrated, distracted. Additionally, the student’s teacher also inputs their observations. The learner profile is saved and accessed and adjusted each time the learner uses the system. 

3. The content to be presented to the learner is then analysed to determine which words are most likely to cause problems. This analysis is based on the latest expert knowledge.The problem words are then changed appropriately with the system adapting the text in the following ways ⁽⁶⁾ , as necessary: 

• Modifying font type, size and colour of the whole text, but also of individual letters 
• Highlighting ( style, speed and the highlighting of syllables)
• Segmentation at paragraph-level, sentence-level, and word-level.
• Hiding parts of words
• Display of help cues e.g pictures

Note that as and when the reader becomes more proficient, these changes are adapted accordingly. 

4. Each time the system is used the face analysis service is activated to proximate learner state and further adjustments may be made to text to accommodate this. For example if the learner is in a distracted state the text may be made bigger and/or paused until the user is back in an attentive state. 

5. The learner has a further opportunity to customise the text if they need to and these changes are then saved to their profile. 

6.  The system is a closed loop “in which we do not only apply pre-defined expert knowledge, but in which experts and teachers themselves learn about the efficiency of their pedagogical strategies. For that, the system provides the possibility to analyze usage data. Experts can then modify the adaptation rules.” ⁽⁷⁾

The diagram below represents the system in action diagrammatically. ⁽⁸⁾   

For further information a short slideshow from one of the developers summarising the objectives and details of the project can be accessed below:

http://www.slideshare.net/aps/agentdysl-a-novel-intelligentreading-system-for-dyslexic-learners

The agent DYSL approach allows for personalised learning, according to the specific needs of the user. It is also adaptive as and when the learner progresses and to their state of attention. Interestingly whilst it is an isolated learning experience for the learner, a common problem associated with technologies that offer personalisation, it also facilitates inclusion as it enables the learner to remain included in the classroom environment.

Despite an extensive search I’ve been unable to ascertain whether the agent DYSL project was ever commercialised. There is some evidence ⁽⁹⁾ that the software had been successful in increasing students scores on reading pace, reading accuracy, motivation and self esteem.  Perhaps it is still in development, or waiting for some more funding or perhaps, as noted by Drigas and Dourou, it needs to be expanded to allow assistance for a greater proportion of learners: “ the main disadvantages of the tool were that it was not tailored to the needs of adults and was limited to assisting children with reading dyslexia, not any other dyslexia difficulties.” ⁽¹⁰⁾ However whether it is commercially available or not, it still provides a real example of how technology can be used to personalise the education experience according to an individuals needs, adapt as the learner requires and in doing these things reduce barriers to equality in education. 

_{(1) Drigas A & Dourou A. A review on ICT’s, E-Learning & Artificial Intelligence for Dyslexic’s Assistance. Accessed via http://www.imm.demokritos.gr/publications/Dyslexics_Assistance.pdf}

_{(2)  http://www.youtube.com/watch?v=DjmM2fZYmgk}

_{(3) Athanasaki, M., Avramouli, M., Karpouzis, K., Kollias, S., Ntalianis, K., Schmidt, A., Symvonis, A., Valcarcel, F. Agent-dysl: A novel intelligent reading system for dyslexic learners, Cunningham, (eds.), Expanding the Knowledge Economy. Proceedings of E-Challenges 2007, IOS Press, 2007}

_{(4) Tzouveli, P., Schmidt, A,. Schneider, M., Symvonis, A., Kollias S Adaptive Reading Assistance for the Inclusion of Students with Dyslexia: The AGENT-DYSL approach.} 

_{(5) Athanasaki, M., Avramouli, M., Karpouzis, K., Kollias, S., Ntalianis, K., Schmidt, A., Symvonis, A., Valcarcel, F. Agent-dysl: A novel intelligent reading system for dyslexic learners, Cunningham, (eds.), Expanding the Knowledge Economy. Proceedings of E-Challenges 2007, IOS Press, 2007}

_{(6) Ibid}

_{(7) Ibid}

_{(8) Ibid}

_{(9) Athanasalis T et al Making assistive reading tools user friendly: a new platform for Greek dyslexic students empowered by customised speech recognition. Springer Science and Business Media, LLC 2012.}

_{(10) Drigas A, and Dourou A. A review on ICT’s, E-Learning and Artificial Intelligence for Dyslexic’s Assistance. Accessed via http://www.imm.demokritos.gr/publications/Dyslexics_Assistance.pdf}