Issues register

EdTech Assessment Toolkit

Dyslexia Detection System (Lidingö, Sweden)

Prediction model

A Swedish company has developed an algorithm to assess the reading comprehension of students to support systematic reading development via eye tracking technology. In 2020, the company Lexplore claimed its software package was able to detect dyslexia in students. To do that, a student would read from a computer screen while a camera tracks the eye movements of the students. The algorithm then determines the student’s reading ability and makes personalised suggestions to increase the student’s reading development. Purchasing the software package for schools is relatively affordable and has been adopted by several schools across the world.

The system has been criticised in its ability to detect dyslexia by researchers (Witzenberger & Gulson, 2021). As of 2023, the company has dropped its claim of being able to detect dyslexia and instead has rebranded its product as a systematic reading development tool.

Snapshot (July 2023)

System task/function: Eye tracking based dyslexia detection
Model: Prediction Model
Deployment: Eye movement pattern detection
Location of application: Lindigö, Sweden
Rationale for introduction: Detect dyslexia and support reading comprehension
Vendor: Lexplore
Pricing: €22 per pupil per year (school organisation), €33 per pupil per year (single school)
Data and computation: Webcam based; AI enabled eye tracking technology.
Inequalities/harms: pre-emptive technology. It cannot be proven wrong, see (Witzenberger & Gulson, 2021).
Status: Rebranded to ‘systematic reading development’ tool
Authority/regulation: School and district level
Unintended consequences: Teachers may rely on the software rather than conventional and proven methods to support reading comprehension.
Sanction/redress: The use of the system was suspended in August 2020 and students have been awarded grades predicted by their teachers.

References/further reading

Lexplore (2022) The all-in-one service for systematic reading development. Retrieved from: https://lexplore.com/

Witzenberger, K., & Gulson, K. N. (2021). Why EdTech is always right: students, data and machines in pre-emptive configurations. Learning, Media and Technology, 46(4), 420-434. doi:10.1080/17439884.2021.1913181