10th EAI International Conference on Pervasive Computing Technologies for Healthcare

Research Article

Dytective: Diagnosing Risk of Dyslexia with a Game

  • @INPROCEEDINGS{10.4108/eai.16-5-2016.2263338,
        author={Luz Rello and Miguel Ballesteros and Abdullah Ali and Miquel Serra and Daniela Alarcón Sánchez and Jeffrey P. Bigham},
        title={Dytective: Diagnosing Risk of Dyslexia with a Game},
        proceedings={10th EAI International Conference on Pervasive Computing Technologies for Healthcare},
        publisher={ACM},
        proceedings_a={PERVASIVEHEALTH},
        year={2016},
        month={6},
        keywords={dyslexia; screening; diagnosis; serious games; linguistics},
        doi={10.4108/eai.16-5-2016.2263338}
    }
    
  • Luz Rello
    Miguel Ballesteros
    Abdullah Ali
    Miquel Serra
    Daniela Alarcón Sánchez
    Jeffrey P. Bigham
    Year: 2016
    Dytective: Diagnosing Risk of Dyslexia with a Game
    PERVASIVEHEALTH
    EAI
    DOI: 10.4108/eai.16-5-2016.2263338
Luz Rello1,*, Miguel Ballesteros2, Abdullah Ali3, Miquel Serra4, Daniela Alarcón Sánchez5, Jeffrey P. Bigham1
  • 1: Carnegie Mellon University
  • 2: Universitat Pompeu Fabra
  • 3: University of Maryland Baltimore County
  • 4: Department of Basic Psychology
  • 5: Change Dyslexia
*Contact email: luzrello@gmail.com

Abstract

More than 10% of the population has dyslexia, and most are diagnosed only after they fail in school. This work seeks to change this through scalable early detection via machine learning models that predict reading and writing difficulties by watching how people interact with a linguistic web-based game: Dytective. The design of Dytective is based on (i) the empirical linguistic analysis of the errors that people with dyslexia make, (ii) principles of language acquisition, and (iii) specific linguistic skills related to dyslexia. Experiments with 243 children and adults (95 with diagnosed dyslexia) revealed differences in how people with dyslexia read and write. We trained a machine learning model that was able to predict dyslexia with 83% accuracy in a held-out test set with 100 participants. Currently, we are working with schools to put our approach into practice at scale to reduce school failure as a primary way dyslexia is diagnosed.