Novel use of ocular movements

Session Details

Session Title: Training and Innovation

Session Date/Time: Tuesday 10/10/2017 | 16:30-18:00

Paper Time: 17:18

Venue: Room 4.4

First Author: : A.Zafar GREECE

Co Author(s): :    I. Asslanides   G. Kymionis                 

Abstract Details

Purpose:

To study a novel, fast, objective, non-invasive method, named Rapid Assessment of Difficulties and Abnormalities in Reading (RADAR) that screens for features associated with the aberrant visual scanning of reading text seen in dyslexia.

Setting:

Optotech ltd and Emmetropia eye institute, Heraklion, Crete, Greece The study had the Ethics approval from the local Ethics committee.

Methods:

Seventy-eight children (8.5-12.5 years old; 42 girls and 36 boys) participated in the study with native Greek speaking children in Greece. Nine of the participating children (3 control, 6 dyslexic) were rejected blind to the diagnosis due to unreliable eye-movement recording or lack of cooperation with the experimenters. Of the remaining 69 children, 32 (15 girls, 17 boys) were diagnosed as dyslexic by the official governmental agency for diagnosing learning and reading difficulties in Greece. Participants were asked to silently read a text in the Greek language, presented on a computer monitor, while their eye movements were recorded.

Results:

The parameters analyzed were: i) fixation duration, ii) saccade length, iii) the short backward saccades (less than 4-character-long refixations representing mostly within-word backwards eye movements), and iv) the total number of fixations during the reading of the text. The circular validation was performed using 65 participants to train our test score classifier and one individual each time for testing. We have also completed a smaller study in English language. Our results show 93.8% sensitivity and 94.6% specificity towards the dyslexia diagnosis.

Conclusions:

Our study demonstrates that the RADAR method can separate effectively dyslexic from non-dyslexic readers, based on a series of eye-tracking parameters obtained during the silent reading of a standard text. We measured and compared eye movement patterns of dyslexic and non-dyslexic readers 8.5-12.5-year-old. We evaluate the power of different eye tracking parameters in discriminating among typical and dyslexic readers, ii) we assess the stability of these parameters under retesting, and iii) we combine parameters that have high discriminability and stability into a score that can classify 8.5-12.5-year-old Greek readers as typical or atypical with high sensitivity (93.8%) and specificity (94.6%).

Financial Disclosure:

NONE