Ordinal pattern transition networks in eye tracking reading signals.

Area of Science:

• Neuroscience
• Cognitive Science
• Biomedical Engineering

Background:

• Eye tracking is an emerging technology for non-invasive neurocognitive diagnosis.
• It offers improved assessment of latent neurophysiological information.
• Analyzing eye movement data for cognitive insights is an evolving field.

Purpose of the Study:

• To apply ordinal patterns transition networks for identifying dyslexia during text reading.
• To characterize eye movement transitions for distinguishing dyslexic from typically developed subjects.
• To explore the potential of eye tracking in understanding neurocognitive conditions.

Main Methods:

• Collected eye tracking data during simple text reading experiments from subjects with and without dyslexia.
• Analyzed the temporal evolution of eye movements using ordinal patterns.
• Characterized transitions between ordinal patterns and used their relative frequencies as features for a classifier.

Main Results:

• A classifier achieved almost 100% accuracy in distinguishing typically developed from dyslexic subjects.
• The classification was based on the relative frequencies of specific eye movement transition patterns.
• This approach highlights differences in cognitive behavior during reading.

Conclusions:

• Ordinal patterns transition networks offer a highly accurate method for dyslexia identification via eye tracking.
• The findings provide insights into the cognitive underpinnings of dyslexia.
• This methodology can be extended to analyze other neurocognitive conditions.