Fixation Sequences as Time Series: A Topological Approach to Dyslexia Detection
The hosts discuss a study applying topological data analysis to eye-tracking recordings from reading Danish texts. The researchers modeled fixation sequences as time series, using persistence diagrams to quantify movement stability. By introducing non-horizontal filtrations, they captured temporal dynamics missed by traditional methods. Results showed these topological features outperformed baseline models, providing a robust foundation for improved diagnostic tools for dyslexia.