Stringology-Based Motif Discovery from EEG Signals: an ADHD Case Study

Researchers Anat Dahan and Samah Ghazawi have published a novel computational framework on arXiv that applies stringology—the study of efficient string processing algorithms—to the analysis of electroencephalography (EEG) signals. The primary aim is to introduce quantitative measures for understanding neural dynamics, with this paper serving as a proof-of-concept focused on Attention-Deficit/Hyperactivity Disorder (ADHD). The framework adapts two specific algorithms, Order-Preserving Matching (OPM) and Cartesian Tree Matching (CTM), to detect recurrent temporal patterns, or 'motifs,' in brainwave data. These methods are designed to preserve the relative ordering and hierarchical structure of signal patterns while remaining invariant to amplitude scaling, offering a temporally precise representation that complements traditional spectral analysis.

The researchers applied their framework to a publicly available dataset of multichannel EEG recordings from individuals with ADHD and matched controls. The analysis revealed three systematic, group-specific differences in neural activity. First, the ADHD group exhibited significantly higher motif frequencies, suggesting increased repetitiveness. Second, OPM analysis showed ADHD motifs were shorter and had greater gradient instability, reflected in larger inter-sample amplitude changes. Third, CTM analysis demonstrated reduced hierarchical complexity in ADHD, characterized by shallower tree structures with fewer levels. These findings suggest ADHD involves distinct alterations in the structure and organization of temporal brain patterns. The proposed framework provides a new, complementary computational tool with clear potential for developing objective, data-driven biomarkers to aid in the diagnosis and study of neurodevelopmental disorders.