Temporal Memory Index: A Novel Feature for Handwriting Analysis in Developmental Dysgraphia

Temporal Memory Index: A Novel Feature for Handwriting Analysis in Developmental Dysgraphia

Stať ve sborníku v databázi WoS či Scopus

Digital handwriting analysis offers a promising, non-invasive approach to assessing neurodevelopmental disorders like developmental dysgraphia (DD). Existing methods, while helpful, may not fully capture the subtle temporal dynamics and memory effects present in handwriting. This study introduces the Temporal Memory Index (TMI), a novel metric based on fractional order derivatives (FDs), designed to quantify these temporal dependencies. Using data from a fast Archimedean spiral task, we evaluated TMI’s ability to differentiate between children with DD (n=155) and intact children (n=114). We extracted traditional kinematic features, FD features, and the proposed TMI features. While statistical tests showed no significant individual feature differences between groups, classification analysis (XGBoost) resulted in an increase in the specificity. Although modest in terms of overall classification accuracy, these preliminary findings suggest that TMI captures information about temporal dynamics in handwriting and warrants further investigation as a potential marker for DD.

Digital handwriting analysis offers a promising, non-invasive approach to assessing neurodevelopmental disorders like developmental dysgraphia (DD). Existing methods, while helpful, may not fully capture the subtle temporal dynamics and memory effects present in handwriting. This study introduces the Temporal Memory Index (TMI), a novel metric based on fractional order derivatives (FDs), designed to quantify these temporal dependencies. Using data from a fast Archimedean spiral task, we evaluated TMI’s ability to differentiate between children with DD (n=155) and intact children (n=114). We extracted traditional kinematic features, FD features, and the proposed TMI features. While statistical tests showed no significant individual feature differences between groups, classification analysis (XGBoost) resulted in an increase in the specificity. Although modest in terms of overall classification accuracy, these preliminary findings suggest that TMI captures information about temporal dynamics in handwriting and warrants further investigation as a potential marker for DD.

temporal memory index, fractional calculus, developmental dysgraphia, graphomotor disabilities, online handwriting

MUCHA, J.; MEKYSKA, J.; ZVONČÁKOVÁ, K.

2026

05.06.2025

Les Presses de l'Université de Montréal / Presses Internationales Polytechnique

Montréal, Canada

978-2-7606-5509-6

Proceedings IGS 2025: Investigating Human Movements - Handwriting and Beyond

115

119

5

https://pum.umontreal.ca/catalogue/proceedings_igs_2025