This is a past event
Presenter: Iacopo Caporossi, University of Siena.
Title: A nonlinear dynamical approach to the symptoms graph.
Abstract: This talk presents a theoretical and data-driven investigation into the temporal evolution of psychological disorders through the lens of dynamical systems and network theory.
We first introduce a mathematical framework designed to capture the time-dependent behaviour of symptom fluctuations in a patient, focusing on two key psychological dynamics: hysteretic regimes, in which the system’s state depends on its past trajectory, and metastable regimes, where symptoms linger near quasi-stable configurations before transitioning. Through theoretical simulations, we illustrate how these regimes can emerge naturally from the model and how they shape the observable evolution of disorder intensity over time.
Building on this, we employ network theory applied to psychology to formalize symptoms as nodes within a symptoms graph, emphasizing the role of graph connectivity in constraining the system’s dynamics. Finally, we present a case study based on a longitudinal dataset containing daily mood self-assessments from a patient with depression undergoing varying antidepressant dosages. For each observed items, we estimate the number of metastable states (called metastates) detectable in its temporal trajectory and examine how these estimates relate to properties of the symptoms graph. We explore the correlation between symptom connectivity and the degree of metastability, offering new insights into how network structure may modulate the stability and variability of symptoms psychological states.
- Speaker
- Iacopo Caporossi
- Hosted by
- Department of Physics
- Venue
- Meston 311