Lyapunov Spectral Analysis of Speech Embedding Trajectories in Psychosis

TL;DR

This study uses Lyapunov spectral analysis of speech embeddings to distinguish psychotic from healthy speech, revealing stable dynamical differences that could aid in understanding disordered cognition.

Contribution

It introduces a novel application of Lyapunov exponents to speech embeddings, demonstrating their effectiveness in differentiating psychotic from healthy speech across models.

Findings

Word-level embeddings show no positive Lyapunov exponents.

Answer-level embeddings reveal positive Lyapunov exponents and higher-dimensional attractors.

Spectral analysis robustly separates psychotic from healthy speech.

Abstract

We analyze speech embeddings from structured clinical interviews of psychotic patients and healthy controls by treating language production as a high-dimensional dynamical process. Lyapunov exponent (LE) spectra are computed from word-level and answer-level embeddings generated by two distinct large language models, allowing us to assess the stability of the conclusions with respect to different embedding presentations. Word-level embeddings exhibit uniformly contracting dynamics with no positive LE, while answer-level embeddings, in spite of the overall contraction, display a number of positive LEs and higher-dimensional attractors. The resulting LE spectra robustly separate psychotic from healthy speech, while differentiation within the psychotic group is not statistically significant overall, despite a tendency of the most severe cases to occupy distinct dynamical regimes. These findings indicate that nonlinear dynamical invariants of speech embeddings provide a physics-inspired probe of disordered cognition whose conclusions remain stable across embedding models.