Capturing Thought Dynamics: Modeling the Structure of Spontaneous Associations

Abstract

Abstract

Human thought unfolds in patterns that vary across individuals; some move fluidly across ideas, whereas others return to narrower or more repetitive paths. Here, we asked whether structural aspects of associative thinking-its breadth, trajectory, and temporal pace-can account for individual differences in repetitive thought styles such as rumination. Participants completed brief free-association tasks, and their responses were quantified using a model trained on large-scale free association data to derive individual structural profiles of associative flow. Across two independent samples, these profiles predicted self-reported measures of rumination and repetitive thinking, with nonlinear models accounting for meaningful inter-individual variation and showing generalization across tasks and datasets. These findings suggest that individual differences in repetitive thought are reflected not only in what people think about but also in how associations are organized and unfold over time. Structural modeling therefore provides a viable approach to characterizing spontaneous cognition through patterns of associative organization.