A Dynamic Model of Spontaneous Stuck Thought Open Access

Abstract

While worry and rumination are highly prevalent features of internalizing disorders, their cognitive mechanisms remain unclear. Moreover, uncovering these mechanisms has proved challenging experimentally, as worried and ruminative thoughts often occur in the absence of measurable behaviors. To date, free-association paradigms and semantic word associations have shown some promise for uncovering mechanisms of worry and rumination, though this work remains in early stages. Here, we analyze word-association data using a dynamic attractor-state modeling that conceptualizes repetitive negative thinking as a phenomenon of spontaneously navigating a multidimensional semantic space while in the presence of a strong maladaptive attractor space. The previously validated Free Association of Semantics Task (FAST) was used to collect word-associations from two samples of participants: 79 online and 65 in-person. This task required users to submit single word responses to prompts based on the psychoanalytic procedure of “free association”. Submitted words were first embedded using a pre-trained GloVe model, and the resulting multi-dimensional semantic space was reduced to 7 dimensions, explaining 90% of the data variance. Using dynamic attractor-state modeling, data simulations were conducted and compared to the collected data using cosine similarity. An optimizer helped find the best fitting parameters, which were later clustered using unsupervised k-means. Consistently, one of three resulting clusters yielded higher levels of perceived pathology, lower levels of enjoyment, and revealed a pattern of thought inflexibility and repetition, which resembled worried or ruminative thinking. Indeed, individuals whose data mostly clustered into that cluster also self-reported higher measures of rumination and worry. Our results support a conceptual model of repetitive negative thinking derived from attractor-state dynamics, thereby providing promise to a novel method for measuring rumination and worry severity.

Table of Contents

Introduction Materials & Methods Participants Task Data Preprocessing Modeling Dynamic Modeling Kernel Density Estimation Unsupervised Clustering Statistical Analyses Results Rating Scales Model Clustering Overview Cluster Descriptions Cluster Interpretations Cluster Testing Discussion Limitations and Next Steps Tables Figures Citation Diversity Statement References

About this Master's Thesis

Rights statement

• Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.

School	Laney Graduate School
Department	Psychology
Subfield / Discipline	Clinical Psychology
Degree	M.A.
Submission	Master's Thesis
Language	English
Research Field	Psychology, Clinical Psychology, Cognitive
Keyword	Spontaneous Thought Worry Rumination Mind-Wandering Dynamic Modeling
Committee Chair / Thesis Advisor	Michael Treadway, Emory University
Committee Members	Philip Kragel, Emory University Alexandra Cohen, Emory University Jessica Cooper, Emory University

Last modified

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	A Dynamic Model of Spontaneous Stuck Thought ()	2023-05-03 11:58:29 -0400	Press to Select an action Download

Supplemental Files

Thumbnail	Title	Date Uploaded	Actions