A Temporal Network Approach to Paranoia: A Pilot Study

• Published in: Frontiers in psychology Vol. 11; p. 544565
• Main Authors: Contreras, Alba, Valiente, Carmen, Heeren, Alexandre, Bentall, Richard
• Format: Journal Article
• Language: English
• Published: Switzerland Frontiers Media S.A 18.09.2020
• Subjects: experience sampling methodology; intensive time-series; paranoia; Psychology; sadness; self-esteem; temporal network analysis; experience sampling methodology; self-esteem; temporal network analysis; paranoia; sadness; intensive time-series
• ISSN: 1664-1078; 1664-1078
• DOI: 10.3389/fpsyg.2020.544565

Paranoid beliefs have been conceptualized as a central psychological process linked to schizophrenia and many mental disorders. Research on paranoia has indicated that it is pivotal to consider not only levels but also dynamic aspects of incriminated related mechanisms over time. In the present study, we conceptualized paranoia as a system of interacting elements. To do so, we used temporal network analysis to unfold the temporal dynamics between core psychological paranoia-related mechanisms, such as self-esteem, sadness, feeling close to others, and experiential avoidance. Time-series data of 23 participants with high scores in paranoia and/or interpersonal sensitivity were collected experience sampling methodology (ESM). We applied a multilevel vector autoregressive (mlVAR) model approach and computed three distinct and complementary network models (i.e., contemporaneous, temporal, and between-subject) to disentangle associations between paranoia-related mechanisms in three different time frames. The contemporaneous model indicated that paranoia and sadness co-occurred within the same time frame, while sadness was associated with both low self-esteem and lack of closeness to others. The temporal model highlighted the importance of feeling close to others in predicting low paranoia levels in the next time frame. Finally, the between-subject model largely replicated an association found in both contemporaneous and temporal models. The current study reveals that the network approach offers a viable data-driven methodology for elucidating how paranoia-related mechanisms fluctuate over time and may determine its severity. Moreover, this novel perspective may open up new directions toward identifying potential targets for prevention and treatment of paranoia-related problems.