Assessing Person-Specific Symptom Networks From Brief Narratives: Development and Initial Evaluation of the Narrative-to-Network (N2N) Approach
Abstract
Person-specific symptom networks formalize which problems characterize an individual and how those problems are perceived to influence one another, making them increasingly relevant for individualized case conceptualization, treatment planning, and personalized psychotherapy. However, existing approaches typically rely on structured causal assessment or intensive longitudinal data, both of which can increase assessment burden. We developed and initially evaluated the Narrative-to-Network (N2N) framework for deriving directed person-specific networks from brief spoken narratives in which individuals describe their psychological difficulties and perceived relations among them in their own words. Following a development phase (N = 22), an independent evaluation sample (N = 50) completed a narrative assessment and a Perceived Causal Networks (PECAN) assessment. Narratives lasted approximately three minutes. Manual coding showed relatively high agreement for node identification (Dice = .76) and edge direction conditional on a shared relation (agreement = .81), with lower agreement for edge presence (Dice = .62). Narrative-derived and PECAN networks showed partial but incomplete convergence (mean Jaccard = .29). Most problems identified in narratives were also identified through PECAN (62.6%), whereas only 36.7% of PECAN problems appeared in the corresponding narratives. Despite containing substantially fewer problems and relations, narrative-derived networks were perceived as highly representative; among participants with complete ratings (n = 39), mean representativeness scores were 33.26 for narrative-derived networks and 34.10 for regulated PECAN networks, with no statistically significant difference (p = .527). Narrative node count was positively associated with psychological distress both concurrently (ρ = .34) and at follow-up (ρ = .40). These findings provide initial evidence that structured person-specific symptom networks can be reproducibly derived from brief prompted spoken narratives and may offer an economical, complementary representation of individuals’ difficulties.
Related
- Decision-Supported Collaborative Care for Gestational Diabetes: A Narrative Review of Development Pathways and Evaluation in Routine Clinical Practice.
- Development and Pilot Evaluation of a Standardized and Accessible Rubric (OMNI) for Assessing Generative AI in Health Care Contexts: Four-Stage Mixed Methods Study
- Development and educational evaluation of patient-specific 3D-printed breast phantoms for oncoplastic surgical training.
- Collecting health narratives at scale: A multi-study evaluation of speech-to-text in health surveys
- Development and evaluation of indirect-ELISA based on recombinant nucleoprotein for detection of Ganjam virus (GANV) specific antibodies in small ruminants
BibTeX
@misc{vogel2026assessing,
title = {{Assessing Person-Specific Symptom Networks From Brief Narratives: Development and Initial Evaluation of the Narrative-to-Network (N2N) Approach}},
author = {Felix Vogel and Carla Paulus and Julian Reichert and Jonas Tesarz and Stefanie Jungmann},
howpublished = {PsyArXiv},
year = {2026},
month = sep,
doi = {10.31234/osf.io/x8g3n_v1},
url = {https://osf.io/preprints/psyarxiv/x8g3n_v1/},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Assessing Person-Specific Symptom Networks From Brief Narratives: Development and Initial Evaluation of the Narrative-to-Network (N2N) Approach
AU - Felix Vogel
AU - Carla Paulus
AU - Julian Reichert
AU - Jonas Tesarz
AU - Stefanie Jungmann
PY - 2026
DA - 2026/09/26
PB - PsyArXiv
DO - 10.31234/osf.io/x8g3n_v1
UR - https://osf.io/preprints/psyarxiv/x8g3n_v1/
AB - Person-specific symptom networks formalize which problems characterize an individual and how those problems are perceived to influence one another, making them increasingly relevant for individualized case conceptualization, treatment planning, and personalized psychotherapy. However, existing approaches typically rely on structured causal assessment or intensive longitudinal data, both of which can increase assessment burden. We developed and initially evaluated the Narrative-to-Network (N2N) framework for deriving directed person-specific networks from brief spoken narratives in which individuals describe their psychological difficulties and perceived relations among them in their own words. Following a development phase (N = 22), an independent evaluation sample (N = 50) completed a narrative assessment and a Perceived Causal Networks (PECAN) assessment. Narratives lasted approximately three minutes. Manual coding showed relatively high agreement for node identification (Dice = .76) and edge direction conditional on a shared relation (agreement = .81), with lower agreement for edge presence (Dice = .62). Narrative-derived and PECAN networks showed partial but incomplete convergence (mean Jaccard = .29). Most problems identified in narratives were also identified through PECAN (62.6%), whereas only 36.7% of PECAN problems appeared in the corresponding narratives. Despite containing substantially fewer problems and relations, narrative-derived networks were perceived as highly representative; among participants with complete ratings (n = 39), mean representativeness scores were 33.26 for narrative-derived networks and 34.10 for regulated PECAN networks, with no statistically significant difference (p = .527). Narrative node count was positively associated with psychological distress both concurrently (ρ = .34) and at follow-up (ρ = .40). These findings provide initial evidence that structured person-specific symptom networks can be reproducibly derived from brief prompted spoken narratives and may offer an economical, complementary representation of individuals’ difficulties.
ER -