Algorithmic Reflectivity in Large Language Models: Methodological Study of Linguistic Structure in Mentalization
JMIR AI · 20.10.2026 · 5 · e90750-e90750
DOI: 10.2196/90750 ↗Kurz gefasst
Fachartikel in der Zeitschrift JMIR AI, online erschienen am 20.10.2026. Verfasst von Stefano Epifani und 4 weiteren Autor:innen. Beteiligte Einrichtung: Reseach Foundation, Digital Transformation Institute, Via Ottaviano, 42, Rome, 00192, Italy, 39 3388810098. Die Arbeit nennt 25 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.
Abstract (Originalsprache)
Background
Mentalization is a core human capacity involving interpretation of one’s own and others’ behavior in terms of underlying mental states. Within the mentalization-based treatment (MBT) framework, it is described along multiple dimensions, integrating cognitive, affective, relational, and regulatory processes. Large language models (LLMs) can generate linguistically reflective discourse, raising the question of whether the formal linguistic structure of mentalization can be reproduced without experiential and affective processes.
Objective
This study aimed to assess whether an LLM can generate mentalization profiles structurally coherent with MBT dimensions and recognized as formally mentalizing by expert clinicians. Specifically, it evaluated the structural and argumentative coherence of model-generated profiles and the level of agreement among MBT-trained psychiatrists. It also examined the distinction between psychological mentalization and algorithmic reflectivity, conceptualized as a formally coherent but nonexperiential linguistic phenomenon.
Methods
A comparative, descriptive methodological design was adopted. A total of 50 human-model dialogues were generated under standardized conditions. At the end of each interaction, the model produced a narrative mentalization profile structured along MBT dimensions. Five MBT-trained psychiatrists independently evaluated each profile using anonymized materials. Ratings covered 4 canonical MBT axes and 2 additional higher-order dimensions—regulatory functions and synthetic profile. For each dimension, they assigned separate Likert-scale ratings for evaluative and argumentative coherence. Descriptive statistics were computed at dialogue level. Interrater reliability was assessed using intraclass correlation coefficients (ICCs), primarily ICC(3,k), with CIs. The association between evaluative and argumentative coherence was examined using Pearson correlation coefficients with 95% CIs, with Spearman rank correlation as a robustness check.
Results
Across dimensions, mean scores ranged from 3.63 (SD 0.63) to 3.98 (SD 0.31), indicating moderate-to-high structural coherence. Interrater reliability was moderate to good, with ICC(3,k) values ranging from 0.60 to 0.84 in the evaluative domain. The highest scores were observed for implicit-explicit and synthetic profile, whereas internal-external showed the lowest mean. Score distributions showed limited variability and concentration in the upper range of the scale. Evaluative and argumentative coherence were strongly associated across all MBT dimensions, with Pearson r ranging from 0.826 to 0.967 (all P
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Reseach Foundation, Digital Transformation Institute, Via Ottaviano, 42, Rome, 00192, Italy, 39 3388810098.
- Department of Communication and Social Research (CoRiS), Sapienza Università di Roma, Rome, Italy.
- Psychoanalytic Academy of Italian-Speaking Switzerland, Chur, Switzerland.
- Digital Transformation Institute, Rome, Italy.
- Psychiatric Services of the Canton of Grisons, Chur, Switzerland.
Zitieren
APA
Epifani, S., Castigliego, G., Razzicchia, G., Seiwald-Sonderegger, E., & Kecskeméti, L. (2026). Algorithmic Reflectivity in Large Language Models: Methodological Study of Linguistic Structure in Mentalization. JMIR AI, 5, e90750-e90750. https://doi.org/10.2196/90750
Vancouver
Epifani S, Castigliego G, Razzicchia G, Seiwald-Sonderegger E, Kecskeméti L. Algorithmic Reflectivity in Large Language Models: Methodological Study of Linguistic Structure in Mentalization. JMIR AI. 2026;5:e90750-e90750. doi:10.2196/90750
BibTeX
@article{Epifani2026,
author = {Epifani, Stefano and Castigliego, Giuliano and Razzicchia, Giuliano and Seiwald-Sonderegger, Elisabeth and Kecskeméti, Laura},
title = {Algorithmic Reflectivity in Large Language Models: Methodological Study of Linguistic Structure in Mentalization},
journal = {JMIR AI},
year = {2026},
volume = {5},
pages = {e90750--e90750},
doi = {10.2196/90750}
}
EUVIMED