AUTHOR=Hadar-Shoval Dorit , Elyoseph Zohar , Lvovsky Maya 

TITLE=The plasticity of ChatGPT’s mentalizing abilities: personalization for personality structures

JOURNAL=Frontiers in Psychiatry

VOLUME=Volume 14 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/psychiatry/articles/10.3389/fpsyt.2023.1234397

DOI=10.3389/fpsyt.2023.1234397

ISSN=1664-0640

ABSTRACT=Contribution to the field statement This study explores how ChatGPT, a large language model, can generate mentalizing-like responses tailored to specific personality structures and psychopathologies. It focuses on Borderline Personality Disorder (BPD) and Schizoid Personality Disorder (SPD), both of which involve distinct emotional intensity and regulation patterns. The study used ChatGPT's free version 23.3 to evaluate mentalizing-alike responses using the Levels of Emotional Awareness Scale (LEAS) . The findings indicate that ChatGPT was able to describe the emotional experiences of individuals with BPD as more intense, complex, and rich than those with SPD. This suggests that ChatGPT can generate mentalizing responses consistent with various psychopathologies, aligning with existing clinical knowledge . However, the study also raises concerns about potential biases and stigmas associated with mental diagnoses, which may impact the effectiveness and validity of chatbot-based interventions in clinical practice. It emphasizes the need for responsible development in mental health interventions, which considers diverse theoretical perspectives . Overall, this study contributes to the field by demonstrating the potential of ChatGPT in generating mentalizing-like responses tailored to different psychopathologies. It highlights the importance of considering diverse perspectives in the responsible use of chatbot-based interventions in mental health.