EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

The ChatGPT Effect: Investigating Shifting Discourse Patterns, Sentiment, and Benefit–Challenge Framing in AI Mental Health Support

Sanguk Lee iD, Minjin Mj Rheu iD, Jie Zhuang

Behavioral Sciences · 28.08.2025 · 15 (9) · 1172

DOI: 10.3390/bs15091172 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Behavioral Sciences, online erschienen am 28.08.2025. Verfasst von Sanguk Lee und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Communication Studies, Texas Christian University, Fort Worth, TX 76109, USA. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Willmon Grant (Department of Communication Studies at Texas Christian University), Willmon Grant, an internal funding source from the Department of Communication Studies at Texas Christian University, Hankuk University of Foreign Studies Research Fund of 2025.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

AI has the potential to enhance mental health by scaling support. However, its implementation brings uncertainties and challenges that require careful review to ensure safety. This study examined evolving public views on AI mental health support by analyzing relevant Reddit posts (n = 517). Following the release of ChatGPT in 2022, discussions about AI in the context of mental health surged, with a noticeable shift in preference toward large language models (LLMs) over conventional therapy chatbots. Users appreciated AI for its emotional support, companionship, and accessibility, while also expressing concerns about adverse effects and lack of conversational depth and emotional connection. Distinct patterns in how benefits and challenges were discussed emerged between experienced and non-experienced AI users, as well as between AI-focused and mental health-focused communities. AI-experienced users acknowledged both the benefits and limitations, whereas AI communities emphasized the positives and mental health communities highlighted the lack of conversational depth. These findings underscore the need for tailored communication strategies to set realistic expectations about the utility of AI in mental healthcare among different stakeholders. This research provides insights into developing ethical AI systems that complement traditional care while addressing current limitations.

Schlagwörter

Stichwörter der Autor:innen

AI for mental healthRedditlarge language modelsmental health

Autor:innen und Institutionen

  • Department of Communication Studies, Texas Christian University, Fort Worth, TX 76109, USA.
  • Division of Media & Communication, Hankuk University of Foreign Studies, Seoul 02450, Republic of Korea.
  • School of Communication, Loyola University Chicago, Chicago, IL 60660, USA.

ORCID-Profile: Sanguk Lee, Minjin Mj Rheu

Förderung

  • Willmon Grant (Department of Communication Studies at Texas Christian University)
  • Willmon Grant, an internal funding source from the Department of Communication Studies at Texas Christian University
  • Hankuk University of Foreign Studies Research Fund of 2025

Zitieren

APA

Lee, S., Rheu, M. M., & Zhuang, J. (2025). The ChatGPT Effect: Investigating Shifting Discourse Patterns, Sentiment, and Benefit–Challenge Framing in AI Mental Health Support. Behavioral Sciences, 15(9), 1172. https://doi.org/10.3390/bs15091172

Vancouver

Lee S, Rheu MM, Zhuang J. The ChatGPT Effect: Investigating Shifting Discourse Patterns, Sentiment, and Benefit–Challenge Framing in AI Mental Health Support. Behavioral Sciences. 2025;15(9):1172. doi:10.3390/bs15091172

BibTeX
@article{Lee2025,
  author = {Lee, Sanguk and Rheu, Minjin Mj and Zhuang, Jie},
  title = {The ChatGPT Effect: Investigating Shifting Discourse Patterns, Sentiment, and Benefit–Challenge Framing in AI Mental Health Support},
  journal = {Behavioral Sciences},
  year = {2025},
  volume = {15},
  number = {9},
  pages = {1172},
  doi = {10.3390/bs15091172}
}
Kostenlos starten →