Sharma, Shivank https://orcid.org/0009-0008-4429-406X
Taylor, Matthew https://orcid.org/0000-0001-9480-4832
Sadiq, Zara
Kashyap, Tanya
Shao, Hanlei
Sulaiman, Zeba
Wang, Yiqiao
Kumarendran, Balachandran
Griffiths, Sian Lowri https://orcid.org/0000-0003-0031-7174
Article History
Received: 29 April 2025
Revised: 12 December 2025
Accepted: 10 February 2026
First Online: 24 April 2026
Competing interests
: The authors declare that they have no conflicts of interest relevant to the research, authorship, and/or publication of this systematic review.
: Artificial intelligence (AI) was used solely to assist with administrative and analytical aspects of the literature review process. The authors used Claude AI, developed by Anthropic, and Grammarly to enhance the manuscript’s readability and clarity. Specifically, ChatGPT (OpenAI, GPT-5, 2025 version) was employed for the May 2024 –September 2025 updated search to: 1. Screen and manage bibliographic data exported from Ovid/Embase, Medline, PsycINFO, and Web of Science, 2. Identify duplicates, parse publication dates , and filter articles published between May 2024 and September 2025 (including online-first records) , 3. Summarise abstracts to verify whether each study met predefined inclusion criteria (human social-defeat exposure, inflammatory markers, and mental health outcomes), and 4. Generate structured summary tables (e.g., inclusion/exclusion decisions and variable extraction fields). All AI outputs were verified manually by the authors before inclusion, and no AI system was used to generate or modify the scientific interpretation, results, or conclusions. The final data synthesis, wording, and critical appraisal were entirely written and approved by the human authors.