Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
June 15, 2026BMC PsychiatryOpen Access

How do large language models answer ADHD-related questions? A comparative study of ChatGPT, Gemini, and DeepSeek

View Full Paper
Ask AI
Bookmark
Share

Authors

BBBerrin BilgiçSTSerkan TuranSASibelnur Avcil

Discussion

Loading...

Member takes

Overview

Comparative study evaluates accuracy and reliability of ChatGPT, Gemini, and DeepSeek in answering ADHD-related questions, suggesting varied effectiveness.

Key Points

  • This study evaluates and compares the performance of large language models in answering ADHD-related questions.
  • Cross-sectional comparative design evaluating 22 ADHD-related questions across three models.
  • Responses were assessed by two child psychiatry specialists using standardized prompts.
  • Descriptive statistics and non-parametric repeated-measures analyses were employed for comparison.
  • Mean accuracy scores were 91% for ChatGPT, 89% for Gemini, and 87% for DeepSeek.
  • Reproducibility scores were 89% for ChatGPT, 86% for Gemini, and 84% for DeepSeek.
  • Significant differences in quality and usefulness were observed, with ChatGPT receiving the highest expert scores.

Cite This Study

Bilgiç et al. (2026) studied this question.

synapsesocial.com/papers/6a2f96eca1cfeec49082809chttps://doi.org/10.1186/s12888-026-08280-x
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Evaluation of large language models in emergency medicine scenarios: a comparative analysis of ChatGPT-4o, ChatGPT-o3mini, Gemini 2.0-pro, and DeepSeek-R12026 · 1 citations
  2. 2Performance of ChatGPT, Gemini and DeepSeek for non-critical triage support using real-world conversations in emergency department2025 · 22 citations
  3. 3ChatGPT’s Response Consistency: A Study on Repeated Queries of Medical Examination Questions2024 · 60 citations
  4. 4ChatGPT and the rise of large language models: the new AI-driven infodemic threat in public health2023 · 678 citations
  5. 5Health information seeking in the digital age: An analysis of health information seeking behavior among US adults2017 · 680 citations