Artificial Intelligence Mimics Human Social Perception, Opening New Avenues in Neuroscience Research

A new study reveals that AI models like ChatGPT can evaluate social interactions with accuracy comparable to humans, advancing neuroscience research and practical applications in healthcare and security.
Recent advancements in artificial intelligence have demonstrated its capacity to analyze and interpret social interactions between individuals in images and videos with a level of accuracy comparable to human perception. A pioneering study conducted by researchers at the University of Turku in Finland, published in the journal Imaging Neuroscience, highlights how models like ChatGPT can assess various social features, including facial expressions, body language, cooperation, and hostility in diverse visual media.
In this investigation, the AI evaluated 138 social traits from numerous videos and photographs, and its assessments were compared against over 2,000 human evaluations. Remarkably, the AI's evaluations closely aligned with human judgments and were even more consistent than those of a single human observer. Postdoctoral researcher Severi Santavirta noted that the AI's evaluations could be considered highly trustworthy, although collaborative human assessments still outperform AI in accuracy.
An important aspect of this research was its exploration of how AI-driven social evaluations can aid neuroscience. By modeling brain networks involved in social perception using functional brain imaging data, researchers found that social assessments made by AI mirrored those derived from human judgments. This suggests AI's potential to facilitate large-scale, cost-effective neuroscience experiments, reducing manual effort and speeding up data processing. For example, what traditionally might require thousands of hours of human work can now be achieved in mere hours by AI.
Beyond research, the capabilities of AI in understanding social interactions have promising applications across various sectors. In healthcare, AI could monitor patient well-being continuously through video analysis. In marketing, it could predict audience responses to audiovisual content, and in security, it could identify abnormal situations via surveillance footage. AI's ability to operate tirelessly around the clock presents a significant advantage for continuous monitoring and analysis.
Overall, this study underscores how artificial intelligence is becoming an essential tool in both neuroscience and practical fields, offering new pathways to understand and interpret complex social behaviors efficiently and reliably.
source: https://medicalxpress.com/news/2025-09-ai-social-situations-similar-humans.html
Stay Updated with Mia's Feed
Get the latest health & wellness insights delivered straight to your inbox.
Related Articles
US Dementia Incidence Declines But Prevalence Continues to Rise Amid Population Aging
A recent study reveals that while new US dementia cases have declined since 2015, the number of people living with dementia continues to grow due to population aging, with significant disparities among different communities.
New Sex-Specific Pathway Uncovered in Melanoma Metastasis with Implications for Female Cancer Therapies
A groundbreaking study uncovers a sex-specific pathway involving E-cadherin loss, ERα, and GRPR that drives melanoma metastasis in women, opening new avenues for targeted therapies.
Artificial Light Disrupts Natural Connection Between Menstrual Cycles and Lunar Phases, Study Finds
A groundbreaking study reveals that artificial lighting diminishes the natural synchronization of women's menstrual cycles with lunar phases, potentially affecting fertility and health.
Prompt Treatment of Rare Flu-Related Brain Complication Critical for Children, Study Highlights
A Stanford-led study reveals that rapid treatment improves survival and recovery in children with influenza-associated brain swelling, emphasizing the importance of vaccination and early intervention.



