Multimodal Large Language Models as Synthetic Participants in Video-Based Studies: An Evaluation
On the map: Multimodal LLMs / vision-language models → Multimodal emotion understanding
Abstract (arXiv)
Multimodal large language models (MLLMs) have shown strong performance on objective tasks such as video understanding and reasoning. However, it remains unclear whether they can approximate subjective human responses, which depend not only on content comprehension but also on individuals' social contexts. To address this gap, we evaluate MLLMs as synthetic participants in an emerging task: assessing perceived sensory engagement with short videos. Grounded in the Perceived Message Sensation Value (PMSV) framework, we compare ratings from recruited human participants and profile-conditioned MLLM simulations (n=673) using a 17-item scale measuring emotional arousal, dramatic impact, and novelty. We find that even leading MLLMs (Gemini 3 Flash and Qwen 3 Omni) show limited agreement with human participants. The models exhibit distinct downward mean-shift and central-tendency biases in their rating distributions. They both introduce and flatten subgroup differences, while showing inconsistent sensitivity to participant profiles. Prompting strategies affect these metrics differently, modestly improving some aspects while worsening others. These results highlight both the challenges and opportunities of developing MLLMs as synthetic participants in video-based research. Data and code: https://github.com/MINDLab25/mllm-human-simulation-eval
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Multimodal emotion understanding, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 7 of its 18 references that are in the map. First posted on arXiv 1 May 2026.
Builds on
7 of 18 references.
References · 7
- 2023MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGIcited by 2,080
- 2022Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?cited by 1,434
- 2023Large Language Models Cannot Self-Correct Reasoning Yetcited by 1,017
- 2025Qwen3-Omni Technical Reportcited by 497
- 2025Finetuning LLMs for Human Behavior Prediction in Social Science Experimentscited by 39
- 2024Algorithmic Fidelity of Large Language Models in Generating Synthetic German Public Opinions: A Case Studycited by 14
- 2026A Computational Model of Message Sensation Value in Short Video Multimodal Features that Predicts Sensory and Behavioral Engagementcited by 1
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2025Customizing Visual Emotion Evaluation for MLLMs: An Open-vocabulary, Multifaceted, and Scalable Approach
- 2025Correspondence of high-dimensional emotion structures elicited by video clips between humans and Multimodal LLMs
- 2025Can Large Language Models Grasp Concepts in Visual Content? A Case Study on YouTube Shorts about Depression ↗
- 2025Computational emotion analysis with multimodal LLMs: Current evidence on an emerging methodological opportunity
- 2025MME-Emotion: A Holistic Evaluation Benchmark for Emotional Intelligence in Multimodal Large Language Models
- 2024HumanVBench: Probing Human-Centric Video Understanding in MLLMs with Automatically Synthesized Benchmarks
- 2025HumanVideo-MME: Benchmarking MLLMs for Human-Centric Video Understanding
- 2025Multimodal Large Language Models for End-to-End Affective Computing: Benchmarking and Boosting with Generative Knowledge Prompting