Perceptual constancy is the ability to maintain stable perceptions of objects despite changes in sensory input, such as variations in distance, angle, or lighting. This ability is crucial for visual understanding in a dynamic world. Here, we explored such ability in current Vision Language Models (VLMs). In this study, we evaluated 155 VLMs using 236 experiments across three domains: color, size, and shape constancy. The experiments included single-image and video adaptations of classic cognitive tasks, along with novel tasks in in-the-wild conditions. We found significant variability in VLM performance across these domains, with model performance in shape constancy clearly dissociated from that of color and size constancy.
@article{arxiv.2502.10273,
title = {Probing Perceptual Constancy in Large Vision-Language Models},
author = {Haoran Sun and Bingyang Wang and Suyang Yu and Yijiang Li and Qingying Gao and Haiyun Lyu and Lianyu Huang and Zelong Hong and Jiahui Ge and Qianli Ma and Hang He and Yifan Zhou and Lingzi Guo and Lantao Mei and Maijunxian Wang and Dezhi Luo and Hokin Deng},
journal= {arXiv preprint arXiv:2502.10273},
year = {2026}
}