Exploring Spatial Schema Intuitions in Large Language and Vision Models
This research addresses the underexplored problem of embodiment in LLMs for AI researchers, providing incremental insights into how these models relate to human spatial intuitions.
The study investigated whether large language models (LLMs) capture implicit human intuitions about spatial schemas, despite lacking embodiment, by reproducing psycholinguistic experiments and found correlations between model outputs and human responses, with notable distinctions in polarized responses and reduced correlations in vision language models.
Despite the ubiquity of large language models (LLMs) in AI research, the question of embodiment in LLMs remains underexplored, distinguishing them from embodied systems in robotics where sensory perception directly informs physical action. Our investigation navigates the intriguing terrain of whether LLMs, despite their non-embodied nature, effectively capture implicit human intuitions about fundamental, spatial building blocks of language. We employ insights from spatial cognitive foundations developed through early sensorimotor experiences, guiding our exploration through the reproduction of three psycholinguistic experiments. Surprisingly, correlations between model outputs and human responses emerge, revealing adaptability without a tangible connection to embodied experiences. Notable distinctions include polarized language model responses and reduced correlations in vision language models. This research contributes to a nuanced understanding of the interplay between language, spatial experiences, and the computations made by large language models. More at https://cisnlp.github.io/Spatial_Schemas/