Psychological Imagination Networks Show Cross-Population Centrality and Clustering Alignment in Humans That Large Language Models Fail to Replicate
arXiv:2510.04391v5 Announce Type: replace-cross Abstract: Mental imagery vividness is a stable individual trait, yet whether imagined scenarios share relational structure across human and synthetic large language model (LLM) populations remains unknown. We applied psychological network analysis to vividness ratings from two validated questionnaires: the Vividness of Visual Imagery Questionnaire (VVIQ-2) and the Plymouth Sensory Imagery Questionnaire (PSIQ), across geographically and linguistically distinct human samples (Florida, Poland, and London; total N = 2,743) and six large language mode
The proliferation of advanced large language models necessitates a deeper understanding of their cognitive and experiential limitations compared to human cognition, particularly in nuanced areas like imagination.
This research provides a critical benchmark for the fidelity of LLM-generated internal states and imaginative capabilities, highlighting a significant divergence from human psychological structures.
The perceived gap between human imagination and LLM simulation of it is quantified, suggesting limitations in LLMs' ability to replicate complex human mental processes beyond linguistic pattern matching.
- · Cognitive science researchers
- · Developers of psychologically grounded AI
- · Neuroscience
- · Proponents of LLMs as near-human intelligence
- · Applications requiring deep human-like imaginative understanding
Further research will focus on identifying the architectural or training data deficiencies contributing to LLMs' failure to replicate human imaginative structures.
This understanding could lead to the development of novel AI architectures specifically designed to foster more human-like internal experiential states and imaginative capacities.
A clearer delineation of human vs. AI cognitive strengths might reshape philosophical discussions on consciousness and the unique aspects of human intelligence.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.CL