Internal World Models as Imagination Networks in Cognitive Agents

2025-10-05

AI Summary

This work investigates the computational objectives of imagination and structural disparities in internal world models (IWMs) between humans and large language models (LLMs). Method: We propose the theoretical hypothesis that “imagination serves as an IWM access mechanism” and introduce, for the first time, a psychological network analysis framework—constructing and comparing imagination networks for humans and LLMs under varied prompting and memory conditions, using the Vividness of Visual Imagery Questionnaire and centrality metrics (expected influence, intensity, proximity). Contribution/Results: Human imagination networks exhibit a robust positive correlation between clustering coefficient and centrality, whereas LLM networks lack this structural signature—revealing a fundamental architectural divergence between human and current AI IWMs. Our approach establishes a quantifiable, graph-theoretic paradigm and structural benchmark for modeling human-like imagination in artificial systems.

Technology Category

• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Machine Learning: Large Multimodal Models (LMMs)
• Humans and AI: Other Foundations of Human Computation & AI

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

What is the computational objective of imagination? While classical interpretations suggest imagination is useful for maximizing rewards, recent findings challenge this view. In this study, we propose that imagination serves to access an internal world model (IWM) and use psychological network analysis to explore IWMs in humans and large language models (LLMs). Specifically, we assessed imagination vividness ratings using two questionnaires and constructed imagination networks from these reports. Imagination networks from human groups showed correlations between different centrality measures, including expected influence, strength, and closeness. However, imagination networks from LLMs showed a lack of clustering and lower correlations between centrality measures under different prompts and conversational memory conditions. Together, these results indicate a lack of similarity between IWMs in human and LLM agents. Overall, our study offers a novel method for comparing internally-generated representations in humans and AI, providing insights for developing human-like imagination in artificial intelligence.

Problem

• Investigating computational objectives of imagination in cognitive agents
• Comparing internal world models between humans and large language models
• Developing methods to assess imagination networks in AI systems

Innovation

• Using psychological network analysis for internal models
• Constructing imagination networks from questionnaire reports
• Comparing human and AI imagination through centrality measures