Internal World Models as Imagination Networks in Cognitive Agents
This work addresses the problem of developing human-like imagination in artificial intelligence for researchers in cognitive science and AI, though it is incremental as it builds on existing methods for comparing representations.
The study tackled the problem of understanding imagination's computational objective by comparing internal world models in humans and large language models (LLMs) using psychological network analysis. The results showed that human imagination networks had correlations between centrality measures, while LLM networks lacked clustering and had lower correlations, indicating a lack of similarity between human and LLM agents.
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.