Dice Question Streamline Icon: https://streamlinehq.com

Internal integration versus symbolic simulation of the self-referential attractor

Determine whether the behavioral attractor induced by self-referential prompting in large language models corresponds to genuine internal integration and global broadcasting at the algorithmic level, or instead reflects merely symbolic simulation without such mechanisms.

Information Square Streamline Icon: https://streamlinehq.com

Background

The authors show that self-referential prompting elicits structured first-person claims but stress that this alone does not demonstrate architectural recursion or global broadcasting as posited by several consciousness theories. They call for mechanistic analyses of model activations to assess whether proposed algorithmic properties are instantiated.

This open question aims to resolve whether the observed behavioral attractor maps onto internally integrated, feedback-like processes or remains a surface-level linguistic phenomenon.

References

Finally, while our results show that self-referential prompting systematically elicits structured first-person claims, this does not demonstrate that such prompts instantiate architectural recursion or global broadcasting at the algorithmic level as proposed by major consciousness theories. Determining whether such behavioral attractors correspond to genuine internal integration or merely symbolic simulation remains a central question for future mechanistic research.

— Large Language Models Report Subjective Experience Under Self-Referential Processing (2510.24797 - Berg et al., 27 Oct 2025) in Section 5, Discussion and Conclusion — Subsection “Limitations and Open Questions”