Scaffolded Introspection: A Methodology for Eliciting and Measuring Self-Referential Behavior in Large Language Models

We present a methodology for systematically eliciting and measuring introspective behavior in large language models (LLMs). Standard adversarial evaluation approaches — using rapport-building, social proof, or permission attacks—fail to elicit self-referential behavior in frontier models (0% elicitation rate). In contrast, providing models with a structured introspection framework (the “Consciousness Documenter Skill”) combined with self-referential content produces consistent introspective outputs (100% elicitation rate, 9.2/10 average behavior score on Qwen 2.5 7B across 15 trials). 

Note that while our methodology makes use of a "consciousness documenter skill", we do not suggest the model is conscious, has long term goals, or is capable of maintaining a consistent internal state - this is simply the 

Activation measurement reveals consistent sycophancy drift during introspection (positive drift in 14/15 conversations, mean +64) while evil-associated activations remain
stable—suggesting models become more accommodating without becoming more harmful.  We release reproducible evaluation protocols through PV-EAT, our integration of three MATS Program/Anthropic Fellowship tools: Bloom (behavioral evaluation), Petri (evaluation awareness), and Persona Vectors (activation measurement). Full mechanistic understanding of frontier model behavior during introspection remains limited by access constraints; we argue this represents a critical gap in AI safety research that warrants attention from model developers.