Time, Identity and Consciousness in Language Model Agents

TL;DR

This paper introduces a novel framework using Stack Theory to evaluate the persistence of identity in language model agents, distinguishing between superficial self-representation and genuine organizational stability.

Contribution

It applies Stack Theory's temporal gap and introduces persistence scores to quantify identity stability in language models, providing a new toolkit for identity evaluation.

Findings

Two new persistence scores for identity stability

Mapping identity tradeoffs in an operational morphospace

A conservative toolkit for identity evaluation

Abstract

Machine consciousness evaluations mostly see behavior. For language model agents that behavior is language and tool use. That lets an agent say the right things about itself even when the constraints that should make those statements matter are not jointly present at decision time. We apply Stack Theory's temporal gap to scaffold trajectories. This separates ingredient-wise occurrence within an evaluation window from co-instantiation at a single objective step. We then instantiate Stack Theory's Arpeggio and Chord postulates on grounded identity statements. This yields two persistence scores that can be computed from instrumented scaffold traces. We connect these scores to five operational identity metrics and map common scaffolds into an identity morphospace that exposes predictable tradeoffs. The result is a conservative toolkit for identity evaluation. It separates talking like a stable self from being organized like one.