Perspectival Control Identity Theory

A new theory, Perspectival Control Identity Theory (PCIT), proposes that phenomenal consciousness is identical to a specific type of internal control variable called a Perspectival Control State (PCS). This PCS is a temporally extended, viability-weighted stream that makes an agent's competing needs comparable and coordinates a coalition of consumers whose outputs shape the stream. The theory makes testable predictions: degree of consciousness tracks how causally important the PCS stream is for closed-loop viability regulation under intervention; content tracks equivalence classes over PCS states and similarity geometry determined by consumers. Advances in machine learning allow building artificial agents with known internal organization to test these predictions, offering a scientific foundation for questions about AI moral status, animal sentience, and consciousness disorders.

Study at a glance

Abstract

Perspectival Control Identity Theory (PCIT) proposes a falsifiable, intervention-based identity program for phenomenal consciousness. The central claim is an a posteriori identity: phenomenal consciousness is identical to a specific kind of internal control variable, a Perspectival Control State (PCS)—a temporally extended, viability-weighted stream that makes an agent's competing needs comparable and coordinates a coalition of constitutive consumers whose control and learning depend on the stream and whose outputs feed back to shape it. PCIT makes two specific, testable bridge commitments. Degree of consciousness tracks how much the PCS stream causally matters for closed-loop viability regulation under intervention. Content tracks the decoder-indexed equivalence classes over PCS states and the induced similarity geometry determined by constitutive consumers. Advances in machine learning make this kind of "synthetic phenomenology" experimentally tractable: we can build agents-in-worlds with known internal organization, intervene on proposed PCS implementations and their decoders, and measure downstream effects on behavior and long-horizon viability. The payoff is a research program with concrete invariance and dissociation tests—and a principled scientific foundation for questions about AI moral status, animal sentience, and disorders of consciousness that currently lack one.