Can We Test Consciousness Theories on AI? Ablations, Markers, and Robustness

TL;DR

This paper uses artificial agents with specific neural architectures to test and compare key predictions of consciousness theories like GWT, IIT, and HOT, revealing their functional distinctions and interactions.

Contribution

It introduces a synthetic neuro-phenomenology approach, creating agents to experimentally test and dissociate the functional roles of different consciousness theories.

Findings

Self-Model lesion causes blindsight-like behavior, supporting HOT.

Workspace capacity is causally necessary for information access, aligning with GWT.

Broadcast amplification increases internal noise, indicating fragility in GWT-style broadcasting.

Abstract

The search for reliable indicators of consciousness has fragmented into competing theoretical camps (Global Workspace Theory (GWT), Integrated Information Theory (IIT), and Higher-Order Theories (HOT)), each proposing distinct neural signatures. We adopt a synthetic neuro-phenomenology approach: constructing artificial agents that embody these mechanisms to test their functional consequences through precise architectural ablations impossible in biological systems. Across three experiments, we report dissociations suggesting these theories describe complementary functional layers rather than competing accounts. In Experiment 1, a no-rewire Self-Model lesion abolishes metacognitive calibration while preserving first-order task performance, yielding a synthetic blindsight analogue consistent with HOT predictions. In Experiment 2, workspace capacity proves causally necessary for information access: a complete workspace lesion produces qualitative collapse in access-related markers, while partial reductions show graded degradation, consistent with GWT's ignition framework. In Experiment 3, we uncover a broadcast-amplification effect: GWT-style broadcasting amplifies internal noise, creating extreme fragility. The B2 agent family is robust to the same latent perturbation; this robustness persists in a Self-Model-off / workspace-read control, cautioning against attributing the effect solely to $z_{\text{self}}$ compression. We also report an explicit negative result: raw perturbational complexity (PCI-A) decreases under the workspace bottleneck, cautioning against naive transfer of IIT-adjacent proxies to engineered agents. These results suggest a hierarchical design principle: GWT provides broadcast capacity, while HOT provides quality control. We emphasize that our agents are not conscious; they are reference implementations for testing functional predictions of consciousness theories.