Do leading theories of consciousness target the same empirical data?

Ascertain whether neuroscientific theories of consciousness—including global neuronal workspace theory, higher-order theories, integrated information theory, and self-model theories—are aimed at explaining the same empirical data, rather than targeting disparate phenomena, to enable meaningful theory comparison.

Background

The paper argues that many prominent theories of consciousness have been developed around specific paradigms and subsets of phenomena (e.g., contents, state, self), which complicates direct empirical comparison. This fragmentation raises concerns about whether different theories even address the same empirical observations.

Clarifying whether these theories target overlapping empirical datasets is necessary to evaluate and compare their explanatory power and predictions, a key motivation for developing a minimal theory under the active inference framework.

References

This state of affairs thus poses a double challenge: not only are theories of consciousness difficult to arbitrate between - on the basis of empirical evidence (Yaron et al., 2022) - it is also sometimes not clear whether such theories are aiming to explain the same empirical data (Seth & Bayne, 2022).

— On the Minimal Theory of Consciousness Implicit in Active Inference (2410.06633 - Whyte et al., 9 Oct 2024) in Section 1. Introduction