Back to Compare

Theory Selection Criteria

How we decide which theories to include, which to exclude, and why transparency about these choices matters more than pretending to be neutral.

Our position: We prioritize testable theories over metaphysical frameworks. This is a deliberate choice.

The Three Gates

Every theory must pass three gates to be included. Without defensible criteria, every exclusion becomes political. With criteria, it becomes a conversation about whether the theory meets the standard.

1. Self-Nomination

The framework must explicitly present itself as a theory of consciousness. If the authors don't call it a consciousness theory, we don't interpret on their behalf. This prevents us from co-opting models that were designed for other purposes.

2. Peer Review

There must be at least one substantive peer-reviewed publication. Not necessarily Nature, but the field must have had a chance to scrutinize the theory. Preprints and blog posts alone are not sufficient.

3. Testability

The theory must make empirically testable predictions — predictions that could, in principle, be wrong. This is the line between a theory and a philosophical position. If you can't name a prediction that could fail, it's a position, not a theory.

What We Include and Why

The following 12 theories currently pass all three gates:

TheoryWhy It's Included
Integrated Information Theory (IIT)Mathematically precise predictions about integrated information; extensive peer-reviewed literature
Global Workspace Theory (GWT)Testable via masking, attentional blink, and inattentional blindness paradigms; decades of empirical work
Higher-Order Thought Theory (HOT)Predicts dissociations between first-order and higher-order representations; active experimental program
Predictive Processing (PP)Generates testable predictions about prediction error and conscious perception; growing empirical base
Perceptual Control Theory (PCT)Explicitly models conscious control of perception; testable control-hierarchy predictions
Illusionism (ILLU)Makes the testable prediction that introspective reports systematically misrepresent underlying processes
Orchestrated Objective Reduction (Orch-OR)Specific predictions about quantum coherence in microtubules; peer-reviewed since the 1990s
Dual-Level Causal Theory (DLCT)Testable predictions about causal relationships between neural levels; peer-reviewed publications
Irruption Theory (IRRUP)Predicts specific conditions under which consciousness emerges; published in peer-reviewed venues
Free Energy Principle (FEP)Generates predictions about surprise minimization and conscious inference; extensive literature
Recurrent Processing Theory (RPT)Specific predictions about recurrent neural activity and visual awareness; well-established experimental evidence

What We Exclude and Why

Exclusion is not a judgment of intellectual value. Some of the most important ideas in consciousness studies don't meet our criteria — because our criteria are designed for a specific purpose: enabling empirical comparison.

Applying the Same Standard to Everyone

These criteria must apply uniformly — including to theories associated with this project's contributors. If a theory doesn't pass the gates, it doesn't pass, regardless of who proposed it. This is what makes the criteria credible.

How to Propose a Theory

We welcome proposals for new theories to include. To propose a theory, it should meet the three gates:

  1. The theory explicitly self-nominates as a theory of consciousness
  2. At least one substantive peer-reviewed publication exists
  3. The theory makes empirically testable predictions

If you believe a theory meets these criteria, you can submit it through the Upload Theory page (requires sign-in), or open an issue on our GitHub repository. Include references to the relevant publications and a brief description of the theory's testable predictions.

Why Transparency Matters

Nobody believes a consciousness theory comparison tool is neutral. Pretending to be neutral invites attacks. Being transparent about our choices invites productive engagement.

The question isn't “are we objective?” — nobody is. The question is “are we transparent about our framework?” That's what we hope will earn trust.