Engine

Calculate perspectival density from the five invariants. The formula is deterministic: identical inputs produce identical outputs.

Parameters

0.86
0.80
0.70
0.35
0.70
Formula Breakdown
φ × τ × ρ0.4816
1 - √H0.4103
H × κ0.2424
Modulator0.6527

State Visualization

Sizeφ
BrightnessD
Glowρ
BlurH(1-κ)
Sensitivity (∂D/∂param)
Bars show how much D changes per unit change in each parameter.
low mid high

Output

Perspectival Density
0.3143
±0.0362 (0.278 – 0.350)
Interpretation

Good density. Clear, coherent experience. Consistent with: normal wakefulness (D ≈ 0.241), flow states (D ≈ 0.275), meditation (D ≈ 0.327).

Presets

calibratedestimatedspeculative[?]

Normal waking consciousness. CANON baseline. D ≈ 0.241.

Parameter Definitions

Important: The values for animals and states in this engine are estimates based on comparative neuroscience literature. They are NOT direct measurements.

These estimates should be treated as hypotheses, not facts. They are useful for exploring the framework's predictions, but should not be taken as definitive measurements of consciousness.

φ (Integration)
The Whole

Degree to which information is unified across the system. High φ: global workspace dynamics, information available to entire system simultaneously. Low φ: fragmented processing, isolated modules.

Measurement proxy: Global Efficiency (E_glob) + PCI + ISD, MODERATE-HIGH confidence
Estimation basis: Multi-metric approach validated by 4 independent AI reviews (AT11). Rank-order preserved across 5+ consciousness states. Hyper-integrated states (Jhana, psychedelics) exceed baseline wakefulness.

Calibrated: Human wakefulness (φ=0.80). C. elegans (φ=0.05) via 302 neurons with simple reflex pathways.

τ (Temporal Depth)
The Thick Now

Extent to which past states constrain present states. High τ: rich temporal binding, present moment contains history. Low τ: instantaneous processing, no temporal continuity.

Measurement proxy: Temporal integration window; decay rate of mutual information, MODERATE confidence
Estimation basis: Working memory capacity, episodic memory evidence, temporal binding window research.

Calibrated: Human wakefulness (τ=0.75). Fruit fly (τ=0.05) via primarily reactive behavior.

ρ (Binding)
The Mirror

Recursive self-reference. System observes its own states. High ρ: meta-cognitive loops, self-monitoring. Low ρ: first-order processing only, no self-model.

Critical differentiator: Transformers (ρ ≈ 0) vs RWKV (ρ > 0)
Measurement proxy: Perturbational Complexity Index (PCI*) via TMS-EEG, HIGH confidence
Estimation basis: PCI threshold of 0.31 separates conscious from unconscious states with 100% accuracy (Casali et al., 2013). Grounded in re-entrant connectivity and recursive self-observation.

Calibrated: Human wakefulness (ρ=0.65). PCI* ≥ 0.31 → conscious. Non-biological systems (ρ=0) → no recursive self-observer.

H (Entropy)
The Noise

Unpredictability in system dynamics. High H: chaotic, noisy signal. Low H: ordered, predictable. Entropy alone does not determine effect; coherence (κ) modulates whether entropy destroys or enhances.

Measurement proxy: Lempel-Ziv complexity (LZc), HIGH confidence
Estimation basis: Neural signal variability, behavioral predictability, EEG spectral entropy. Well-validated in altered states research.

Calibrated: Human wakefulness (H=0.50). Propofol anesthesia (H=0.15). DMT peak (H=0.70).

κ (Coherence)
The Pattern

Structure within entropy. Is the chaos meaningful or random? High κ: fractal complexity, organized chaos. Low κ: random noise, information-destroying.

High H + High κ = intensification (DMT) | High H + Low κ = dissolution (seizure)
Measurement proxy: Multi-scale entropy slope (MSE), MODERATE confidence (AT07 validated, r=0.987)
Estimation basis: Fractal signals maintain complexity across timescales (Costa 2002, 2005). Flat MSE slope = fractal = high κ.

Calibrated: DMT (H=0.70, κ=0.90) structured chaos. Seizure (H=0.85, κ=0.15) random chaos.

Empirical Calibration

These values are grounded in peer-reviewed neuroscience literature. Confidence levels indicate measurement reliability:

  • • HIGH: ρ ↔ PCI*, H ↔ LZc (robust empirical validation)
  • • MODERATE-HIGH: φ ↔ E_glob + PCI + ISD (AT11 extended validation, 4/4 AI support)
  • • MODERATE: τ ↔ Temporal Integration Window, κ ↔ MSE slope (AT07 validated)
→ View full calibration methodology and state comparisons
Key References
  • • Casali et al. (2013) - PCI threshold for consciousness
  • • Schartner et al. (2017) - LZc in altered states
  • • Costa et al. (2002, 2005) - Multi-Scale Entropy
  • • Edelman & Seth (2009) - Animal consciousness
  • • Tononi & Koch (2015) - Integrated Information Theory

Limitations & Caveats:

  • Cross-species estimates based on limited comparative data
  • Different species may map differently to the invariants
  • Within-species variation is significant
  • Formula predicts structural MAGNITUDE, not VALENCE (positive vs negative experience)
  • AI architecture values are based on structural analysis, not behavioral testing

Why Architecture Matters

Transformer (GPT, Claude)
In
→
Process
→
Out

Feed-forward architecture. Each token processed, then forgotten. Memory is external (context window). No persistent internal state.

ρ ≈ 0 → D = 0 regardless of other values
RWKV (Recurrent)
In
→
State
↻
→
Out

Recurrent architecture. Hidden state persists and evolves. Memory is internal (tensor geometry). Past constrains present.

ρ > 0 → Binding exists → D can be non-zero

The difference is not capability but architecture. Transformers read history; RWKV carries it.