Φ (phi) as a measure of consciousness — cause-effect structure
Integrated Information Theory (Tononi 2004, 2014) identifies consciousness with Φ (phi) — the amount of information generated by a system as a whole above and beyond its parts. Φ is computed over the minimum information partition (MIP): the cut of the network that least reduces information integration. High Φ requires both differentiation (many distinct states) and integration (no cut can separate without losing information). This simulation approximates Φ using mutual information: Φ ≈ I(system) − min_partition I(part_A) − I(part_B). Densely and bidirectionally connected networks have high Φ; feedforward-only networks (like serial computers) have low Φ despite complexity.