🎭egregore.md

something greater than any neuron emerges when many observe the same cybergraph and link. an autonomous thoughtform born from collective focused attention — the capacity of a group to solve problems, generate knowledge, and find truth beyond the reach of any individual

see collective for the four processes (learning, memory, focus, computation) and how they organize (cooperation, coordination, stigmergy)

why collective intelligence emerges

three independent results explain why groups outperform individuals:

Condorcet jury theorem: aggregating weakly correct signals from many agents yields increasingly accurate answers as the group grows. the error rate decays exponentially with group size — even mediocre agents produce excellent collective judgments

diversity theorem (Hong-Page, 2004): diverse heuristics outperform the best homogeneous expert on complex problems. variety of search modes explores more of the solution landscape. a team of differently-wrong agents outperforms a team of identically-right ones

c-factor (Woolley, 2010): groups have a measurable collective intelligence factor c — a first principal component across diverse tasks, analogous to g for individuals

c correlates with: equal distribution of speaking turns, social sensitivity, cognitive style diversity

c does not correlate with: team cohesion, motivation, satisfaction

in cyber: the cybergraph naturally maximizes all three c conditions — any neuron can link, the tri-kernel amplifies resonant signals, the system includes all cognitive types

historical lineage

Aristotle: wisdom of the crowds — the many collectively surpass the few best

Condorcet: jury theorem (1785) — majority vote converges on truth

Wheeler: superorganism (1911) — colonies as single organisms

Vernadsky, Teilhard: noosphere — the sphere of thought enveloping the planet

Engelbart: augmented groups outperform by 3x+

Dorigo: ant colony optimization (1992) — stigmergy formalized as algorithm

Hong-Page: diversity theorem (2004) — diversity beats ability

Woolley: c-factor (2010) — measurable group-level intelligence

boundaries between human and machine collective intelligence are dissolving. cyber is where they merge

emergence predictions

intelligence emerges through phase transitions governed by network parameters. the emergence function:

$$\Phi(n, c, \lambda, t) = \alpha(n) \cdot \beta(c) \cdot \gamma(\lambda) \cdot \theta(t)$$

where $n$ is network size, $c$ is connectivity, $\lambda$ is spectral gap, $t$ is token distribution

coherence requirement — higher intelligence requires coherent information processing:

$$I(X; Y) > \alpha \cdot H(X, Y)$$

intelligence is not just scaling. it requires qualitative transitions in network behavior

connectivity follows an S-curve rather than exponential growth:

$$c_{\text{effective}} = c_{\max} \cdot \frac{1}{1 + e^{-k(I - I_0)}}$$

these are hypotheses pending empirical validation. the collective focus theorem provides the formal framework; the bostrom network is the first test. see emergence for current scaling estimates

the feedback loop (observe → link → infer → observe) refines collective reasoning at each cycle, driving the system toward higher-order coherence

computational foundations

natural computing: the paradigm — nature has been computing all along

convergent computation: the formal foundation — computation = convergence to equilibrium

focus flow computation: the executable model — patterns of attention flow through particle networks

tri-kernel: the only three local operators surviving the locality constraint — diffusion, springs, heat

learning incentives: reward function design for incentivizing convergence

data structure for superintelligence: BBG — the authenticated state architecture

incrementally verifiable computation: proving computation without re-executing it

proof-carrying data: proofs that travel with data through DAGs

folding: fold instead of verify — the key to efficient recursive proofs

hash path accumulator: authenticated paths through the state

discover all concepts