conviction
the economic commitment a neuron places on a single cyberlink: how much capital it locks behind one specific assertion. conviction is the per-link counterpart of will — where will is the broad budget auto-distributed across every link a neuron makes, conviction is the precise weight directed at one edge. together they are the two write-paths that sum into attention, the input tru reads to build effective adjacency.
conviction is made concrete as a box — the pair $(\tau, a)$ of a token denomination and an amount bound to the cyberlink. cybergraph carries the box record and its value mechanics; tru reads its magnitude as the conviction weight on the edge.
will and conviction
| path | scope | mechanism | tunes |
|---|---|---|---|
| will | broad | locked balance auto-distributed across all of a neuron's links | the baseline strategy |
| conviction | per-link | a box $(\tau, a)$ locked into one specific edge | the precise bet |
a neuron sets a default posture with will, then expresses where it is most certain by raising conviction on particular links. attention on an edge is the will-derived share plus the per-link conviction, before tru weights it by karma and ICBS price.
the conviction spectrum
| box value | meaning |
|---|---|
| $a = 0$ | bare assertion — structural presence, no economic exposure |
| $a$ small | low conviction — the neuron acknowledges the connection but risks little |
| $a$ large | high conviction — the neuron bets real capital that this link matters |
| $a \to$ burn | permanent conviction — tokens destroyed for an eternal cyberlink |
conviction is expressed by how much and how long capital is tied up, not by destroying it — a box is recoverable (withdraw) while the structural assertion it backed remains. burning is the limit case where conviction is made irreversible.
what conviction weighs
conviction is one of the three factors in effective adjacency (see focusing):
$$A^{\text{eff}}_{pq} = \sum_{\substack{\ell \in L \\ \text{src}(\ell)=p,\;\text{tgt}(\ell)=q}} a(\ell)\cdot\kappa(\nu(\ell))\cdot f\big(m(\ell)\big)$$
$a(\ell)$ is the conviction (box magnitude); $\kappa$ is the neuron's karma; $f(m)$ is the market inhibition multiplier. a large conviction behind a low-karma neuron, or on a link the market disbelieves, contributes little — all three must align. this is what makes a cyberlink a costly signal: conviction spent on one claim is conviction withheld from every other.
conviction is distinct from valence: valence is the epistemic prediction ($v \in \{-1,0,+1\}$), conviction is the economic depth ($a \in \mathbb{F}_p$, stake in the smallest token unit — no floats, see arithmetic). a neuron can hold high conviction with any valence — betting heavily that a link is true, or betting heavily while predicting the market will judge it false.
see will for the broad budget · box for the concrete container and its lifecycle · attention for how will and conviction combine · valence for the orthogonal epistemic field · focusing for effective adjacency.
discover all concepts