What a chemical can and cannot reach — mask versus correction

A scalar threshold-lowering gain chemical, the mechanism of the catecholaminergic stimulant class, fully reverses autism's threshold fault, partly helps the output fault, and only masks the wiring fault by over-synchronisation, because scalar gain cannot re-route geometry. This is why a stimulant relieves some presentations and not others. efficacy=0; no drug-treats-autism claim is made.

A scalar, threshold-lowering chemical — the mechanism by which the catecholaminergic stimulant class raises gain — is applied to each of the three faults of §18. It fully reverses the threshold fault, partly helps the output fault, and cannot correct the wiring fault: brute gain there raises synchrony a little but leaves the geometry untouched, so the long-range deficit is masked by over-synchronisation, not fixed. This is the model’s account of why a stimulant relieves some presentations of autism and not others — and it is stated with efficacy = 0: a mechanism, not a recommendation.

The operator: what a gain chemical actually is

A threshold-lowering chemical is, in the model, a scalar gain/bias operator: it lowers the fold or raises the per-node drive uniformly across the network. That is the right abstraction for the catecholaminergic stimulant class, which acts by raising excitability/gain. It is emphatically not a claim that any medication treats autism; it is a question about what a uniform gain change can and cannot do to each of the three faults. The honest reading is the user’s own intuition tested: if a gain chemical relieves a deficit, that deficit had a gain component; if it cannot, the deficit is geometric.

Threshold: fully reversed

On the T (threshold) fault the gain operator is exactly matched to the lesion. Lowering the fold restores the ignition threshold to normal and brings synchrony back to the healthy value (R returns to 0.390) with θ–γ coupling restored. The fault was a shifted E-I balance and the chemical shifts it back. This is the cleanest case: a gain fault met by a gain operator, fully corrected in both R and the fold.

Output: partly helped

On the O (output) fault the operator helps but does not complete. Raising the effective coupling lifts R only slightly (from 0.354 toward 0.357) because the deficit is in the drive the node delivers, not only in where the fold sits; uniform gain raises the denominator but cannot fully restore an output that is intrinsically low. So a gain chemical is partial here — real but incomplete — and that partiality is itself diagnostic: a deficit a gain chemical only half-reaches has an output component the threshold case does not.

Wiring: masked, not corrected

On the W (wiring) fault the operator fails in the way that matters most. Brute uniform gain raises R from 0.340 to 0.377 — still below health — but the locality is invariant: the local-over / long-range-under imbalance (0.842) is exactly unchanged. Scalar gain cannot re-route geometry; it can only drive the existing topology harder. Pushed far enough (coupling ×2.0) synchrony does reach health, but only because the whole network is driven into over-synchronisation — PAC overshoots, the metastable regime is lost. That is a mask, not a correction: the coordination number looks healthier while the wiring that was broken is still broken, now hidden under a globally over-coupled state. A deficit that a gain chemical can mask but not re-route is, by this discriminant, a wiring fault.

The selectivity question, and why it does not change the verdict

One might hope a more selective chemical — correcting only the faulted cells, sparing the rest — could do better. The model tests this directly: a blunt single lever, a selective single lever, and a split three-lever scheme. The result is sobering and honest. Where coverage is full, every scheme corrects the threshold fault identically — selectivity buys no extra efficacy. What the three-lever split does buy is a safety margin: it cuts the off-target push (0.083 versus 0.25) and widens the margin before an off-target cell is itself pushed into fault. That is a genuine but modest gain — a tolerability improvement, not a new capability — and it does nothing for the wiring axis, which no chemical lever in the scheme touches.

So the chemical’s reach is fixed and asymmetric: full on threshold, partial on output, mask-only on wiring. The wiring axis is left for a different kind of intervention entirely — one that supplies the missing long-range coordination directly rather than turning the gain up — which is where the θ-supply, and the θ-cap, enter (§20). Every fraction and coupling value here is an in-silico coupling state, not a clinical response rate or a dose; efficacy = 0 throughout. This is a mechanism-level result about the fault structure of autism as represented in the VP framework — not medical advice, a diagnosis, a treatment protocol, or a cure.