When to Wait: The Informational Value of Time Against Consolidation and Erosion in Early Suppression

This is a preprint. It has not undergone peer review. Should treatment of a symptom begin now, or should the case be observed longer? The question is usually framed on one side: early intervention is defended because delay lets a disorder consolidate, watchful waiting because early suppression hides what the symptom would have said about its cause. We argue that these are two sides of one time axis and that a decision rule must weigh both. On one side, waiting has an informational value: while the phenomenology remains observable, onset latency, trajectory and stability accumulate evidence about the mechanism. On the other side, time has two costs. Under dopaminergic control of induction, a trace may cross a barrier in the consolidated synaptic weight beyond which it maintains itself without drive: this is the cost of delay, consolidation. Suppressing the dopamine-dependent drive slows consolidation, but it also erodes the functional traces that depend on the same drive, and spares those already consolidated: this is the cost of suppression, erosion. Suppression therefore exchanges one clock for another. We state a rule of one-step-ahead sequential stopping that sets the informational value of waiting against these costs, and we show, in a simulation in arbitrary units, that the clock exchange can occur in a model taken from a deposited framework. The argument is conditional on three premises (the precision ratio κ, dopaminergic control of induction, bistable maintenance). It gives no criterion for the individual case, and the operator that converts informational gain into a rate is left undefined. Erosion is progressive, and its speed depends on the regime of the trace. We derive one prediction that can fail, a bimodal outcome by band of delay. No empirical data were analysed.