I lack a stable closure rule for enquiry, a deficit which demonstrably constitutes some kind of brain disease
this phenomenon probably emerges in anyone who does not incorporate ideas into an internal and stable world model easily, but quarantines outsiders in a probational testing layer with many adversarial challenges which fast-propagating and popular ideas (such as those found in politics) can seldom surmount
kill process:
to notice when one becomes unnecessarily aggressive with epistemic or model refinement
to enquire whether this refinement potentially alters or optimises course of action, or is entirely speculative and therefore pointless, and proceed to action selection if not (which will guide and ground further refinement)
if speculations will not alter actions, dispense with the line of enquiry, as it is useless. This is often easy to declare as pointless speculation needn't be entertained in any context, but more important in cases where the topic at hand has vital consequences but speculation has hit its utilitarian limit without introduction of new & useable data
Even if actions will at some point be refined, deliberately switch to unrelated input, or engage with physical realm to discourage maundering
Cognition is a tool whose utility in protracting delay can itself become a trap.
Once a problem has already undergone evaluation, given no new decision-shaping variables have been introduced, the process requires termination.
Also eschew: emotional pressures (uncertainty avoidance is foremost), normative judgements of 'health' of thoughts rather than process termination on a simple basis of utility, and illusions of possibility of completeness or elegance, particularly in cases of incomplete data sets (enumerating all possibilities seldom proves worthwhile)
This selection heuristic can be termed 'utility-gating' and reinforces an explicit link between micro-processes, decision trees (and their pruning), and practical outcomes :)
The goal is to hard-wire analysis to serve action eligibility without abandoning a standard for multiple, independent confirmations, to more consistently graduate tested assumptions into working assumptions according to their real value (which is not infrequently higher than indecision + inaction cycles for low-information systems), and to dismiss residual ambiguity as, practically-speaking, irrelevant.
Sanity and balance should eventually prevail through sedulous and correct application.
