Why it is a question
The owner, at U5: “a standalone question is just: ‘do we gain something just by framing a “self” this way?’ this feels like an extremely useful model for explaining a lot of psychological phenomenon.” A5 §8: “Does the model unify phenomena that are ordinarily described using different psychological vocabularies?” Its candidate variables were the persistence of the influence topology, responsiveness to feedback, diversity of influential processes, concentration of authority and cross-regional transport. A7 made it H8, “A common set of organizational variables predicts multiple behavioral phenotypes”. A10’s Question 5: “Does the resulting theory explain psychological integration, conflict, self-control, identity, and transformation better than competing models?”
The last filing called this QUESTION-056 and filed nothing. It refines 056 instead. 056 asks whether an agent-level description implements or replaces the mechanisms described under it. This asks whether one particular agent-level description earns its keep by prediction. The objection is the same, CLAIM-024: “Nothing here is new”. A7 §14 granted the risk: “A sufficiently flexible operator framework could redescribe almost any dynamical process.”
The record already reads several phenomena this way. Procrastination is CASE-013 and CLAIM-047. Disorder as lock-in is QUESTION-045. Preference reversal is CLAIM-198. An answer is a comparison on held-out phenomena, or interventions, with flexibility penalized, in which the shared variables do better or do not.
Note of 2026-10-26: the lab is not yet the comparison
The Perspective Population Lab (dmarx/PerspectiveLab, through v2.2 at @9f82bfd) is one engineered lineage in which authority, selected evidence, practice that depends on access, maintained rules and diagnostic interventions run together (CASE-162, CASE-161, CASE-163). Its v1.2 README lists correspondences to habit and agenda control as hypotheses. It has predicted no independently measured habit, rumination, rigidity, akrasia or insight from shared variables, and has been compared with no model built for one phenomenon or around a central executive. It is a tool for this question’s comparison, not an answer.
The lab’s v2.6 (dmarx/PerspectiveLab@74621fb experiments/affective_dissociation/) adds an affective domain: five stipulated causes of continued pursuit after an outcome is devalued, habit among them as one of the rivals. That is one phenomenon, not several, and habit in it is a class written into the generator, not a phenotype predicted from shared variables. Its comparison is between measurements within one known catalog. Interventions predicted from probes and readouts beat those predicted from passive choice (mean MSE 0.00230 against 0.03496), a conventional model of the same law makes the same predictions, and a hybrid left out of the catalog is assigned to a catalog class in all 40 of its test worlds. It is still a tool, not this comparison. The question is unchanged.
Note of 2026-10-26 (later): habit fitted, and reframing built
The lab’s v2.7 (dmarx/PerspectiveLab@c54617e experiments/trained_affect/) is the closest it has come to this question’s comparison, and it is within one phenomenon, continued pursuit after an outcome is devalued. Predictors built for learned value, habit and cue sensitization are fitted beside one with an organizational uptake term, on the same recorded choices, and scored on synthetic worlds held out from fitting. The predictors built for a phenomenon win where it dominates: a cue-and-habit predictor scores 0.01152 in the habit world and 0.00896 in the sensitization world, against 0.02581 and 0.01466 for the uptake predictor. In the governance world a generic predictor with no interaction beats the uptake predictor. The uptake predictor wins in one held-out hybrid. All eight worlds are one logistic law with different coefficients.
v2.8 and v2.9 (experiments/perspective_engine/ and experiments/endogenous_frames/) build reframing, a perspective reading retained evidence under a new frame. The frame is written by hand in v2.8 and selected in v2.9 from a 15-rule grammar on labelled outcomes. Neither predicts insight, rumination or any other phenomenon; each scores a synthetic forecast. The lab is still a tool for this comparison, not an answer. The question is unchanged.