Virtuous Feedback Loop

Why do some cycles create output while others increase the ability to choose what happens next?

A Validated Virtuous Feedback Loop (VVFL) joins a values-aligned setpoint to a reality-tested gauge. Each cycle improves both action and future perception. Its outcome is greater situational wisdom and agency.

Three Resolutions

The same mechanism appears at three useful resolutions.

Consciousness atom

PERCEIVE -> QUESTION -> ACT -> changed PERCEPTION

Perception supplies a frame. Questioning tests the frame and chooses a consequential gap. Action lets reality answer. A cycle is not closed until that answer changes what can be perceived next.

Operational FLOW

INTENTION -> CAPTURE -> PRIORITISE -> COMMIT-ACTION
          -> MEASURE -> QUESTION-LEARN -> EVOLVE

Stage	Contract	Proof output
Intention	Name the valued future and authority boundary.	Setpoint worth serving
Capture	Catch the signal without judging it.	Source-faithful signal
Prioritise	Screen by purpose, then choose what wins attention.	Decision and accepted loss
Commit-Action	Run one bounded quest to its review point.	Changed real-world state
Measure	Compare dated evidence with the durable setpoint.	Gauge reading and variance
Question-Learn	Ask what the result reveals about frame, method, or control.	Evolved question
Evolve	Change the next input, setpoint, control, or capability.	Higher starting baseline

Physical eye

hopper -> filter -> pump -> gauge -> controller -> return line

• The hopper captures widely, then purpose screens the intake.
• The filter chooses one priority and closes debate for the action leg.
• The pump commits attention and moves the chosen quest.
• The gauge reads dated reality against a durable standard.
• The controller questions the frame and changes the system, not merely the wording.
• The return line carries proof into the next cycle as changed perception.

The mark is therefore an eye, a question mark, and a control system. The operating instruction is: see, question what seeing hides, act, receive reality, and see differently.

Entry Conditions

Enter where evidence says the loop is blocked. Do not force every job to restart at Capture.

Entry condition	Route	Stop condition
Raw signal, problem, possibility, success, or failure	Capture, then Purpose	Signal is kept, parked, or discarded by intent before ranking
Desired future but no falsifiable move	Inner Space, then Priorities	One quest, proof gate, and accepted loss are explicit
Position is unclear	Map Reality or Navigation	Options, constraints, bearing, and missing evidence are visible
Condition cannot carry committed action	Flow Coach	State and work block match the challenge
Priority is settled	Attention and Value	One action reaches its declared review point
Outcome exists but meaning is unclear	Standards	Dated gauge, durable setpoint, and variance remain distinct
Variance or surprise exists	Evolve	Controller, input, setpoint, or capability changes

Human authority stops remain binding. The loop may recommend, route, and expose variance. It does not silently change a person's values, irreversible commitment, or declared setpoint.

Agency Gauge

Measure each quest on two coordinates:

• Position: Did options, relationships, resources, information, or proximity to the desired future improve?
• Condition: Did energy, capability, conviction, attention, or readiness improve?

An action can ship while both coordinates stay fixed. That is useful output, but it is not evidence that agency compounded.

Failure Modes

Loop	Setpoint	Gauge	Result
Runaway	Missing or hidden	Missing	Amplification without consequence control
Corrective	Fixed	Present	Stability around an inherited target
Virtuous	Values-aligned and revisable	Reality-tested	Wisdom and agency improve the next cycle

Common breaks:

• Purpose bypass: a raw signal reaches priority without an intent screen.
• Fantasy: a vivid picture has no falsifiable quest or reality gauge.
• Condition blindness: attention is committed when the actor cannot carry it.
• Gauge drift: a dated reading overwrites the durable standard.
• Controller theatre: reflection produces nicer prose but changes no input, control, setpoint, or capability.
• Open return line: proof is recorded but never changes the next question.

Method

1. State the valued setpoint and who retains authority over it.
2. Name the current entry condition and active stage.
3. Route through only the stages needed to reach one measured quest.
4. Compare the dated gauge with the durable setpoint; name the variance.
5. Change the next question, input, control, or capability before declaring the loop closed.

The cheapest falsifier is a teach-back: the actor can name position and condition, the chosen quest, its proof gate, and how its result changes the next cycle.

Proof of Done

The loop is closed when the dated gauge is compared with the durable setpoint, the variance reaches the controller, and the next cycle begins with a changed question, input, control, setpoint, or capability.

Master the Inner Loop → Master the Game

▽DreamerQuestions→?CoachPrinciples→▷EngineerProtocols→○RealistStandards→•StorytellerPlatform

The Tight Five →

Context

• depends-on Inner Space — supplies the experiential entry into conscious perception and rehearsal.
• applies-to Agency — makes position and condition actionable and measurable.
• depends-on Purpose — screens the setpoint through meaning, intention, conviction, drives, and flow.
• pairs-with Tight Five Loops — compresses question, principle, protocol, standard, and platform.
• measured-by Standards — keeps the setpoint and gauge distinct.

Questions

What changed in the controller, not only in the output?

• Which stage accepted the last valid handoff?
• What did reality reveal that the original picture hid?
• How will the next cycle begin with better perception?

Changes my mind: A completed action cannot be traced through a measured handoff into changed future perception.

Next question: Which controller change will make the next consequential question easier to see?