Info detectors: activation, hold, and reset algorithms

This document lists all info-picture detectors, their numeric fields in InfoPicture, and three condition groups for each:

• Activation — when the detector becomes active (field ≠ 0).
• Hold — when the detector stays active.
• Reset — when the detector must return to 0.

General principles

• Detector values are integers (or floats for “levels”). Zero always means “inactive / not detected”.
• Each InfoDet_updateXxx(...) in info_detectors.js:
  • takes parameters from the “information gathering” site (vitals, contexts, EM effects, waiting state, etc.);
  • itself decides whether to set a non-zero value or reset to 0 from the three conditions.
• Callers must pass fresh data on every material change (pulse, OR, EM frame close, wait start/end, etc.), even when “nothing happened” — so the detector can reset itself.

Detectors are loosely grouped as:

• Snapshot (state) — can be recomputed from the current state snapshot (BadNormWell, BasicContextsActived, semantics, etc.).
• Event — capture a moment/fact in time (wait ended without answer, novelty spike, etc.). Hold and reset matter especially for these.

1. Comprehension mode, Theme, Situation

1.1. Target comprehension mode (main consciousness cycle + “Mode” cell)

• Source: main cycle thinkingMode; UI — Consciousness_getMainCycleThinkingMode().
• Activation:
  • When starting a reaction/automatism that needs target comprehension (e.g. OR with semantically salient stimulus or starting a comprehension cycle for unsolved_problem): call InfoDet_updateTargetMode(true).
• Hold:
  • While the target comprehension cycle runs or automatism consequence evaluation (waiting period) is active.
• Reset:
  • When the target cycle ends (problem solved or deemed irrelevant) and no new target comprehension is needed → InfoDet_updateTargetMode(false).

1.2. Current theme (InfoPicture.themeId, “Theme” cell)

• Field: themeId: 0 — no theme, >0 — theme image ID from the situations tree.
• Activation:
  • On each SituationsTree_onPerceptionTreeActivated() call, best‑match ready theme image for current mode and tree.BaseID/EmotionID/ActivityID:
    • if a ready image exists (ready and score>0) → InfoDet_updateCurrentTheme(themeId).
• Hold:
  • While the current (Base, Emotion, Activity) combo stays closest to the same ready theme image.
• Reset:
  • If on the next situations-tree activation:
    • no ready images (all count ≤ 3),
    • or no ready image yields non-zero score,
      → InfoDet_updateCurrentTheme(0).

1.3. Current situation (InfoPicture.situationId, “Situation” cell)

• Field: situationId: 0 — none, >0 — situation image ID from the situations tree.
• Activation:
  • In SituationsTree_onPerceptionTreeActivated() a component vector is built from info-picture detector fields; among ready images pick best‑match:
    • if score>0 and ready=true → InfoDet_updateCurrentSituation(situationId).
• Hold:
  • While the current detector set stays closest to the same situation.
• Reset:
  • On the next call, if:
    • no ready situations for current mode,
    • or score==0 for all,
      → InfoDet_updateCurrentSituation(0).

2. Vitals and basic contexts

2.1. Critical vital (InfoPicture.criticalVital, “Crit. vital” cell)

• Field: criticalVital:
  • 0 — no critical vital;
  • >0 — vital ID whose value is >70 and maximal among all.
• Activation:
  • In updateImportanceMetrics(badVitals) or getActiveBasicContexts(cpuls):
    • scan VitalsArr and find a vital with value > 70 and maximum value;
    • if found → InfoPicture.criticalVital = vitalId.
• Hold:
  • While repeated calls still have a vital with value > 70; on “leader” change update detector to new ID.
• Reset:
  • If on the next call no vital exceeds 70 → criticalVital = 0.

2.2. Bad state (InfoPicture.badState, “Bad” cell)

• Field: badState: 0 — not Bad, 1 — Bad (BadNormWell=1).
• Activation:
  • In InfoDet_updateBadState(lastBadNormWell, badNormWell):
    • when badNormWell === 1 → badState = 1.
• Hold:
  • While BadNormWell stays 1.
• Reset:
  • On any BadNormWell != 1 in the next call → badState = 0.

2.3. Sudden worsening (InfoPicture.suddenChange, “Sudden chg.” cell)

• Field: suddenChange:
  • 0 — none;
  • 1 — moment of switch to Bad (per PDF); a more detailed level may be stored.
• Activation:
  • In InfoDet_updateBadState(lastBadNormWell, badNormWell):
    • if lastBadNormWell !== 1 and badNormWell === 1 → suddenChange = 1.
• Hold:
  • Event detector: may keep 1 for one comprehension-cycle step or one/few pulses (your choice) if the event must be “seen”.
• Reset:
  • On next call without another switch to Bad → suddenChange = 0.

2.4. Play mode (InfoPicture.playMode, “Play” cell)

• Field: playMode: 0 — off, 1 — Play basic context active.
• Activation:
  • In getActiveBasicContexts after BasicContextsActived:
    • if BasicContextsActived contains context id == 11 → playMode = 1.
• Hold:
  • While Play stays in BasicContextsActived.
• Reset:
  • On next call, if Play context inactive → playMode = 0.

2.5. Learning mode (InfoPicture.learningMode, “Learning” cell)

• Field: learningMode: 0 — off, 1 — learning mode active.
• Activation:
  • In InfoDet_updateLearningAndSearch(BasicContextsActived, VitalsArr):
    • if basic context Curiosity (id=12) is active or vital “Interest” has value > 70 → learningMode = 1.
• Hold:
  • While at least one holds (Curiosity or high Interest).
• Reset:
  • With no Curiosity and Interest≤70 → learningMode = 0.

2.6. Search interest (InfoPicture.searchInterest, “Search” cell)

• Field: searchInterest: 0 — off, 1 — Interest vital >70.
• Activation:
  • In InfoDet_updateLearningAndSearch:
    • if vital with Id “Interest” (agreed id, e.g. 10) has value > 70 → searchInterest = 1.
• Hold:
  • While Interest>70.
• Reset:
  • On next call when Interest≤70 → searchInterest = 0.

2.7. Fear (InfoPicture.fear, “Fear” cell)

• Field: fear: 0 — off, 1 — Fear basic context active.
• Activation:
  • In getActiveBasicContexts:
    • if BasicContextsActived contains context id=6 (Fear) → fear = 1.
• Hold:
  • While Fear context stays active.
• Reset:
  • On next call without id=6 → fear = 0.

2.8. Aggression (InfoPicture.aggression, “Aggression” cell)

• Field: aggression: 0 — off, 1 — Aggression basic context active.
• Activation:
  • If BasicContextsActived contains context id=7 (Aggression) → aggression = 1.
• Hold:
  • While Aggression context active.
• Reset:
  • On next call without id=7 → aggression = 0.

2.9. Defense (InfoPicture.defense, “Defense” cell)

• Field: defense: 0 — off, 1 — defense mode active.
• Activation:
  • If fear == 1 or aggression == 1 → defense = 1.
• Hold:
  • While fear or aggression active.
• Reset:
  • When fear == 0 and aggression == 0 → defense = 0.

3. Semantics and orientation reflex

3.1. Console stimulus (InfoPicture.consoleStimulus, “From console” cell)

• Field: consoleStimulus:
  • 0 — no stimulus;
  • >0 — winning stimulus image FinalImageId.
• Activation:
  • In runOROnStimulusActivation after choosing winnerId:
    • InfoDet_updateConsoleStimulus(winnerId).
• Hold:
  • May hold until next OR (last salient stimulus).
• Reset:
  • On explicit absence of active stimuli (all cleared) or by policy on branch change:
    • call InfoDet_updateConsoleStimulus(0).

3.2. High semantic importance (InfoPicture.highSemanticImportance)

• Field: highSemanticImportance: 0 — none; >0 — meanImportance under current conditions.
• Activation:
  • In orientation_reflex.js after semWinner:
    • if meanImportance > 0 → highSemanticImportance = meanImportance.
• Hold:
  • While current salient stimulus (winnerId) and conditions (well, emotionId) unchanged.
• Reset:
  • On stimulus/context change (new OR) or no semantics → highSemanticImportance = 0.

3.3. High-importance object (InfoPicture.highImportanceObject)

• Field: highImportanceObject: 0 — none; >0 — salient image ID (FinalImageId) with |semantics|>5.
• Activation:
  • In orientation_reflex.js:
    • if Math.abs(meanImportance) > 5 → highImportanceObject = winnerId.
• Hold:
  • While same stimulus and conditions similar.
• Reset:
  • On stimulus/context change or |meanImportance| ≤ 5 → highImportanceObject = 0.

3.4. Image novelty (InfoPicture.imageNovelty)

• Field: imageNovelty: 0 — none; >0 — novelty value if above threshold.
• Activation:
  • In orientation_reflex.js after novelty:
    • if novelty > threshold (from code) → imageNovelty = novelty.
• Hold:
  • Until next OR.
• Reset:
  • On new OR when novelty ≤ threshold → imageNovelty = 0.

3.5. Conflicting semantics (InfoPicture.conflictingSemantics)

• Field: conflictingSemantics: 0 — none; 1 — conflicting semantics detected.
• Activation:
  • Semantics analysis (understanding module) finds:
    • mean importance near 0 with large samplesCount,
    • under different conditions both sizable positive and sizable negative values.
  • Then → conflictingSemantics = 1.
• Hold:
  • Until conflicting-semantics situation resolves.
• Reset:
  • If next analysis shows no conflict signs → conflictingSemantics = 0.

4. Action effects and automatisms

4.1. Positive / negative action effect (positiveActionEffect, negativeActionEffect)

• Fields:
  • positiveActionEffect: 0 or >0 — magnitude of last positive effect;
  • negativeActionEffect: 0 or <0 — magnitude of last negative effect.
• Activation:
  • In closeLastFrame (episodic_memory.js) after computing last.effect:
    • if effect > 0 → positiveActionEffect = effect, negativeActionEffect = 0;
    • if effect < 0 → negativeActionEffect = effect, positiveActionEffect = 0;
    • if effect == 0 → both 0.
• Hold:
  • Holds the last completed EM frame’s value.
• Reset:
  • On a new frame without effect or explicit history reset → both = 0.

4.2. Negative motor automatism effect (negativeMotorEffect)

• Field: negativeMotorEffect: 0 — none; <0 — negative z during motor automatism waiting.
• Activation:
  • When a motor automatism fires and z = getDiffImportance() is measured in the waiting period:
    • if z < 0 → negativeMotorEffect = z.
• Hold:
  • Until a new z is taken for another motor automatism.
• Reset:
  • On next measurement with z >= 0 or waiting ends without negative effect → negativeMotorEffect = 0.

4.3. Negative mental automatism effect (negativeMentalEffect)

• Same as motor, for automatisms started from comprehension (mental):
  • if z < 0 in waiting → negativeMentalEffect = z,
  • on absence/completion → 0.

4.4. Low automatism usefulness (lowAutomatismUsefulness)

• Field: lowAutomatismUsefulness: 0 — none; |value| ≤ 1 — low usefulness.
• Activation:
  • On each new effectSample written to automatism usefulness:
    • if 0 < |effectSample| ≤ 1 → lowAutomatismUsefulness = effectSample.
• Hold:
  • Until the next sample shows a normal absolute effect (>1).
• Reset:
  • When |effectSample| > 1 or effectSample == 0 → lowAutomatismUsefulness = 0.

4.5. Suitable automatism found (foundSuitableAutomatism)

• Field: foundSuitableAutomatism: 0 — none; 1 — comprehension cycle in target mode found and uses an automatism.
• Activation:
  • When a suitable automatism is chosen and started in target mode → foundSuitableAutomatism = 1.
• Hold:
  • While the comprehension cycle keeps using that automatism.
• Reset:
  • When the cycle ends or the automatism is abandoned → foundSuitableAutomatism = 0.

5. Waiting and operator responses

5.1. No reaction to stimulus (noReactionToStimulus)

• Field: noReactionToStimulus: 0 — none; 1 — no reaction to stimulus recorded.
• Activation:
  • In EpisodicMemory_onPulse, when last.actionImageId === null and waiting expired without reaction → noReactionToStimulus = 1.
• Hold:
  • Until the next wait starts or explicit reset.
• Reset:
  • On reaction or starting a new wait → noReactionToStimulus = 0.

5.2. Operator ignoring (operatorIgnoring)

• Field: operatorIgnoring: 0 — none; 1 — operator ignored the action.
• Activation:
  • In EpisodicMemory_onPulse, if last.actionImageId != null (there was reaction/automatism) and waiting ended without response stimulus → operatorIgnoring = 1.
• Hold:
  • Until a response stimulus arrives in a new waiting period.
• Reset:
  • In orientation_reflex.js, on response stimulus during waiting (EpisodicMemory_isInWaitingPeriod):
    • operatorIgnoring = 0.

5.3. Operator answer received (operatorAnswerReceived)

• Field: operatorAnswerReceived: 0 — none; 1 — operator answer received.
• Activation:
  • In orientation_reflex.js, when OR fires during waiting and the frame closes with response stimulus → operatorAnswerReceived = 1.
• Hold:
  • Until the next waiting frame or explicit reset.
• Reset:
  • On starting a new waiting period or processing the answer → operatorAnswerReceived = 0.

5.4. Active answer waiting (activeWaitingAnswer)

• Field: activeWaitingAnswer: 0 — none; 1 — automatism result waiting period active.
• Activation:
  • When creating an EM frame tied to an automatism (especially mental, started from comprehension) → activeWaitingAnswer = 1.
• Hold:
  • While waiting is not finished (answer/timeout).
• Reset:
  • On frame close (with or without answer) → activeWaitingAnswer = 0.

5.5. Run of episodes without effect (seriesWithoutEffect)

• Field: seriesWithoutEffect: 0 — none; >0 — length of latest run of frames with effect == 0.
• Activation / hold:
  • EM logic increments on each closed frame with effect == 0 and clears on non-zero effect; on each length change call InfoDet_updateSeriesWithoutEffect(length).
• Reset:
• On non-zero effect (effect != 0) or explicit EM reset → seriesWithoutEffect = 0.

6. Comprehension and hard problems

These detectors depend on the comprehension cycle (conscience_level_2) and unsolved_problem handling.

6.1. Teacher learning (teacherLearning)

• Field: teacherLearning: 0 — off; 1 — teacher-learning mode on.
• Activation:
  • When comprehension records an unsolved problem for a given image that stays unsolved through an uninterrupted attempt loop (several target-comprehension steps in a row).
• Hold:
  • While “hand initiative to operator” mode runs: system tracks actions and elicits responses.
• Reset:
  • On problem solved (operator answer or internal fix) or step limit reached → teacherLearning = 0.

6.2. Dissatisfaction with the status quo (dissatisfaction)

• Field: dissatisfaction: 0 — none; 1 — dissatisfaction present.
• Activation:
  • When Calm basic context is active, no salient stimuli, and a passive-mode comprehension cycle finished (reached end without new ideas/changes).
• Hold:
  • Until external conditions change or a new stimulus/interest appears.
• Reset:
  • On new stimuli, context change, or target cycle start → dissatisfaction = 0.

6.3. Misunderstanding (misunderstanding)

• Field: misunderstanding: 0 — none; 1 — misunderstanding state.
• Activation:
  • If comprehension reaches the end in target mode or iteration steps exceed a limit (e.g. 10) and the problem is unsolved.
• Hold:
  • Until the cycle restarts with different conditions or the problem is solved.
• Reset:
  • On problem solved or cycle restart with new configuration → misunderstanding = 0.

6.4. Doubt in default automatism (doubtInDefaultAutomatism)

• Field: doubtInDefaultAutomatism: 0 — none; 1 — doubt present.
• Activation:
  • When level-2 comprehension starts in target mode for a branch where the default automatism gives ambiguous/unacceptable results.
• Hold:
  • While level 2 considers alternatives to the default automatism.
• Reset:
  • On choosing a new default or confirming the old → doubtInDefaultAutomatism = 0.

6.5. Hard problem (hardProblem)

• Field: hardProblem: 0 — none; >0 — difficulty level/counter for the image’s problem.
• Activation / hold:
  • Per image (stimulus/situation) count cases when:
    • unsolved_problem exists for that image,
    • target cycle ends without resolution,
    • comprehension step count exceeds threshold.
  • On each such case increment hardProblem for the current image and pass to InfoDet_updateHardProblem(level).
• Reset:
  • On problem solved (image marked resolved) → level cleared, image removed from “hard problems” list, hardProblem = 0.

This document defines target algorithms for all info detectors; InfoDet_update... implementations in info_detectors.js should follow these rules, and call sites in prepare.js, orientation_reflex.js, episodic_memory.js, conscience_level_2.js, etc. should pass the right parameters whenever relevant state changes.