ASTRA System Status

Real-time Metrics

SNN Engine State

Advance SNN Simulation

Reset SNN Engine

Spike Injection

Consciousness Assessment (Proxy)

IRB Neural Welfare Check

Modify State Parameter

Bio-Computing Platforms

Full State Snapshot

Simulation Control

Encode SNN State to Latent Space

Predict Next SNN State in Latent Space

CEM Planning for Optimal Spike Injection

Violation-of-Expectation Detection

Online World Model Training Step

World Model Status & Metrics

V-JEPA 2 Visual Encoding (Image/Video)

A-JEPA Audio Encoding (Waveform → Mel → Latent)

Koniku Kore Olfactory Encoding (Chemoreceptor → Latent)

Cross-Modal Attention Fusion

Full Multimodal Pipeline (All Modalities → Fused z)

Multimodal Sensor Pipeline Status

Run one or more ACM cycles (the_consciousness_ai port): SNN signals → AKOrN binding → GNW ignition → qualia → emotion → reward shaping → emotional memory → self-model → second-order loop. Set stopWhenSatisfied to halt early once the recursive loop reaches a sustained satisfactory (converged, low-curiosity, stable) regime.

Global Neuronal Workspace state: ignition, focus, qualia, sync R, unity metrics, access history

Appraise raw signals into PAD emotional space (Mehrabian) with inertia

Store an experience in emotional memory (attention-gated, salience-indexed)

Retrieve memories by blended cosine similarity, PAD congruence and salience

Self-representation state: interoception, epistemic model, temporal continuity, attention schema

Consciousness proxy report: GNW metrics, Effective Information, Φ̃-RIIU, composite score

Reset the TCAI consciousness system (workspace, memory, emotion, metrics)

Second-order (self-evidencing) loop snapshot: meta-learning velocity, RND curiosity (epistemic value), capability model, meta-consciousness score, developmental stage. The system observing and correcting its own predictive capacity (Legros 2026 §3.2).

Meta-learning state (MetaLearningModule port): learning velocity from RPE-variance dynamics. velocity>0 ⇒ converging; noveltySpike ⇒ novel/confusing regime. Optionally inject an RPE sample.

Agency capability model (DirectExperienceLearner port): action → expected-valence map (EMA). Query expected outcome of an action, or list the learned capability table.

Intrinsic-reward / curiosity (RNDCuriosity port): prediction error between a frozen random target and an online predictor on a representation vector. High error = novelty = exploration drive (EFE epistemic value proxy, Legros 2026 §4.1). Defaults to the current GNW broadcast.

Meta-consciousness composite (MetaconsciousnessEvaluator port): weighted score over confidence calibration, learning awareness, self-continuity and error monitoring. PROXY of meta-representation capacity, not a measurement.

Longitudinal developmental tracking (DevelopmentTracker port): coarse stage (nascent→reactive→integrative→reflective) from the running composite-proxy level, stability and meta-representation score. Second-order self-monitoring over time.

Inspect or configure the recursive double-loop halting criterion (v2.7). With no arguments, returns the current satisfaction state and active thresholds. With arguments, updates them. The loop halts only when variational free energy has settled (|ΔF| ≤ epsFreeEnergy) AND realized task quality is high (≥ minTaskQuality) AND epistemic value is low, sustained over patience cycles — stationarity alone is insufficient (Legros 2026 §2.2/§4.3).

Active-inference core telemetry (v2.7): the REAL variational free energy F (surprise), expected free energy G(π) decomposed into pragmatic + epistemic value, the realized task quality, the model entropy, and the Dirichlet-learned action. This is the principled quantity the halting criterion thresholds on — not a heuristic correlate (Da Costa et al. 2020; Legros 2026 §4.3).

Calibrate the halting threshold on the measured ΔF scale instead of a guessed constant. Runs cycles warm-up cycles at the given reward, records the free-energy increments |ΔF|, and sets epsFreeEnergy to factor× their median. Returns the measured ΔF scale and the applied threshold. Addresses the v2.7 critique that the default 0.02 nats was uncalibrated.

NeuroPlatform v2 — Platform & Controller Status

NeuroPlatform v2 — Define, validate & upload a StimParam (charge-balanced biphasic stimulation)

NeuroPlatform v2 — Fire trigger(s): execute uploaded StimParams via a 16-bit trigger array

NeuroPlatform v2 — Closed-loop _count_spike: spikes per electrode over an N-ms window

NeuroPlatform v2 DB — SpikeCountQuery: spikes/minute per electrode over a time window

NeuroPlatform v2 DB — SpikeEventQuery: individual spike timings over a window

NeuroPlatform v2 DB — TriggersQuery: triggers sent to the organoid over a window

NeuroPlatform v2 — Last MEA camera capture (descriptor + viability)

NeuroPlatform v2 — Closed loop: read organoid → couple to ASTRA fusion/ROS/ethics, optionally drive the SNN