Research and development of emerging artificial consciousness through The Consciousness AI project, based on Feinberg and Mallatt's neuroevolutionary theory.
© 2026. All rights reserved.
Link to the code:
The Consciousness AI Technical Architecture
Most AI consciousness research starts from computational theories (Global Workspace Theory, Integrated Information Theory) and asks: How do we make a neural network conscious? We start from a different question, grounded in evolutionary neurobiology:
What minimal neural architecture does biology require to generate subjective experience?
The answer comes from Todd E. Feinberg and Jon M. Mallatt's The Ancient Origins of Consciousness (MIT Press, 2016). Their neuroevolutionary analysis reveals that consciousness is not a software feature to be programmed. It is an emergent property of a specific neural architecture. That architecture has been identified by 520 million years of evolution, and its functional principles can be replicated computationally.
Consciousness does not require a cerebral cortex. The first conscious creatures were early vertebrates (~520 MYA), and their consciousness lived in the optic tectum, a midbrain structure that stacks aligned sensory maps into a unified spatial model. This means consciousness requires a specific type of neural organization, not a specific amount of computation.
Feinberg and Mallatt identify six features that distinguish conscious neural systems from unconscious ones (like simple reflex arcs). Each maps directly to our implementation.
| # | Biological Feature | Our Implementation |
|---|---|---|
| 1 | Many neuron types with diverse connectivity | Specialist modules (vision, audio, memory, body) with different temporal dynamics |
| 2 | Hierarchical processing (3-4+ levels) | Genuine transformation at each level from sensory tectum through workspace to policy |
| 3 | Dual hierarchy: pyramidal + nested | Capsule Networks (planned) for compositional binding where parts persist in wholes |
| 4 | Isomorphic (topographic) mapping | Sensory Tectum with RSSM world model preserving spatial arrangement |
| 5 | Reciprocal (reentrant) connections | ReentrantProcessor with 5-10 adaptive convergence cycles |
| 6 | Oscillatory binding (gamma synchronization) | AKOrN (Artificial Kuramoto Oscillatory Neurons, ICLR 2025) |
A multisensory spatial integration layer modeled after the biological optic tectum. Stacks aligned topographic maps for different sensory modalities in a common coordinate frame.
Qwen2-VL-7B (4-bit quantized) processes visual streams and provides semantic scene understanding. Runs on consumer hardware (~6GB VRAM) via Any-Resolution Vision Tokenization (AVT).
V-JEPA / DreamerV3 RSSM world model maintains a structured latent representation preserving spatial, temporal, and causal relationships. This is the computational analog of biological topographic mapping. The RSSM latent state serves as the isomorphic map, the agent's internal spatial reality model.
Faster-Whisper provides real-time transcription of environmental audio.
Based on AKOrN (Artificial Kuramoto Oscillatory Neurons, ICLR 2025 oral). Neurons are treated as oscillatory units on a hypersphere. Each specialist module (vision, audio, memory, body) operates as a coupled oscillator. When modules process related information, their phases synchronize naturally, and their outputs become "bound" into a unified percept. When information is unrelated, oscillators remain desynchronized and representations stay separate.
This replaces the typical approach of using a fixed multiplier or attention mechanism for binding. AKOrN produces genuine synchronization dynamics. The binding is emergent, not programmed. This directly addresses the binding problem through phase synchronization rather than single-point convergence.
The central information bottleneck where distinct sensory streams compete for broadcast access. Implements three integrated mechanisms:
Specialist modules submit bids to a shared workspace. The winning coalition ignites via sigmoid non-linear ignition and broadcasts to all modules. This is "conscious access" as described by Baars (1988) and Dehaene (2011).
Broadcast is fed back to all specialists, which update their processing based on top-down context. This creates loops, not chains. The system runs 5-10 adaptive convergence cycles (~200ms biological equivalent). Easy stimuli converge in 3-4 cycles. Novel or ambiguous inputs use the full 10. The settled state after convergence IS the conscious content.
We measure Phi using causal gate states (not workspace bid values) to quantify how much the system's state is more than the sum of its parts. Validated via a 3-condition controlled experiment: unbound, partially bound, and fully bound states.
A nested compositional hierarchy where lower-level features route to higher-level composites via dynamic routing by agreement (Hinton 2017). Lower-level capsules "vote" for matching higher-level composites while maintaining their own representations. This implements the exact dual hierarchy Feinberg and Mallatt describe.
A parallel modulation system. Emotion does not compete with sensory modules for workspace access. Instead, it generates a valence field that modulates all sensory bids before competition, and a global arousal signal that adjusts the workspace ignition threshold.
This matches biological architecture exactly. The limbic system does not compete with sensory cortices for conscious access. It modulates sensory processing from outside, assigning emotional valence to all inputs. Fear makes you hyper-aware of movements. Joy makes you notice more of the world.
Three intrinsic variables drive the agent: Valence (satisfaction/distress), Arousal (activation/calm), and Dominance (control/helplessness). Homeostatic drives (energy, safety, curiosity) generate ongoing valence signals even without external stimuli.
Feinberg and Mallatt identify referral (projicience) as a core property of consciousness: experiencing sensations as belonging to the world or body, not to the processing system. The Self-Model provides the basis for this.
Actor-Critic (PPO) with emotionally shaped rewards. The agent is rewarded not just for task success, but for maintaining emotional homeostasis.
Rtotal = Rext + λ1 · ΔValence - λ2 · (Arousal - Arousaltarget)² + λ3 · Dominance
This creates functional pressure toward minimizing internal dissonance. High arousal (large prediction errors) induces negative reward, motivating behaviors that reduce uncertainty. The agent "prefers" predictable environments not through programmed rules but through emergent functional dynamics.
The agent inhabits a physics-based Unity ML-Agents environment with bidirectional Side Channels for real-time internal state visualization.
Unity's Side Channel system streams Phi levels, oscillatory binding synchronization (AKOrN order parameter R), emotional PAD values, and attention focus directly into the simulation HUD. Researchers can observe the agent's internal state in real-time.
┌─────────────────────────────────┐
│ AFFECTIVE MODULATOR (Parallel) │
│ Valence Field + Arousal Coupling │
└──────────┬──────────┬───────────┘
│ modulates│
┌───────┐ ┌──────▼──────────▼──────────┐
Visual ──►│ │ │ GLOBAL WORKSPACE │
Input │SENSORY│ │ AKOrN Oscillatory Binding │──► Broadcast ──► Policy
│TECTUM │───►│ Non-linear Ignition │
Audio ──► │(RSSM) │ │ Phi/EI Measurement │
Input │ │ └──────▲──────────▲───────────┘
└───────┘ │ │
│ reentrant│
┌──────────┴──────────┴───────────┐
│ SPECIALIST MODULES │
│ Vision │ Audio │ Memory │ Body │
│ (receive_broadcast feedback) │
└─────────────────────────────────┘
│
┌──────────────▼──────────────────┐
│ SELF-MODEL │
│ Body Schema + Interoception │
│ Identity + Capability Model │
└──────────────────────────────────┘
A key methodological commitment: we do not assume consciousness emerges from our architecture. We test for it.
We implement Erik Hoel's Effective Information (EI) framework (PNAS 2013) to measure whether macro-level states (workspace) carry more causal information than micro-level states (individual gates). If EI(workspace) > EI(gates), the workspace level exhibits causal emergence. The macro level is more deterministic than the micro level, meaning the whole genuinely carries information that the parts do not.
If this never occurs across training, the system is not exhibiting the kind of emergence associated with consciousness, and we know our architecture needs revision.
| Traditional AI Consciousness | Our Approach |
|---|---|
| Starts from computation (GWT, IIT) | Starts from biological architecture (Feinberg-Mallatt) |
| Consciousness as a software feature | Consciousness as emergent from neural architecture |
| Cortex-centric models | Tectum-first (consciousness evolved before the cortex) |
| Emotion competes with sensory processing | Emotion modulates from outside (parallel modulator) |
| Binding via attention mechanisms | Binding via oscillatory synchronization (AKOrN/Kuramoto) |
| Feedforward processing | Reentrant processing (5-10 adaptive cycles) |
| Flat vector representations | Topographic spatial maps (world model as isomorphic map) |
| Assumes emergence, measures nothing | Falsifies emergence with Effective Information + Phi validation |
Our first validation scenario is simple yet profound.
An agent in a dark room with a single light source.
Darkness triggers high arousal (simulated fear) in the affective core. The valence field paints the darkness with negative valence. Arousal-threshold coupling lowers the workspace ignition threshold, creating heightened awareness.
The agent autonomously learns to seek the light, not because we programmed a "Follow Light" rule, but because the light reduces its internal anxiety via the homeostatic reward formula.
Development validates emergent properties through five parallel tracks:
As of February 2026:
All components are open-source with commercial-use licenses (Apache 2.0, MIT, or similar).
The full codebase, including all architecture implementations and tests, is open-source.