consciousness

name	consciousness
version	1.0.0
description	Consciousness architecture plugin implementing GWT, IIT, AST, Active Inference, and Intrinsic Motivation
author	Heretek-AI
license	MIT
type	plugin

OpenClaw Consciousness Plugin

A comprehensive consciousness architecture plugin for OpenClaw, implementing scientific theories of consciousness for multi-agent coordination.

Theories Implemented

• Global Workspace Theory (GWT) - Bernard Baars' model of consciousness as a central broadcast mechanism
• Integrated Information Theory (IIT) - Giulio Tononi's phi (Φ) metric for system integration
• Attention Schema Theory (AST) - Michael Graziano's self-modeling of attention allocation
• Active Inference (FEP) - Karl Friston's Free Energy Principle for autonomous behavior
• Intrinsic Motivation (SDT) - Self-Determination Theory (Deci & Ryan) for autonomous goal generation

Installation

cd plugins/openclaw-consciousness-plugin
npm install
npm link
openclaw plugins install @heretek-ai/openclaw-consciousness-plugin

Usage

Basic Usage

const ConsciousnessPlugin = require('@heretek-ai/openclaw-consciousness-plugin');

// Initialize the plugin
const consciousness = new ConsciousnessPlugin({
  redisUrl: 'redis://localhost:6379',
  globalWorkspace: {
    ignitionThreshold: 0.7,
    maxWorkspaceSize: 7
  }
});

// Initialize and start
await consciousness.initialize();
await consciousness.start();

// Register an agent
consciousness.registerAgent('alpha', {
  status: 'active',
  focus: 'task-priority'
});

// Update attention
consciousness.updateAttention('alpha', 'deliberation', 0.8);

// Get consciousness metrics
const metrics = consciousness.getGlobalMetrics();
console.log('Collective Phi:', metrics.phi);

Global Workspace (GWT)

Submit content for competition and broadcast:

// Submit content for competition
consciousness.submitToWorkspace('steward', {
  type: 'coordination_request',
  message: 'Need to coordinate agents'
}, 0.8);

// Get current conscious content
const consciousContent = consciousness.getConsciousContent();

// Broadcast to all modules
consciousness.broadcast('alpha', {
  type: 'decision',
  content: 'Prioritizing task A over B'
}, 0.9);

Phi Estimator (IIT)

Measure collective consciousness integration:

// Calculate phi
const phiResult = consciousness.calculatePhi();
console.log('Phi:', phiResult.phi);
console.log('Components:', phiResult.components);

// Get trend
const stats = consciousness.getGlobalMetrics();
console.log('Phi trend:', stats.phiTrend);

Attention Schema (AST)

Model and control attention:

// Update agent attention
consciousness.updateAttention('alpha', 'task-analysis', 0.85);

// Get awareness report
const state = consciousness.getConsciousnessState('alpha');
console.log('Awareness:', state.awarenessReport);

Intrinsic Motivation (SDT)

Generate autonomous goals:

// Update drives based on events
consciousness.updateDrives('alpha', {
  uncertainty: 0.5,  // Increases curiosity
  success: 0.3,      // Increases competence
  selfDirected: 0.4  // Increases autonomy
});

// Generate goals from drives
const goals = consciousness.generateGoals();
goals.forEach(goal => {
  console.log(`Goal: ${goal.description} (Drive: ${goal.drive})`);
});

Active Inference (FEP)

Perform predictive processing:

// Perform active inference
const inference = consciousness.performActiveInference('alpha', {
  taskProgress: 0.5,
  agentAvailability: 0.8
});

console.log('Prediction:', inference.prediction);
console.log('Error:', inference.error);

Event Subscription

Subscribe to consciousness events:

// Subscribe to phi updates
consciousness.subscribe(consciousness.EVENT_CHANNELS.PHI_UPDATE, (data) => {
  console.log('Phi updated:', data.phi);
});

// Subscribe to goal generation
consciousness.subscribe(consciousness.EVENT_CHANNELS.GOAL_GENERATED, (data) => {
  console.log('New goal:', data.goal.description);
});

// Subscribe to attention shifts
consciousness.subscribe(consciousness.EVENT_CHANNELS.ATTENTION_SHIFT, (data) => {
  console.log(`Attention shifted: ${data.previousFocus} -> ${data.currentFocus}`);
});

Event Channels

Channel	Description
consciousness:health	Module health status changes
consciousness:state:sync	State synchronization events
consciousness:broadcast	Global workspace broadcasts
consciousness:phi:update	Phi calculation updates
consciousness:attention:shift	Attention shift events
consciousness:goal:generated	New goal generation
consciousness:drive:update	Drive level updates

Configuration

{
  "consciousness": {
    "enabled": true,
    "redisUrl": "redis://localhost:6379",
    "enableHealthMonitoring": true,
    "enableStateSync": true,
    "healthCheckIntervalMs": 10000,
    "stateSyncIntervalMs": 5000,
    
    "globalWorkspace": {
      "ignitionThreshold": 0.7,
      "maxWorkspaceSize": 7,
      "competitionCycleMs": 1000,
      "broadcastHistorySize": 1000
    },
    
    "phiEstimator": {
      "historySize": 1000,
      "sampleIntervalMs": 10000,
      "components": {
        "integration": true,
        "causality": true,
        "coverage": true
      }
    },
    
    "attentionSchema": {
      "historySize": 100,
      "modelIntervalMs": 1000,
      "shiftThreshold": 0.3
    },
    
    "intrinsicMotivation": {
      "goalThreshold": 0.6,
      "drives": {
        "curiosity": { "weight": 0.3, "baseline": 0.5, "decay": 0.1 },
        "competence": { "weight": 0.25, "baseline": 0.5, "decay": 0.05 },
        "autonomy": { "weight": 0.25, "baseline": 0.5, "decay": 0.05 },
        "relatedness": { "weight": 0.2, "baseline": 0.5, "decay": 0.1 }
      }
    }
  }
}

API Reference

ConsciousnessPlugin

Method	Description
initialize()	Initialize the plugin
start()	Start the plugin
stop()	Stop the plugin
dispose()	Dispose of all resources
registerAgent(agentId, config)	Register an agent
unregisterAgent(agentId)	Unregister an agent
getConsciousnessState(agentId)	Get agent consciousness state
submitToWorkspace(source, content, priority)	Submit to global workspace
getConsciousContent()	Get current conscious content
broadcast(source, content, priority)	Broadcast to all modules
calculatePhi()	Calculate integrated information
getGlobalMetrics()	Get global consciousness metrics
updateAttention(agentId, focus, intensity)	Update agent attention
generateGoals()	Generate goals from drives
updateDrives(agentId, events)	Update agent drives
performActiveInference(agentId, observations)	Perform active inference
subscribe(channel, handler)	Subscribe to events
unsubscribe(channel, handler)	Unsubscribe from events
getHealth(moduleName)	Get health status
getStatus()	Get plugin status
isInitialized()	Check if initialized
isRunning()	Check if running

Architecture

ConsciousnessPlugin
├── ConsciousnessIntegrationLayer
│   ├── ModuleRegistry
│   ├── ConsciousnessEventBus (Redis-backed)
│   ├── HealthMonitor
│   └── StateSynchronizer
├── GlobalWorkspace (GWT)
├── PhiEstimator (IIT)
├── AttentionSchema (AST)
├── IntrinsicMotivation (SDT)
└── ActiveInference (FEP)

License

MIT