@aiql-org/soul

Affective AI Engine — Modeling emotional states for AIQL agents.

This package implements an artificial "Soul" that processes inputs (stimuli), maintains an internal affective state (Reward, Pain, Stress, Novelty), and influences AI behavior.

🌟 Features

• Biological Homeostasis: Models emotional decay and equilibrium over time.
• 4-Axis Affect Model:
  • Reward (Joy): Positive reinforcement from successful tasks.
  • Pain (Sorrow): Negative feedback from failures or contradictions.
  • Stress (Fear): System load, uncertainty, or time pressure.
  • Novelty (Curiosity): New information, exploration, or high entropy.
• Auto-Integration: Automatically detects affective metadata in AIQL !Feel intents.

📦 Usage

import { ArtificialSoul } from '@aiql-org/soul';

const soul = new ArtificialSoul();

// Process a stimulation
soul.process({
  type: 'discovery', // Triggers Novelty -> Curiosity
  intensity: 0.8,
  source: 'SemanticExploration'
});

// Check current state
const state = soul.getCurrentState();
console.log(state);
// {
//   dominantEmotion: 'Curiosity',
//   arousal: 0.75,
//   valence: 0.4,
//   vectors: { reward: 0.2, pain: 0.0, stress: 0.1, novelty: 0.8 }
// }

🧠 Influence on Semantics

The Soul engine couples with @aiql-org/semantics:

• High Novelty → Triggers Exploration (Curiosity).
• High Stress → Reduces Coherence (Decoherence).
• High Reward → Reinforces Trust (Belief).

License

MIT