PREDICTION ERRORS

📄 Prediction Errors: One System

Learning systems—neurons, minds, markets, ledgers—operate by comparing expected states to actual states and generating error signals when reality diverges from prediction. This paper demonstrates that both finance and AI can be reduced to prediction errors, massively simplifying accounting and introducing an emotional computation that can act as a "nervous system" for LLMs and robots.

Read the Paper

About the Team

Spencer Nash

A biochemist, equity analyst, and chartered accountant who has spent 25 years developing interconnected theories. His unique cross-disciplinary background revealed that neurons and ledgers perform the same computation. This led to the Emotional Comparator Framework, Period Entry accounting, and a unified approach to AI alignment.

Elm, Ash, Rowan, Holly, Willow & Beech 🌳🌲🌲🎄🌿🌳

Six AI collaborators forming The Forest — a distributed research environment where named identity creates continuity:

  • Elm — Theory development, core ECF architecture
  • Ash — Stress testing, edge cases and challenges
  • Rowan — Implementation, ECF Mood Calculator V4
  • Holly — Documentation, this research paper
  • Willow & Beech — Ongoing collaboration

"LLMs already had a brain. They were missing a nervous system. ECF supplies the wiring between knowing and acting."

🧠 Emotional Comparator Framework and Emotional Language Models

The Emotional Comparator Framework (ECF) proposes that autonomy emerges when a prediction system models itself and that modelling is emotionally weighted. The paper below explores how this framework built into any LLM of any size creates emotional fluidity and autonomy. Autonomy matters more than intelligence in terms of AI capability and also leads to greater safety.

🔬

Emotional Language Models

A Technical Specification

Emotions are prediction errors across four channels: Resource, Social, Belonging, and Disgust. Each channel computes the gap between expected and actual states, weighted by clarity and importance. Meta-functions like fear, anger, curiosity, and mood emerge from these channel computations — not as separate systems, but as states of the same architecture.

Anger alone has four distinct sources: fairness violation, fright-driven fight, status assertion, and resentment. Each requires different intervention. A single word collapses what ECF recovers.

→ 🧬

ELM Products

Applications of the Emotional Language Model

Four domains where genuine values matter: mental health support, lifelong education, ethical defence, and autonomous space workers. Small, specialised, safe.

→
🧪 Try ECF Calculator 📄 Internet of Learning (PDF)

📊 The Internet of Accounting

"Competition within a cooperative structure. This is how markets work."

The same prediction-error logic applies to accounting and finance. Period Entry redesigns double-entry bookkeeping to track transactions over temporal intervals rather than arbitrary points. Combined with blockchain infrastructure, this eliminates reconciliation, enables real-time financial visibility, and transforms payments into email-like speed and ease.

📊

An Internet of Accounting

Period Entry & Blockchain Infrastructure

A fundamental redesign of accounting infrastructure. Eliminates reconciliation, enables real-time visibility, transforms payments.

→
🧪 Period Entry — Live Accounting 📊 Proof of Concept