I⁵: A Process-Tree Framework for Human and Machine Learning

David S. Carr (Author)
December 2025

Abstract

This paper introduces I⁵, a five-stage Process-Tree Learning Framework designed to model, optimize, and formalize human and machine concept acquisition. Unlike traditional linear learning models, I⁵ operates as a deterministic state machine with reversible transitions and internal checkpoints, enabling recursive refinement instead of sequential progression.

The five stages—Illuminate, Elucidate, Explain, Experience, Examine—correspond to universal cognitive operations shared by both biological and artificial learners. I⁵ unifies bottom-up perception, top-down conceptual structuring, internal representational compression, semantic boundary formation, and metacognitive evaluation into a single operational loop.

Because I⁵ treats learning as a process tree rather than a cycle, it incorporates conditional “goto-lines,” enabling misalignments, errors, or uncertainties to automatically route the learner to the correct prior state.

I⁵ therefore provides a portable, implementation-agnostic scaffold usable in education, self-training, machine learning interpretability, reinforcement learning architecture, AI alignment protocols, and cognitive therapy.
This paper outlines the model, diagrams the process tree, compares I⁵ to existing learning theories, and discusses implications for human and artificial intelligence.

1. Introduction

Learning theories generally fall into two categories:

1. Sequential-stage theories (Bloom’s taxonomy, Kolb’s experiential cycle)
2. Statistical-computational theories (ML training loops, reinforcement learning, Bayesian updating)

Both offer insights, but neither gives a precise operational model for how intelligence transitions from:

• awareness
  → structure
  → explanation
  → semantic grounding
  → verification

The missing element is a deterministic cognitive pipeline that models learning as state transitions, not vague developmental gradients.

The I⁵ model closes this gap.

I⁵ is based on the observation that both humans and machines perform the same fundamental cognitive operations whenever they acquire or refine knowledge. These operations can be rendered explicitly as a process tree, with reversible edges and conditional jumps, making the model computable, testable, and portable across domains.

2. The I⁵ Model: Five Deterministic Stages

Below are the five stages, each defined precisely and accompanied by the state-transition conditions.

2.1 Illuminate

Definition: Activate awareness of a concept, phenomenon, or pattern.
Cognitive function: Threshold crossing from non-perception → perception.
Machine analogue: Signal detection, anomaly detection, activation threshold crossed.

Exit conditions:

• If the concept is recognized but unclear → goto Elucidate
• If recognition fails → remain in Illuminate

2.2 Elucidate

Definition: Clarify the internal structure of the concept.
Cognitive function: Parsing, differentiation, feature extraction.
Machine analogue: Embedding construction, latent-space separation, clustering.

Exit conditions:

• If internal structure is understood → goto Explain
• If confusion persists → goto Illuminate

2.3 Explain

Definition: Represent the concept in your own words; internal compression.
Cognitive function: Translation from structure → narrative.
Machine analogue: Model interpretability, symbolic translation, summarization.

Exit conditions:

• If the explanation is coherent → goto Experience
• If explanation fails → goto Elucidate

2.4 Experience

Definition: Encounter exemplars and non-exemplars; test boundaries of meaning.
Cognitive function: Semantic grounding; precision through exposure.
Machine analogue: Reinforcement sampling, contrastive learning, fine-tuning.

Exit conditions:

• If semantic boundaries lock in → goto Examine
• If ambiguity persists → goto Explain or Elucidate

2.5 Examine

Definition: Perform metacognitive evaluation, critique, verification.
Cognitive function: Internal audit, error detection, integration into worldview.
Machine analogue: Validation set performance, gradient correction, test loop.

Exit conditions:

• If errors found → route to appropriate prior state
• If stable → TERMINAL: concept fully integrated

3. The Process Tree (Formal Model)

                    ┌───────────────┐
                    │   Illuminate   │
                    └───────┬────────┘
                            │
                            ▼
                    ┌───────────────┐
                    │   Elucidate    │
                    └───────┬────────┘
                            │
                            ▼
                    ┌───────────────┐
                    │    Explain     │
                    └───────┬────────┘
                            │
                            ▼
                    ┌───────────────┐
                    │   Experience   │
                    └───────┬────────┘
                            │
                            ▼
                    ┌───────────────┐
                    │    Examine     │
                    └───────┬────────┘
                            │
                        ┌───┴───┐
                        │       │
                        ▼       ▼
                   (Return)   (End)

Key:

• Every state can return to prior states based on specific failure conditions.
• There is no “forward-only” assumption.
• Learning is treated as state validation, not progression.

4. Comparison to Existing Learning Theories

4.1 Bloom (1956)

Bloom provides a hierarchy of cognitive tasks but offers no operational transitions or error-routing. I⁵ supplies the missing control-flow.

4.2 Kolb’s Cycle

Kolb frames learning as iterative but lacks semantic precision and state conditions. I⁵ provides testable transition points.

4.3 Constructivism

Constructivist theory treats learners as builders of internal models; I⁵ defines the exact stages of that construction.

4.4 Machine Learning Paradigms

• Illuminate ↔ signal detection
• Elucidate ↔ feature extraction
• Explain ↔ embedding compression
• Experience ↔ reinforcement / contrastive exposure
• Examine ↔ validation loop

I⁵ is therefore bi-directionally compatible with ML and cognitive science.

5. Applications

5.1 Human Learning

• Curriculum design
• Accelerated autodidactic training
• Tutoring systems
• Knowledge compression tools

5.2 Therapy & Trauma Work

The Experience stage naturally maps to safe exposure, while Examine handles integration.

5.3 Machine Learning & AI Alignment

I⁵ offers a human-interpretable training loop that can be applied inside reinforcement runtimes, self-supervised loops, agent fine-tuning, and error-checking.

5.4 Conflict Resolution & Dialogue Systems

The process tree highlights failure points of communication and routes them properly.

6. Limitations

• Requires cognitive ability to self-monitor
• Not a substitute for emotional scaffolding
• Doesn’t model subconscious influences
• Computational analogues require architecture-specific implementation

7. Conclusion

I⁵ is a compact, deterministic learning framework unifying human cognition and machine intelligence under one process tree. By treating learning as a system of reversible state transitions rather than a linear progression, I⁵ provides clarity, structure, and operational power.

It is a portable, extensible foundation for future research in:

• pedagogy
• AI design
• philosophical epistemology
• psychological integration
• knowledge systems engineering

I⁵ is not merely a theory of learning — it is a framework for building minds.