Neocortex Learning Error Driven Predictive

---
name: neocortex-learning-error-driven-predictive
description: "Neocortex learning framework via error-driven predictive learning using temporal derivatives, corticothalamic circuits, and competitive kinase synaptic plasticity. Activation: neocortex learning, cortical learning, predictive learning, thalamocortical, kinase plasticity, error-driven learning."
---

## Context

Paper: arXiv:2606.08720 - "This is how the Neocortex Learns" by Randall C. O'Reilly (Submitted 7 Jun 2026)

**Three Criteria for Sufficient Account of Neocortex Learning:**
1. **Computationally**: Approximate powerful, general-purpose learning algorithm that scales to human-level intelligence
2. **Algorithmically**: Implementable using known neural circuits within neocortex and associated brain structures
3. **Implementationally**: Detailed neurochemical-level account of algorithmic mechanisms

**Only Framework Meeting All Criteria:**
Error-driven predictive learning via temporal derivatives, driven by corticothalamic circuits, based on competitive kinase synaptic plasticity induction mechanisms.

## Core Methodology

### 1. Error-Driven Predictive Learning Framework
- **Mechanism**: Temporal derivative-driven error signals
- **Architecture**: Corticothalamic circuit implementation
- **Plasticity**: Competitive kinase induction at synapses
- **Implementation**: Axon neural simulation framework (spiking neurons)

### 2. Computational Power
- General-purpose learning algorithm
- Scales to human-level intelligence
- Demonstrated across wide range of cognitively motivated tasks

### 3. Neural Circuit Basis
- Corticothalamic circuits: bidirectional thalamus ↔ cortex connections
- Known, well-established circuit architecture
- Thalamic relay nuclei + cortical layers
- Feedback pathways for error propagation

### 4. Competitive Kinase Plasticity
- Neurochemical implementation level detail
- Kinase competition at synaptic sites
- Temporal derivative signals induce plasticity
- LTP/LTD balance via competitive mechanisms

## Implementation Steps

1. **Model Setup**: Implement spiking neuron network in Axon framework
2. **Corticothalamic Architecture**: Configure bidirectional thalamus-cortex circuits
3. **Temporal Error Signals**: Derive error from temporal derivatives of predictions
4. **Kinase Plasticity**: Configure competitive kinase induction mechanisms
5. **Task Training**: Test across cognitively motivated tasks (decision-making, sequence learning, categorization)

## Key Results

- Successful learning across challenging cognitive tasks
- Meets computational, algorithmic, and implementational criteria
- Spiking neuron implementation demonstrates biological realism
- Scalable to human-level intelligence tasks

## Pitfalls

- **Temporal Derivative Accuracy**: Requires precise timing for error signal computation
- **Kinase Competition Balance**: LTP/LTD balance critical for stable learning
- **Corticothalamic Delay**: Thalamic relay delays affect error signal timing
- **Task Complexity**: Simple tasks may not demonstrate full computational power
- **Spiking Implementation**: Computational cost higher than rate-based models

## Verification

1. Implement Axon framework spiking network
2. Configure corticothalamic circuit with bidirectional connections
3. Verify temporal derivative error signals propagate correctly
4. Check kinase competition plasticity mechanism
5. Test on standard cognitive tasks (sequence prediction, categorization)
6. Compare with alternative learning frameworks (backprop, Hebbian)

## Activation Keywords

- neocortex learning
- cortical learning
- predictive learning
- thalamocortical circuit
- error-driven learning
- temporal derivative
- kinase plasticity
- Axon framework
- spiking neural network learning

## Related Skills

- [[neuromodulated-synaptic-plasticity]] - neuromodulated plasticity mechanisms
- [[predictive-coding-exponential-family]] - predictive coding frameworks
- [[three-factor-snn-learning]] - three-factor learning rules
- [[equilibrium-propagation-lif-snn]] - equilibrium propagation learning

## References

- arXiv:2606.08720 - Original paper
- Axon neural simulation framework documentation
- Competitive kinase plasticity literature
- Corticothalamic circuit anatomy studies