Overview

NeuralEntropy simulates neural spike activity using a 64-neuron motor cortex model and decodes it with a linear decoder to reconstruct 2D cursor velocity. It prints the decoded trajectory for visualization, enabling practical BCI prototyping and learning.

How This Skill Works

A neural simulator generates spike trains for 64 neurons based on cosine tuning. A linear decoder maps spike rates to 2D velocity components (v_x, v_y), producing a trajectory that is printed for review.

When to Use It

• Prototype 2D cursor control in a synthetic BCI environment
• Evaluate how changes in neuron count or tuning affect decoding quality
• Educate learners with a hands-on example of neural decoding
• Debug and validate a linear decoding pipeline using synthetic data
• Benchmark decoding performance and trajectory stability in a controlled setup

Quick Start

1. Step 1: Run the decode command to start the simulation loop with 64 neurons and cosine tuning
2. Step 2: Observe the printed 2D trajectory (v_x, v_y) as the cursor would move
3. Step 3: Modify tuning or neuron count to see immediate changes in the trajectory outputs

Best Practices

• Use cosine-tuned spike generation to reflect motor cortex patterns
• Validate decoder outputs against known synthetic velocities
• Experiment with seeds, length of simulation, and neuron subsets
• Log spike counts and decoded velocities for troubleshooting
• Visualize results via printed trajectory and optional plots to confirm consistency

Example Use Cases

• Prototype a 2D cursor control demo for a teaching or research presentation
• Assess how removing or adding neurons impacts velocity accuracy
• Generate reproducible BCI-like data for a methods paper or lecture
• Debug a linear decoding pipeline using synthetic spike data
• Demonstrate the impact of tuning parameters on decoding behavior

Frequently Asked Questions