Skip to main content
100%

Nervous System Simulation

✓ Published0🌍 Public
NNita
Last edited Apr 4, 2025
Created on Apr 4, 2025
Forked from 3D Cube of Spheres

This interactive WebGL visualization depicts a simulated neural network as a glowing three-dimensional mass of branching neurons and curved connections. Users can click any neuron to trigger it, sending luminous signal pulses that travel along quadratic Bézier curve connections to neighboring cells. Built with Three.js and OrbitControls, the scene uses ray casting for neuron selection and an object-pooling system for efficient pulse rendering. A control panel allows real-time adjustment of firing rate, signal speed, and pulse size.

AI-generated description

Neural Network Simulation

An interactive 3D visualization of a neural network using Three.js. This simulation creates a visually stunning representation of neurons and their connections with realistic signal propagation.

Features

  • Realistic Neural Structure: Neurons are organized in a central core with radiating branches
  • Signal Propagation: Visual representation of neural signals traveling between neurons
  • Dynamic Firing: Neurons fire spontaneously or when receiving signals
  • Interactive: Click on any neuron to manually trigger it
  • Performance Optimized: Uses object pooling for signal visualization
  • Adjustable Parameters: Control firing rate and signal speed in real-time
  • Responsive Design: Adapts to different screen sizes

Controls

  • View Navigation:
    • Rotate: Click and drag
    • Zoom: Mouse wheel
    • Pan: Right-click and drag
  • Neuron Interaction: Click directly on any neuron to trigger it
  • Simulation Controls: Use the panel in the top-right corner to adjust parameters
    • Firing Rate: Controls the probability of spontaneous neuron firing
    • Signal Speed: Adjusts how quickly signals propagate through the network
    • Pulse Size: Change the size of signal pulses
    • Trigger Random Neurons: Manually activate random neurons
    • Auto-Rotation: Toggle automatic camera rotation

Technical Details

The simulation is built using:

  • Three.js for 3D rendering
  • Object Pooling for efficient particle management
  • Quadratic Bezier curves for neural connections
  • Ray casting for interactive neuron selection

The simulation demonstrates principles of neural networks including signal propagation and the interconnected nature of neurons in a visually engaging way.

Browser Compatibility

The simulation works best in modern browsers that support WebGL and ES6 modules.

MIT Licensed

Similar vizzes