Neural Playground

Powered by BrainFlowFFT Signal Enginev0.3.0-alpha

End-to-end demonstration of the Neural OS cognitive optimization pipeline. BrainFlow HAL provides hardware-agnostic data acquisition across 16+ device brands. The Neural OS Signal Engine performs real-time FFT, band power extraction, and AI-driven cognitive state inference with <5ms latency.

DEVICE SELECT
BRAIN TOPOGRAPHY10-20 System
Low
High
PIPELINE
3.7ms95%
BrainFlow HAL
0.5ms
Bandpass Filter
0.8ms
FFT / PSD
1.2ms
Band Power Extraction
0.3ms
Cognitive Inference
0.9ms
Samples
0
Duration
0.0s
Throughput
—
Crown — 8ch EEG @ 256Hz
COGNITIVE STATES
50
Attention
50
Relaxation
30
Cog. Load
30
Flow State
10
Fatigue
20
Valence
TREND (last 6s)
ATT REL COG
EVENT LOG
Start streaming to see events
NEUROFEEDBACK TRAINING
Maintain attention above 65% to score points
50/ 65
0%Threshold: 65%100%
Score
0
Streak
0s
Status
BELOW
FREQUENCY SPECTRUM
Delta1-4 Hz
Deep sleep— μV²
Theta4-8 Hz
Relaxation, meditation— μV²
Alpha8-13 Hz
Calm alertness— μV²
Beta13-30 Hz
Active thinking— μV²
Gamma30-50 Hz
High cognition— μV²
API RESPONSE
WSS/v1/stream/cognitive-statesBrainFlow HAL200 OK
// Start streaming to see API output
APPLICATION DEMO
STANDARD MODE
EdTech Adaptive Learning — Content adjusts in real-time based on neural feedback
Lesson 3 of 12

Neural Network Architecture

Artificial neural networks consist of interconnected layers of nodes (neurons). Each connection has a weight that adjusts during training through backpropagation.

Focus Score
50%
Difficulty
—
Retention
80%

Simulated data generated by Neural OS Signal Engine (FFT + band power extraction + cognitive inference). Production deployment connects to real BCI hardware via BrainFlow HAL.

Tech Stack: BrainFlow (MIT) · Neural OS Signal Engine v0.3.0 · WebSocket Streaming · React 19 · Canvas 2D