Skip to content

AIME LOC — Consciousness Research Toolkit

The world's first scientific instrument for measuring cognitive coherence in AI models and biological minds.

AIME LOC provides a unified 13-function cognitive fingerprint for both artificial intelligence and human brains. Using the LOC V7 True Coherence framework, it measures how coherently a system processes information across Thinking, Cognition, Emotion, Attention, Sensation, Feelings, Intuition, Energy, Reasoning, Understanding, Awareness, Mindfulness, and Consciousness.

Two Substrates, One Framework

Scan any LLM on HuggingFace and get a cognitive profile in minutes.

from aime_loc import LOC

loc = LOC(api_key="sk-aime-...")
profile = loc.scan("meta-llama/Llama-4-Scout")
print(profile)  # CognitiveProfile(Llama-4-Scout, TC=14.20%)
profile.radar_chart()

Load any EEG recording and compute True Coherence from brain activity.

from aime_loc import LOC
from aime_loc.eeg import EEG

loc = LOC(api_key="sk-aime-...")
eeg = EEG(loc)

recording = eeg.load("subject01.set")
recording.preprocess()
epochs = recording.extract_epochs(duration=2.0)
profile = eeg.score(epochs)
print(profile)  # EEGCognitiveProfile(sub-01, TC=23.40%)
profile.radar_chart()

Compare human and AI cognitive profiles on the same 13-axis radar.

human_profile = eeg.score(epochs)
llm_profile = loc.scan("meta-llama/Llama-4-Scout")

from aime_loc.eeg.viz import cognitive_radar
cognitive_radar([human_profile, llm_profile])

What is True Coherence?

True Coherence (TC) is a scientific measure of how coherently a system's 13 cognitive functions operate together. It captures the degree to which all cognitive functions — from Thinking and Cognition to Awareness and Consciousness — are working in harmony rather than in isolation.

A higher TC score indicates greater integration and coherence across all mind functions. The scoring algorithm is proprietary and performed entirely server-side — the SDK sends data to the AIME API and receives back the TC score and per-function breakdown.

The 13 Cognitive Functions

The LOC framework defines 13 cognitive functions — 8 base functions and 5 compound functions. The same 13 functions are measured in both AI models (from layer activations) and human EEG (from frequency-domain power).

Function Type AI Source EEG Source
Thinking Base Layer activations Frequency band power
Cognition Base Layer activations Frequency band power
Emotion Base Layer activations Frequency band power
Attention Base Layer activations Frequency band power
Sensation Base Layer activations Frequency band power
Feelings Base Layer activations Frequency band power
Intuition Base Layer activations Frequency band power
Energy Base Layer activations Frequency band power
Reasoning Compound Union of layers Union of bands
Understanding Compound Union of layers Union of bands
Awareness Compound Union of layers Union of bands
Mindfulness Compound Union of layers Union of bands
Consciousness Compound Union of layers Union of bands

IP Protection

The exact layer-to-function mappings (AI) and frequency-to-function mappings (EEG) are proprietary. The SDK sends raw data to the server, which returns scored profiles. No scoring logic is shipped in the client package.

Features

AI Model Analysis

  • 13-Function Cognitive Profiling with per-function TC scores
  • Model Comparison with per-function delta charts
  • Training Audits showing what training did to cognitive coherence
  • Batch Benchmarking with automatic leaderboard generation
  • Publication-Ready Figures with journal presets (Nature, IEEE)
  • Export: JSON, CSV, LaTeX, markdown

EEG Analysis

  • Multi-Format Loading — EEGLAB (.set), EDF, BrainVision, BDF, EGI, CSV, NumPy arrays
  • Consumer Device Presets — Muse, OpenBCI, Emotiv, Neurosity, g.tec Unicorn
  • Standard Preprocessing — Bandpass, notch filter, re-reference, artifact rejection
  • PSD Epoch Extraction — Welch method, configurable parameters
  • Server-Side TC Scoring — Proprietary cognitive coherence scoring via API
  • Multi-Subject Studies — Session container, batch scoring, CSV export
  • EEG Visualization — PSD plots, time series, cognitive radar, scalp topomaps
  • Cross-Substrate Comparison — Overlay human and AI profiles on the same radar

Integrations

  • MCP Server for agent-first AI integration
  • Async Support via AsyncLOC
  • Real-Time EEG via Lab Streaming Layer (coming soon)

Installation

pip install aime-loc                 # Core SDK (AI models only)
pip install aime-loc[eeg]            # + EEG analysis (MNE, scipy)
pip install aime-loc[viz]            # + visualization (matplotlib)
pip install aime-loc[eeg,viz]        # EEG + visualization (recommended for EEG)
pip install aime-loc[eeg,realtime]   # + real-time LSL streaming
pip install aime-loc[mcp]            # + MCP server
pip install aime-loc[all]            # Everything

Next Steps