We are thrilled to announce the launch of the NeuroQuantix Cognitive State Classifier, an interactive demo now live on Hugging Face Spaces. This tool is a direct result of our research into the connection between language and cognitive effort, allowing users to experience the power of our AI model firsthand.
Input any sentence, and the model will predict whether it is more likely to be processed in a Normal Reading (NR) state or a high-effort Task-Specific Reading (TSR) state.
[ 🚀 Launch the Interactive Demo on Hugging Face Spaces ]
The Science Behind the Demo
This project addresses a core question in computational neuroscience: can we find the “linguistic fingerprints” of cognitive load without direct neural measurements? The model was trained and validated on the ZuCo 2.0 dataset, a rich scientific resource that correlates written text with human EEG and eye-tracking data.
The technology powering this demo is a custom-architected hybrid system. It combines the deep contextual understanding of a fine-tuned BERT transformer with a rich set of engineered linguistic features. This innovative approach proved highly effective in our research, achieving a 94.74% F1-Score in classifying Task-Specific Reading states.
Why This Matters
This demo is more than a technical showcase; it’s a proof of concept for a future where non-invasive, scalable digital biomarkers can be developed from language data. The ability to quantify cognitive effort from text has potential applications in personalized learning, mental health monitoring, and beyond.
We invite you to explore the demo, test it with your own sentences, and see our research in action.
NeuroQuantix 