About

How Resonance started

One person. One question. 66 years of music. And a neural network trained on the human brain.

The project

Resonance is an independent research platform that explores popular music through the lens of neuroscience and data. It analyzes 66 years of Billboard Hot 100 hits, not just how they sound, but how the brain responds to them.

The platform doesn't predict hits. It decodes why they resonate, through interactive explorations built on real data, rigorous statistics, and brain activation maps derived from an fMRI-trained model.

Fabrizio Gè

Fabrizio Gè

MSc Biomedical Engineering · Music Producer

I hold a MSc in Biomedical Engineering from Politecnico di Milano, with a specialization in neuroengineering, machine learning, and neurophysiological signal processing. My thesis focused on EEG-based attention and cognitive load monitoring using wearable BCI systems, developed in collaboration with EssilorLuxottica. It resulted in a patent (co-inventor, pending) and two scientific publications currently under submission.

But before I ever touched an EEG cap, I was already deep into music. I listen obsessively, I produce when inspiration strikes, and together with my university association we created Project Svolta, a music festival now in its 4th edition, born inside the Politecnico di Milano community. Music isn't just what I study. It's the thing that got me here.

Resonance was born where these two worlds meet: the brain and the beat. It's my way of asking, with data and not opinions, why certain sounds move us across generations.

LinkedIn 🎙 Podcast (PoliRadio) ✉ fabrizioge@hotmail.com
Not just research
🎧
Listener
Obsessively
🎹
Producer
When inspiration strikes
🎪
Project Svolta
Music festival · 4th edition
The method
♫

Billboard Hot 100

Every year-end chart from 1960 to 2025. 6,598 tracks, 2,602 unique artists. The most comprehensive historical dataset of popular music in the US. [source]

🧠

TRIBE v2

A deep learning model by Meta Research that predicts cortical activation from audio. 20,484 vertices across 6 functional brain regions. This is what makes Resonance unique: predicted brain responses at scale. [source]

📊

Audio features

Spotify and Deezer APIs provide valence, energy, danceability, loudness, tempo, mode, and more. 44 features per track, both audio-derived and brain-derived. [source]

⬡

Interactive analysis

Every finding is published as an interactive exploration with real data, live charts, and full methodology. No static figures, no paywalls. Explore the data yourself.

Built with Claude

Resonance is a solo project, but not a solo effort. The entire platform, from data pipeline and statistical analysis to visualization code and interactive explorations, is built in collaboration with Claude (Anthropic), working as a research and engineering partner.

This is what modern research can look like: one person with a question, an AI that can build, and the discipline to let the data speak.