Scientists built an AI called “Concept2Brain” that simulates human brain waves from pictures and text
What would your brain waves look like if you saw a photo of an alligator? A new AI tool called Concept2Brain predicts brain-wave responses to pictures or text, generating simulated EEG data that resembles real recordings.
Scientists have created an artificial intelligence system known as "Concept2Brain" that can simulate human brain waves based on visual images and written text. A study published in Nature Communications details how the open-access platform generates synthetic electroencephalography (EEG) recordings that closely resemble real human brain activity during visual and emotional perception.
Electroencephalography, or EEG, measures the tiny electrical voltage changes at the scalp produced by brain activity, allowing researchers to track neural activity millisecond by millisecond. By using Concept2Brain, researchers can now simulate brain-wave experiments on a computer before recruiting human volunteers, saving time and resources.
The tool was trained using EEG recordings from previous passive picture-viewing experiments involving 88 university volunteers, with brain activity recorded from 128 scalp sensors. Concept2Brain links two deep neural networks: one compresses EEG recordings into a 256-number summary while preserving individual differences, and the other translates conceptual numbers into brain-wave summaries, enabling users to upload images or text and download simulated brain responses.
Written by urgent.news from PsyPost's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
