Any stimulus
Video, images, sound and language go in. The model predicts the response they would produce in a human brain, without anyone in a scanner.
Mind Matter Labs builds models that predict how the brain responds to what people see, hear and read, trained on real neural recordings.
Video, images, sound and language go in. The model predicts the response they would produce in a human brain, without anyone in a scanner.
Predictions span named regions across the cortex and unfold over time, so you see which systems respond and when, not a single score.
Fitted on real neural recordings from people watching, listening and reading, and evaluated on people the model has never seen.
The hard part of this field is not fitting a model, it is knowing whether it generalises. We test on people and stimuli the model was never fitted on, the harder test and the more useful one.
We are equally interested in where it fails. A brain model tells you how the brain responds, not whether anyone will like what they see or act on it. We would rather say that plainly than have you find out.
Run in-silico experiments across thousands of stimuli before committing scanner time to the ones that matter.
Understand how audiences' brains process what you make, moment by moment, before it ships.
A pretrained model of typical brain response as a baseline to build on, compare against, and fine-tune.
If you have a question a model of the brain could answer, we'd like to hear it.
Get in touch