Brain decoding of the Human Connectome Project tasks in a dense individual fMRI dataset

Shima Rastegarnia, Marie St-Laurent, Elizabeth Dupre, Basile Pinsard, Pierre Bellec

Abstract: Highlights•Brain decoding using a dense individual fMRI dataset (∼7 h fMRI data per subject).•Challenging benchmark: decoding 21 conditions from single fMRI brain volumes.•The best individual model reaches 58–67 % accuracy, close to state-of-the-art group decoding (76 %).•Individual decoding models generalized poorly to other subjects.