development fMRI dataset


See nilearn.datasets.fetch_development_fmri.


This movie-watching based functional MRI dataset is used for teaching how to use machine learning to predict age from naturalistic stimuli (movie) watching with Nilearn.

The dataset consists of 50 children (ages 3-13) and 33 young adults (ages 18-39). This dataset can be used to try to predict who are adults and who are children.

The data is downsampled to 4mm resolution for convenience. The original data is downloaded from OpenNeuro.

For full information about pre-processing steps on raw-fMRI data, have a look at README at

Full pre-processed data:

Raw data can be accessed from :

See Richardson et al.[1].



functional MRI Nifti images (4D) per subject


TSV file contain nuisance information per subject


Phenotypic information for each subject such as age, age group, gender, handedness.



usage is unrestricted for non-commercial research purposes.