Email (record): Publisher Correction: A 204-subject multimodal neuroimaging dataset to study language processing