Discover datasets

This magnetoencephalography (MEG) dataset comprises recordings from 20 adult participants viewing simple geometric shapes (hexagons, triangles, quadrilaterals) and other visual stimuli. The study investigates the neural mechanisms underlying human perception of regular geometric shapes, examining how the brain processes discrete geometric regularities such as symmetries and parallelism. The dataset supports research into the specialized brain systems engaged in abstract geometric perception and their distinction from general visual processing.

Participants

21

Size

71.1 GB

Version

v1.0.0

Updated

Jul 10, 2026

The Natural Object Dataset (NOD) integrates magnetoencephalography (MEG) and electroencephalography (EEG) data collected from subjects viewing naturalistic stimuli consisting of diverse object images from ImageNet. This multimodal neuroimaging resource complements previously collected fMRI data from the same subjects and stimuli, enabling investigation of neural mechanisms underlying object recognition with both high spatial and temporal resolution across three complementary imaging modalities.

Participants

31

Size

215 GB

Version

v1.0.0

Updated

Jul 22, 2026

This MEG dataset comprises neurophysiological recordings from 90 participants (38 controls, 52 with major depressive disorder) performing a probabilistic learning task during a 306-sensor Elekta Neuromag acquisition. The study investigates reward positivity and ventromedial prefrontal cortex activation patterns in major depressive disorder using reinforcement learning paradigms. Participants were clinically characterized using structured diagnostic interviews (SCID) to establish diagnostic group membership. Note: Anatomical MRI (T1) and resting-state data are planned for future release.

Participants

85

Size

159 GB

Version

v1.0.0

Updated

Jul 10, 2026

This dataset comprises simultaneous naturalistic fMRI and MEG recordings from 60 participants (30 per modality) who watched a 25-minute excerpt from a Chinese reality TV show. The study captures brain activity during both active viewing with comprehension assessment and subsequent spontaneous mental replay, providing a rich multimodal neuroimaging resource for investigating naturalistic cognition and inter-subject correlation analyses in ecologically valid contexts.

Participants

60

Size

112 GB

Version

v1.0.0

Updated

Jul 10, 2026

NeuroMorph is a high-temporal resolution magnetoencephalography (MEG) dataset comprising recordings from 24 subjects totaling over 17 hours of data acquired using a KIT/Yokogawa system with 208 magnetometer channels. The dataset is designed to support morpheme-based linguistic analysis and investigation of language and cognitive processes through visual presentation paradigms, providing a resource for studying neural mechanisms of language processing at fine temporal resolution.

Participants

24

Size

140 GB

Version

v1.0.0

Updated

Jul 10, 2026

FACE-DEC is an OPM-MEG dataset containing neuroimaging data acquired during face processing tasks. The dataset includes raw magnetoencephalography recordings, preprocessing pipelines, decoding analyses, representational similarity analysis (RSA), and statistical comparisons. Analyses were conducted at the whole-brain and sensor levels without individual structural MRI or facial point cloud data. This study was approved by the Institutional Review Board of Peking University. For correspondence, contact [email protected].

Participants

21

Size

27.6 GB

Version

v1.0.0

Updated

Jul 10, 2026

This magnetoencephalography (MEG) dataset examines how heuristics in risky decision-making relate to preferential representation of decision-relevant information. Participants completed a risky decision-making task across multiple scanner runs, with event timing recorded via photodiode and synchronized to MEG recordings. The dataset includes raw MEG data, event timing information, and preprocessed epoched data for both a localizer task and the main choice task, supporting investigation of neural correlates of decision heuristics and information processing biases.

Participants

21

Size

431 GB

Version

v1.0.0

Updated

Jul 10, 2026

This dataset comprises simultaneous magnetoencephalography and subthalamic nucleus local field potential recordings from externalized deep brain stimulation patients with Parkinson's disease, acquired under dopaminergic medication on and off conditions. Participants performed two motor tasks—static forearm extension and self-paced fist clenching—each preceded by resting periods, with some patients recorded during rest only. The dataset enables investigation of basal ganglia-cortex circuit electrophysiology and its modulation by movement and pharmacological state.

Participants

20

Size

162 GB

Version

v1.0.0

Updated

Jul 10, 2026

This magnetoencephalographic (MEG) dataset comprises auditory oddball paradigm recordings from participants with first-episode psychosis and healthy controls, designed to investigate mismatch negativity (MMN) responses to pitch and duration deviants. The study includes three participant groups: healthy controls, first-episode schizophrenia, and first-episode affective psychosis patients. Data were collected as part of the Pittsburgh Early Psychosis Program between December 2017 and September 2021. The dataset includes clinical, neuropsychological, demographic, and medication data in the derivatives.

Participants

60

Size

120 GB

Version

v1.0.0

Updated

Jul 10, 2026

This dataset contains magnetoencephalography (MEG) recordings from a Neuromag phantom used to characterize device-specific artifacts generated by deep brain stimulation (DBS) systems. The study investigates how DBS-related electromagnetic interference affects MEG signal quality and measurement reliability, providing essential reference data for researchers working with concurrent DBS-MEG recordings in clinical and research settings.

Participants

4

Size

6.13 GB

Version

v1.0.0

Updated

Jul 10, 2026