# Dali_MEG_CurrentBiology

## Description

In each of five subject folders: 1. *_PixelBehMI_nuns_not.mat • “Ipixe_beh” stores the pixel-wise MI value <pixel visibility; nuns vs. don’t know>. • “TH” is the statistic threshold for the pixel-wise MI (FWER p < 0.001, one-tailed). • We downsample the bubbles for each spatial frequency band, so the pixel-wise MI is a 5456-d vector (i.e. 64x64 pixels in SF1 + 32x32 pixels in SF2 + 16x16 pixels in SF3 + 8x8 pixels in SF4 + 4x4 pixels in SF5 = 5456 pixels) 2. *_PixelBehMI_volt_not.mat • “Ipixe_beh” stores the pixel-wise MI value <pixel visibility; Voltaire vs. don’t know>. • “TH” is the statistical threshold for the pixel-wise MI (FWER p < 0.001, one-tailed). 3. *_FeatureMegMI.mat • “Ifeat” is the 3D time-by-feature-by-voxel MI matrix. • “TH” is the statistical threshold (FWER p < 0.05, one-tailed). 4. *_BrainFeatures.mat • ‘nuns_feat’: Indices of diagnostic brain features for the perception of “the nuns”. • ‘volt_feat’.: Indices of diagnostic brain features for the perception of “Voltaire”. • ‘nondiag_feat’: Indices of nondiagnostic brain features. 5. *_FullICA_LP_B_NMF_*.PDF • Brain features obtained from the NMF analysis. 6. *_Redundancy_nuns_volt_not.mat • “Ired” is the 3D redundancy matrix (time point x feature × voxel), using all trials. • “TH” is the statistical threshold (FWER p < 0.05, one-tailed). • “time” stores the post-stimulus time (in seconds) for each time point. 7. *_Redundancy_nuns_not.mat • “Ired” is the 3D redundancy matrix (time point x feature × voxel), using “the nuns” and “don’t know” trials. • “TH” is the statistical threshold (FWER p < 0.05, one-tailed). 8. *_Redundancy_volt_not.mat • “Ired” is the 3D redundancy matrix (time point x feature × voxel), using “Voltaire” and “don’t know” trials. • “TH” is the statistical threshold (FWER p < 0.05, one-tailed).