- MneExperiment.set_inv(ori='free', snr=3, method='dSPM', depth=0.8, pick_normal=False, **state)
Set the type of inverse solution used for source estimation
ori ('free' | 'fixed' | 'vec' | float ]0, 1]) –
Orientation constraint (default
'free'; use a number between 0 and 1 to specify a loose constraint).
At each source point, …
free: … estimate a current dipole with arbitrary direction. For further analysis, only the magnitude of the current is retained, while the direction is ignored. This is good for detecting changes in neural current strength when current direction is variable (for example, due to anatomical differences between subjects).
fixed: … estimate current flow orthogonal to the cortical surface. The sign of the estimates indicates current direction relative to the surface (positive for current out of the brain).
vec: … estimate a current vector with arbitrary direction, and return this current as 3 dimensional vector.
float): … estimate a current dipole with arbitrary direction. Then, multiple the two components parallel to the surface with this number, and retain the magnitude.
snr (float) – SNR estimate used for regularization (
λ = 1 / snr). Larger λ (smaller SNR) correspond to spatially smoother and weaker current estimates. 3 is recommended for averaged responses, 1 for raw or single trial data. Set to 0 for unregularized inverse solution (
λ = 0).
method ('MNE' | 'dSPM' | 'sLORETA' | 'eLORETA') – Noise normalization method.
MNEuses unnormalized current estimates.
sLORETA2 and eLORETA 3 normalize each the estimate at each source with an estimate of the noise at that source (default
depth (float) – Depth weighting 4 (
0to disable depth weighting).
pick_normal (bool) – Estimate a free orientation current vector, then pick the component orthogonal to the cortical surface and discard the parallel components.
... – State parameters.
For details, see the MNE documentation on the inverse operator
Free and loose orientation inverse solutions have a non-zero expected value. In that case, when source localizing condition averages, the number of trials affects the expected value. For designs with unequal number of trials per cell, be sure to use equalize_evoked_count appropriately.
Dale A, Liu A, Fischl B, Buckner R. (2000) Dynamic statistical parametric mapping: combining fMRI and MEG for high-resolution imaging of cortical activity. Neuron, 26:55-67. 10.1016/S0896-6273(00)81138-1
Pascual-Marqui RD (2002), Standardized low resolution brain electromagnetic tomography (sLORETA): technical details. Methods Find. Exp. Clin. Pharmacology, 24(D):5-12.
Pascual-Marqui RD (2007). Discrete, 3D distributed, linear imaging methods of electric neuronal activity. Part 1: exact, zero error localization. arXiv:0710.3341
Lin F, Witzel T, Ahlfors S P, Stufflebeam S M, Belliveau J W, Hämäläinen M S. (2006) Assessing and improving the spatial accuracy in MEG source localization by depth-weighted minimum-norm estimates. NeuroImage, 31(1):160–171. 10.1016/j.neuroimage.2005.11.054