Volume Conduction & Source Localization (EEG/MEG Forward & Inverse)

sLORETA

Standardized low-resolution electromagnetic tomography (sLORETA) is a noise-normalized minimum-norm method that standardizes the current estimate by its own estimated variance (derived from the resolution/data covariance) rather than by the noise covariance alone. Its notable theoretical property is zero localization error under ideal conditions (noise-free data, exact head model) for a single point source: the peak of the sLORETA map then coincides with the true source location.

This exact-localization property holds only for a single source and only in the idealized limit; with multiple simultaneous sources, model error or noise, sLORETA — like every linear distributed method — produces blurred maps with finite localization error and cannot resolve nearby sources. It remains popular precisely because of its unbiased single-source localization and its parameter-light formulation. LORETA and eLORETA are related smoothness- or weighting-based variants.

「Zero localization error」 is often over-interpreted: it is a single-source, noise-free, correct-model guarantee about the location of the maximum, not a claim of high resolution or of accuracy under realistic conditions.

Also called
standardized low-resolution electromagnetic tomographyeLORETALORETA family