Establishing the significance of observed effectsis a preliminary requirement for any meaningful interpretationof clinical and experimental Electroencephalographyor Magnetoencephalography (MEG) data. We propose amethod to evaluate significance on the level of sensorswhilst retaining full temporal or spectral resolution. Inputdata are multiple realizations of sensor data. In this context,multiple realizations may be the individual epochs obtainedin an evoked-response experiment, or group study data,possibly averaged within subject and event type, or spontaneousevents such as spikes of different types. In thiscontribution, we apply Statistical non-Parametric Mapping(SnPM) to MEG sensor data. SnPM is a non-parametricpermutation or randomization test that is assumption-freeregarding distributional properties of the underlying data. The method, referred to as Maps SnPM, is demonstratedusing MEG data from an auditory mismatch negativityparadigm with one frequent and two rare stimuli and validatedby comparison with Topographic Analysis of Variance(TANOVA). The result is a time- or frequencyresolvedbreakdown of sensors that show consistentactivity within and/or differ significantly between event orspike types. TANOVA and Maps SnPM were applied tothe individual epochs obtained in an evoked-responseexperiment. The TANOVA analysis established dataplausibility and identified latencies-of-interest for furtheranalysis. Maps SnPM, in addition to the above, identifiedsensors of significantly different activity between stimulustypes.