Nonparametric Statistical Analysis of Map Topographies on the Epoch Level
Establishing the significance of observed effects is a requirement for meaningful interpretation of clinical or experimental data. Averaging is commonly used to increase the signal-to-noise-ratio (SNR) of brain responses recorded in an event-related potential (ERP) or event-related field (ERF) experiment. However, the individual epochs collected contain additional information beyond what is represented by averages. Specifically, consistency between brain responses to a certain stimulus type, as well as differences between stimulus types, can be established by statistically analyzing all of the individual epochs in an ERP or ERF data set. Topographic analysis of variance (TANOVA) and statistical nonparametric mapping (SnPM) are nonparametric permutation or randomization tests which have previously been published but mainly been used to process per-subject averaged EEG data in the context of group studies. This chapter describes how to apply TANOVA to individual epochs on a sample-by-sample basis, even in the context of single-subject data. TANOVA is able to identify latencies of significantly different map topographies.
KeywordsMagnetoencephalography Electroencephalography Event-related fields Event-related potentials Mismatch negativity Mandarin language Statistical analysis Randomization statistics Nonparametric statistics Topographical analysis of variance Statistical nonparametric mapping
- Wagner M, Ponton C, Tech R, Fuchs M, Kastner J (2014) Non-parametric statistical analysis of EEG/MEG map topographies and source distributions on the epoch level. Hum Cogn Neurophysiol 7:1–23Google Scholar