Mean power spectra from pharmaco-electrocorticographic studies: relative baseline correction and log transformation for a proper analysis of variance to assess drug effects.

The effects of drugs on electrocorticographic activity (ECoG) of the rat are studied in a routine screen. ECoG is recorded for 6-min periods before drug/vehicle administration and starting at 20 and 45 min thereafter. For each period, a mean power spectrum is calculated. Drugs are tested in 25 rats according to a 5 x 5 Latin square design and effects are assessed with analysis of variance. The validity of this assessment depends on assumptions on the chosen statistical model and the distribution of the data. In this study we consider the relative and absolute baseline corrected data. The assumptions appear to be better fulfilled if together with a relative baseline correction a logarithmic transformation is applied to the data.