Table of Contents


scalpnorm - normalize ERP data for scalp distribution analyses


scalpnorm nfactors factor1 factor2 ...


scalpnorm Normalizes data for scalp distribution comparisons using an analysis of variance. Normalizes vector length per McCarthy and Wood, 1985 (EEG, 62:203-028), by dividing each electrode value by the square root of the sum of the squares of all the electrode values. Normalization is performed for each subject, for each condition, across electrodes.

scalpnorm reads the data from stdin and writes the normalized data to stdout. Factors are listed in order from slowest moving to fastest moving. Subjects are the fastest moving factor, and scalp sites are the next fastest moving factor in both input and output. Thus, if you had 3 conditions, 26 electrodes, and 5 subjects as input (subjects fastest moving factor, electrodes next fastest, conditions slowest), the command line would be:

scalpnorm 3 3 26 5 < input-file > output-file

If you had 3 conditions, 2 sub-conditions, 26 electrodes, and 5 subjects, the command line would be:

scalpnorm 4 3 2 26 5 < input-file > output-file

In the first example, there are 3 x 5 = 15 normalizations performed; in the second example, 3 x 2 x 5 = 30 normalizations are performed.


None yet reported.


Jonathan C. Hansen

Table of Contents