RESUMO
Although scalp potential distributions do not uniquely determine the location and configuration of neural generators, they are important because they provide the necessary conditions that any hypothesized sources must satisfy and suggest a basis for testing alternate source hypotheses. One problem that could confound the correct interpretation of scalp potentials is the choice of reference electrode. Changing the reference may make activity patterns and waveform components appear and disappear (Pascual-Marqui et al. (1988) Int. J. Neurosci., 43: 237-249). The cortical imaging technique (CIT), a method for approximating potential fields on the cortical surface, was used to test the effects of the choice of reference electrode on these fields. Simulated and empirical evoked potential scalp-recorded referential data were mathematically analyzed for the case in which the reference (linked-ears) was arbitrarily assumed to be at zero potential, and the case in which the reference was the 'average' electrode, the arithmetic mean of all of the scalp-recorded voltages in the referential montage. The results for the two references were similar. This is encouraging because potential measurements relative to a point at infinity (zero potential) are never available and the assumption that any actual reference used for a recording is at zero potential is therefore suspect.
