Z-scores or standard scores are determined in relationship to a data-base of same age individuals. A “normal” curve for a population is bell-shaped and divided into standard deviation units. In terms of z-scores or standard scores, the mean is 0 and plus or minus 2 standard deviation units or z-scores represent about 95% of the population. In terms of z-score neurofeedback, the idea is to provide positive feedback when the parameters of interest (power, coherence or phase) are within a given range (e.g., between -2 and +2 standard scores or z-scores).
Whether a few sites are selected for training (typically two to four sites) or a “whole brain” approach is used with training at 19 sites, feedback is based on staying within the upper and lower limits (thresholds) determined by standard scores. With training, the thresholds for feedback can be adjusted to provide optimum feedback. The z-scores are set either based on a qEEG or in real time as the training session progresses.