I have spent many hours trying to decipher just what we are doing using Neurofeedback.
In fact in the process of trying to find the basis of Neurofeedback. I didn't believe Neurofeedback worked. I started a study. I'd find out if brain blood flow changed with this process. I'd do a blood distribution map, a SPECT, before and after EEG training.
I had everything underway when a radiologist at the chosen SPECT hospital refused to do SPECTs for non-medical reasons. In that study one of the dissertation candidates, Julie Weiner, found Britton Chance's papers on near infrared spectroscopy. Building one was a snap.
In trying to find out what I could do with it, I discovered that cerebral blood flow enhancement is a voluntary function. We do it all the time.
I could increase local cortical blood flow voluntarily. I could exercise my brain! From this Hemoencephalography was born.
I wanted to know what mattered in Neurofeedback. In my search for what matters I found many opinions. I wasn't satisfied until I had a found a physiological change that resulted from the training Attention Deficit Disorder (ADD/ ADHD). I studied the published articles on EEG training that used Test of Variables of Attention (TOVA) as the dependent variable. TOVA has a reputation of test retest reliability so that retesting after training should be free from learning effects.
The result was very strange. The best in the better EEG training group was only 3% above the average for that group. The best in the lower group was only 3% better than the average for its group. The lower group was
70% as good as the upper group.
After developing this graph, I entered the results of my use of Hemoencephalography (HEG), blood flow therapy, for 53 subjects. I was happy to learn that HEG worked significantly better than EEG.
Most surprisingly the HEG group was twice as good as the higher EEG group!
I was delighted when my very good friend, Paul Kwong, found the correlation between HEG and TOVA for the forehead placement was about 0.7 and the chance for error, p was less than .001.
I set out to find out why. For one thing I found it followed the same law as EEG on number of sessions although the gain per session was much higher. TOVA gain increased as the number of sessions increased. Also
training was at the frontal lobe near the eyebrows while EEG trained near the motor strip at the top of the head