Neurofeedback training has been used to treat human epilepsy since the early 1970s . Research studies show that neurofeedback can be used to train the brain to reduce electrical activity linked to the generation of seizures while stimulating activity that instead is more likely to make seizures less frequent.
Recent advances in quantitative electroencephalogram (Q-EEG) and in neurofeedback protocol design today allow clinicians to dramatically reduce seizures or at least reduce the amount of medication required to control them. In particular, it is now possible to precisely determine anomalies in within-region activity (typically slow frequencies in the 3- to 8-Hz range  and between-region connectivity (coherence) in drug-resistant epilepsy reducing or even abolishing seizures in patients appropriately trained, many of whom even become medication-free.
For example, rewarding 12 to 16 Hz activity (SMR) at 20 to 25 mV over central and frontal areas has been found to reduce seizure frequency in clinically refractory patients undergoing sessions three times a week . Another study  found that training with a similar protocol was associated with seizure frequency reduction and normalization of the EEG in a patient with severe epilepsy.
In an important study, Lubar et al  found that training to inhibit 3- to 8-Hz activity and increase SMR was associated with a reduction in seizure frequency. Subjects were subsequently trained to increase 3- to 8-Hz activity and decrease SMR. This training was associated with an increase in seizure activity, which confirmed the specificity of the training effect with regard to seizure reduction or suppression. Importantly, they found that seizure frequency was reduced again after the 3- to 8-Hz frequency inhibit/SMR reward protocol was administered.
Finally, Ayers et al.  in a study with ten patients, found that the inhibition of 4- to 7-Hz activity and the reward of 15- to 18-Hz activity was associated with the suppression of seizures for 10 years.
Altogether, a review of 18 studies where biofeedback was used to treat epileptic seizures shows that out of 174 patients treated, 82% exhibited seizure reduction by 30% .
How Does Neurofeedback Lower Seizure threshold in Epileptic Patients?
It has been proposed that neurofeedback exerts its effects on the seizure threshold through optimization of functional connectivity throughout the brain. During neurofeedback training, the brain literally learns not to have seizures, slowly fine-tuning electrical activity only in the epileptic foci .
The neural mechanisms underlying the efficacy of neurofeedback in epilepsy are well understood as SMR appears to emanate from the thalamus , a small structure within the brain that relays sensory impulses from various parts of the body to the cerebral cortex.
Through a complex modulation of neurotransmitter activity, neurofeedback training has been shown to reduce the amount of sensory information that reaches the thalamus which in turn results in reduced or altered inputs to the areas of the cerebral cortex that control muscle tone [6, 7].
Neurofeedback is known to effectively and reliably reduce epileptic seizures through the reward of a specific EEG frequency range, eventually normalizing the neurochemical imbalances responsible for the low excitation threshold of the motor cortex.
- Sterman, M.B., Basic concepts and clinical findings in the treatment of seizure disorders with EEG operant conditioning. Clin Electroencephalogr, 2000. 31(1): p. 45-55.
- Lubar, J.F., et al., EEG operant conditioning in intractable epileptics. Arch Neurol, 1981. 38(11): p. 700-4.
- Finley, W.W., H.A. Smith, and M.D. Etherton, Reduction of seizures and normalization of the EEG in a severe epileptic following sensorimotor biofeedback training: preliminary study. Biol Psychol, 1975. 2(3): p. 189-203.
- Ayers, M., Long-term follow up of EEG neurofeedback with absence seizures in Society for Neuronal Regulation meetin. 1995: Las Vegas, NV. p. 6.
- Sterman, M.B. and L. Friar, Suppression of seizures in an epileptic following sensorimotor EEG feedback training. Electroencephalogr Clin Neurophysiol, 1972. 33(1): p. 89-95.
- Howe, R.C. and M.B. Sterman, Cortical-Subcortical EEG Correlates of suppressed motor behavior during sleep and waking in the cat. Electroencephalography and Clinical Neurophysiology, 1972. 32: p. 681-695.
- Howe, R.C. and M.B. Sterman, Somatosensory System Evoked Potentials During Waking bahavior and sleep in the cat. Electroencephalography and Clinical Neurophysiology, 1973. 34: p. 605-618.