ABSTRACT
Alpha/theta neurofeedback training is supposed to enhance cognitive performance as well as alleviate psychiatric disorders. Sleep during this training is regarded a confounding factor, although it is usually not controlled. Here, the amount of sleep, the impact of sleep on frequency results and the validity of subjective judgments of having fallen asleep were investigated. 40 healthy young subjects participated in 31 training units of real or mock feedback. It turned out that stage 2 sleep occurred in 10-14 % of training time, in 95 % of the subjects and in 1/3 of all training units. Sleep significantly influenced the theta/alpha ratio and theta but not alpha amplitudes. Only about 2/3 of the subjective judgements were correct and no more than 13 % of the participants could reliably identify sleep. These data demonstrate that sleep can easily occur during alpha/theta neurofeedback training and should be controlled by objective means.
