The German Society of Biofeedback (“Deutsche Gesellschaft für Biofeedback“) is a practice-oriented professional society, proceeding in line with scientifically consolidated research findings. We therefore consider it to be essential to comment critically on (positive and negative) scientific results with regard to biofeedback and neurofeedback from time to time and to provide you with the results or partial results of the current discussion.

The present issue deals with a news published on 18.08.2017 on the homepage of the University of Tübingen (Eberhard Karls Universität Tübingen). The corresponding study was made between February, 2013 and December, 2015 headed by Tübingen scientists ( You can find the relevant publication (in English) under the following link:
Principally, we welcome such a well-designed study as far as the mere operationalizing and the design of the study are concerned (triple-blind design). The following issue, however, is to be regarded as critical:
In order to keep the procedure of neurofeedback constant in the sense of experimental comparability, the authors use "automatic threshold training". Thereby, the used software continually adjusts the threshold (at intervals of 15 seconds) so that a positive feedback of 80% of the training period is achieved (positive feedback: the animation chosen as feedback is moving). Specifically, this means that the threshold adjusts to the amplitudes of the measured frequency bands or the measured   theta/beta ratio provided by the respective brain. Moreover, that means: in only 20% of the time the brain has to adjust to the threshold at all. Clearly stated: the automatic threshold training is no training condition used in practice since the brain’s adjustment process to external stimuli (which is one crucial issue in the case of ADHD) is not trained by that method. This point is also emphasized in the framework of DGBfb neurofeedback workshops.
As to the issue " automatic threshold": For completeness, also SCP neurofeedback has to be mentioned where an automatic threshold is used as well. The difference as opposed to the method as mentioned above lies in the fact that the automatic threshold measured by the software for a certain training interval (“trial”) always refers to a value shortly measured before. This value is to be exceeded for a defined period of time, or the value has to be lower than the one before, depending on the direction of the training. During SCP training, the brain therefore continually receives and reacts to an external request.

Classification of significance: This study’s statement about the efficiency of neurofeedback training ONLY refers to the treatment of adults between 18 and 60 years. A comparison with the neurofeedback treatment of children is therefore not (directly) possible.

Note for comparison of reported values with relevant literature: All the reported theta/beta ratios are logarithmized (natural logarithm) => The e- function must therefore be applied in order to receive the “regular” theta/beta ratios. The ratios cover a range of 2,58-5,42 (values through application of the exponential function to the values as per table 5 of the publication).

In this case, there has not been made a final scientific statement yet, however, we would like to provide you with the status quo of the discussion in brief.

We would like to point out to the fact that the clarification of the issues is mostly only possible in direct scientific discussion, and for that reason we are hereby inviting the authors to our next (possible) annual meeting of the DGBfb e.V. in 2017 or in 2018.

We are looking forward to vivid discussions, in the sense of new ideas and suggestions, and not in the sense of judgement of what is wrong and what is right.
Best regards,

The Board Members of DGBfb e.V.

(WE would like to thank Ms PD Dr. Ute Strehl who brought this study to our attention as well as all the members who provided their input during the discussion).