Perspectives on Type III Statistical Errors: Exaggerating the Effects of Placebo in Neurofeedback

Evaluating the efficacy of electroencephalography neurofeedback (EEG-nf) for the treatment of attention-deficit/hyperactivity disorder (ADHD) has been a topic of vigorous debate over the past few years. However, many of the articles state a lack of efficacy and insist on placebo as the explanation for any positive effects found in the EEG-nf treatment group. Several critical flaws in this analysis are discussed including the existence of non-inert shams, the false no-effect, and placebo as an ad hoc explanation. These flaws lead to Type III statistical errors, which are often repeated in other articles. It is recommended that journals, books, and media articles publishing new research and reviews on the efficacy of EEG-nf be vigilant for these errors in order to improve the quality of the EEG-nf body of research. Requiring researchers and authors reviewing the literature to verify assumptions of non-inert shams, ensure the use of best practices in the EEG-nf treatment groups, and clearly identify ad hoc conclusions can avoid these Type III errors.