Method: Seven children diagnosed with attention-deficit hyperactivity disorder (ADHD) were trained using a standard EEG biofeedback treatment protocol designed to alter SMR/theta ratios and reduce behavioral symptomatology diagnostic of ADHD. During alternate periods they were also trained using a placebo protocol that was identical to the treatment protocol, save that the association between EEG patterns and feedback to the participants was random. Single-case design elements were used to control for the effects of internal validity threats such as maturation, history, and treatment order. Two participants failed to complete all training sessions, and the effects of training on behavior were analyzed both including and excluding these non-completers.
Results: When all participants were included in analyses that controlled for overall trend, EEG biofeedback was found to be no more effective than the placebo control condition involving non-contingent feedback, and neither procedure resulted in improvements relative to baseline levels. When overall behavioral trends unrelated to training were not controlled for and noncompleters were excluded from the analysis, it could be mistakenly concluded that EEG biofeedback is significantly more effective than placebo and that the effect sizes involved are moderate to large. These results indicate that many previous reports of the efficacy of EEG biofeedback for ADHD, particularly those presenting series of single cases, might well have been based on spurious findings.