BACKGROUND: The need for and the interest in non-pharmacological treatments for children with ADHD are increasing. The treatments include electro-encephalogram (EEG) frequency-neurofeedback and Cogmed working memory training. AIM: To investigate the efficacy of frequency-neurofeedback and Cogmed working memory training in children with ADHD. METHOD: Forty-one children with ADHD (aged 8-15 years) were assigned to frequency-neurofeedback or to placebo-neurofeedback in a randomized double-blind trial. We took measurements to find out whether frequency-neurofeedback had reduced the severity of the ADHD-symptoms, and/or had improved neurocognitive ability and global clinical functioning. Fifty-one children with ADHD (aged 5-7 years) were assigned to the active Cogmed JM-working memory training or to the placebo working memory training in a randomised double-blind trial. We took measurements to find out whether Cogmed JM-working memory training had reduced the ADHD symptoms, and/or had improved neurocognitive ability, daily performance and global clinical functioning. RESULTS: The ADHD symptoms and global clinical functioning of the children in both neurofeedback groups improved. However, frequency-neurofeedback did nor produce any significantly better treatment results than did the placebo neurofeedback. At the neurocognitive level, frequency-neurofeedback did not yield any measurements that were significantly superior to those achieved with placebo feedback. Various outcome measurements improved in both groups with memory training. However, the active working memory training was not found to have produced significantly better results than the placebo training with regards to the ADHD symptoms, neurocognitive ability and daily and global functioning. Children from the active working memory training group showed improvements in trained working memory tasks but not on untrained tasks. CONCLUSION: Neither study produced any conclusive evidence for the efficacy of the investigated treatments in children with ADHD. However, both types of treatments can be further improved. Furthermore, the controlled designs may have restricted the embedding of the treatments. Because of possible improvements in the treatments in the future and because of the design restrictions affecting the treatments in their present form, it is still too early to draw any definitive conclusions about the validity and advantages of the two treatment methods.