Neurofeedback and working memory training in children and adolescents with ADHD

Abstract

Background: ADHD is one of the most prevalent neurodevelopmental disorders among children and adolescents. Although pharmacological interventions are highly effective in attenuating symptoms of inattention and impulsivity, adverse side effects are common, warranting the need for alternative treatments. Focused on “training” the brain, Neurofeedback (NF) has received much attention in recent decades, with promising results for ADHD. However, meta-analyses indicate somewhat mixed results, including variations between different neurofeedback protocols.

Aims: The aim of this thesis was to examine the efficacy of two NF protocols as treatment for children and adolescents with ADHD. In addition, this thesis also tried to add further understanding on the self-regulation process during SCP-NF, by qualitatively examining the use of regulatory strategies. Furthermore, this thesis examined how standardized SCP-NF is, by systematically reviewing the literature. The main focus concerned the technical implementation, and in particular how successful self-regulation is evaluated. For this a was conducted.

Methods: In total, 202 children and adolescents were randomized into one of the four groups (n=50-51), Slow Cortical Potential NF (SCP-NF), Live Z-score Training (LZT), Working Memory Training (WMT), or Treatment-as-usual-only (TAU-only). The groups were then compared with each other, and analyzed in a linear-mixed-model. Both primary symptoms from self-, parent- and teacher rating, as well as a battery of cognitive performance tasks, were evaluated at baseline, post-intervention and at a 6-month follow-up. Additionally, a thematic analysis of 133 short semi-structured interviews was conducted for the qualitative study. The systematic review was preregistered at PROSPERO (CRD42021260087) and followed the PRISMA guidelines.

Results: Contrary our expectations, between-group differences were scarce and did not show a distinct pattern, concerning our primary outcome measures in study II. Improved teacher-rating were found for LZT over TAU-only at 6-mont follow-up. For SCP-NF we only found significant differences on meta-cognition compared to TAU-only. Concerning the cognitive tests examined in study III, sustained improvements were only found for spatial working memory in WMT. The thematic-analysis of study I, revealed that numerous different strategies were used in SCP-NF. Furthermore, three prototypical training styles were identified, where only one style described intrinsic motivation and self-perceived improvements. The same profile was the only one that had a positive trend in their self-regulation. For Study IV, of the 800 initial search hits, 63 were included and synthesized. The review revealed broad variations in protocol details, and in how successful self-regulation is evaluated, which limit comparability. Also, data concerning adequate skill application in everyday-situations, is sparsely reported.

Conclusions: We could not find support for NF as broadly implemented treatment for ADHD, when delivered at a high frequency. However, this must be interpreted with caution due to limitations. Future studies should focus on proper self-regulation and its significances on the symptomatology.