L’entraînement de la mémoire de travail est-il bénéfique pour les enfants présentant un trouble déficit de l’attention/hyperactivité ?

Abstract

IntroductionChildhood attention-deficit/hyperactivity disorder is known to be associated with impairment across multiple domains, including social, familial, emotional and academic functioning. Available therapies, and in particular medical treatment, fail to produce lasting improvement in academic and learning outcomes. In this context, recent years have seen a growing interest of experts and practitioners for computer-based cognitive training programs. Indeed, it has been proposed that repeated training of specific executive functions may result in lasting benefits for children with ADHD, especially in the school setting. Cogmed RoboMemo, a program targeting working memory, is the cognitive training that has been the most extensively studied. Literature findingsMeta-analyses have established that the vast majority of children with ADHD present with decreased performance levels compared to typically developing children when performing working memory tasks. Working memory deficits, in turn, have been linked with classroom inattention and decreased academic performance. Based on these findings, models describing working memory deficits as a core feature of ADHD, accounting both for inattentive and hyperactive symptoms, have been proposed. In this theoretical context, interventions specifically targeting working memory deficits in ADHD appear as particularly promising. However, despite this solid theoretical background, meta-analyses of randomized controlled trials relying on the Cogmed RoboMemo program have consistently failed to evidence significant therapeutic effects on key targets, including ADHD symptoms or academic performance. These results suggest that early reports of clinical efficacy with non-blinded assessors were confounded by placebo and expectancy effects. DiscussionDifferent hypotheses have been proposed to explain this lack of significant therapeutic effect. First, it has been observed that Cogmed RoboMemo primarily targets the storage/rehearsal capacity of the child, whereas children with ADHD appear mostly impaired when performing tasks with a strong central executive load (i. e. requiring manipulation and dynamic updating of information). The Cogmed cognitive training program may therefore simply focus on the wrong element of working memory in the context of ADHD. Second, authors have questioned whether RoboMemo practice truly results in increased working memory capacity, proposing that children may simply learn task-specific strategies that they are unable to use in a broader context. In line with this idea, it has also been suggested that training in a non-ecological situation could be insufficient, if children with ADHD do not also learn strategies to apply the capacity acquired in the class context. Finally, one recent randomized control trial suggests that potential methodological limitations (lack of follow-up and/or limited training intensity) may have prevented previous studies from documenting the beneficial effects of the Cogmed program. ConclusionsMeta-analyses available to date indicate that claims regarding the academic, behavioral, and learning benefits associated with Cogmed RoboMemo are unsupported in ADHD. New studies with a follow-up assessment are needed to ensure that beneficial long-term effects of the program have not been overlooked. The development and evaluation of new cognitive training programs targeting the central executive component of working memory is warranted. More research is also necessary to help understand how working memory capacity of children with ADHD could be in improved in an ecological context.