Working memory training in children with neurodevelopmental disorders and intellectual disabilities, the role of coaching: A double-blind randomised controlled trial.
Working-memory training works for kids with NDD and mild ID, but a fancy personal coach adds zero extra benefit.
01Research in Context
What this study did
Kids with autism, ADHD, or intellectual disability used a computer game called Cogmed. The game gets harder each time the child wins.
Half the kids got a coach who talked only to them. The other half got a coach who gave the same talk to every child. No one knew which coach was which.
After five weeks the team tested memory, school work, and behavior to see if the personal coach helped more.
What they found
Every child got better at memory, school tasks, and behavior. The extra personal coach did not add any gain.
A plain coach who keeps the child on task works just as well as a tailor-made coach.
How this fits with other research
Freeman et al. (2015) watched autistic kids for two years and saw their verbal memory stay flat. Howard et al. (2023) now shows you can move those scores if you train hard, so the plateau is not fixed.
Kleinert et al. (2007) and Settanni et al. (2023) both found that coached parent training helps parents teach new skills. Howard et al. (2023) shows the same coach idea does not help when the child is already glued to a computer game.
Dai et al. (2025) blended hospital and home coaching and saw big child gains. Howard et al. (2023) kept coaching inside the game only, so the wider carry-over seen by Dai may need real-life practice, not just screen time.
Why it matters
You can add Cogmed to a middle-school plan for kids with NDD and mild ID. Just make sure an adult is nearby to keep the child on task. Skip the costly one-to-one coach package; the basic coach script is enough. Track memory gains, but do not expect them to spill into class work without extra teaching.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Run the standard Cogmed protocol and use any staff member to give the generic coach prompts—save your individualization time for other goals.
02At a glance
03Original abstract
BACKGROUND: Working memory training (WMT) can offer therapeutic benefits to patients with neurodevelopmental disorders (NDD) and mild to borderline intellectual disability (MBID). However, consistent evidence for treatment benefits of WMT over placebo training is missing. So far, participants in double-blind research designs did receive non-specific coaching, whereas active coaching based on individual training results might increase the efficacy of WMT. Furthermore, the intensity and duration of WMT is often too stressful for these children. This study therefore investigated whether a less intensive but more prolonged WMT, with active personalised coaching and feedback, would reduce behavioural symptoms and improve neurocognitive functioning and academic achievements in children with NDD and MBID. METHOD: A double-blind randomised controlled trial in children (aged 10;0-13;11) with MBID (60 < IQ < 85) and ADHD and/or ASD evaluated the effects of a less intensive but prolonged version of the original Cogmed WMT (30 min a day, 4 days a week, 8 weeks in total). Eighteen participants received active, personalised coaching and feedback, based on their actual individual performance during training. Twenty-two received general non-personalised coaching for the same amount of time. Executive functioning, academic achievements and several behavioural measurements were administered, before and after training, with a 6-months follow-up. RESULTS: We observed a significant effect of time on both primary and secondary outcome measures, indicating that all children improved in working memory performance and other neurocognitive and academic outcomes. The interaction between time and group was not significant. DISCUSSION: This study was unable to document superior effects of active personalised coaching and feedback compared with general non-personalised coaching and no feedback in an adaptive WMT in children with MBID and NDD. The objectively documented changes over time suggest that for these vulnerable children, a regular, structured and structural contact with a coach and adapted exercises is enough to develop therapy fidelity, boost motivation and improve neurodevelopmental task performance. Further research is needed to examine which possible subgroups within this heterogenic group of children profit more from WMT compared with other subgroups.
Journal of intellectual disability research : JIDR, 2023 · doi:10.1111/jir.13047