Cogmed improves working memory and attention in children with Epilepsy

Research institution: Karolinska Institute

Research title: Combined cognitive testing and EEG recording before and after working memory training in children with epilepsy – a randomized placebo-controlled trial

Researchers: Westerberg, H., Dahlin, M., Persson, R., Berglund, G., Molin, K., Skoog, A.B., Forssberg, H., Stödberg, T.

Training program used in research: Cogmed RM

Presented: Poster presented in 2006 at the 4th World Conference for NeuroRehabilitation in Hong Kong

Children with epilepsy (EP) struggle with impaired working memory (WM), attention, mental speed and school performance. It is possible that the transient cognitive impairments experienced by this population are due to clinical interictal epileptiform activity (IEA), ictal and postictal effects of seizures, brain lesions, side effects from antiepileptic drugs, and/or sleep disruption. Stemming from earlier studies showing that cognitive abilities can be improved by WM training (Klingberg et al., 2002, 2005; Olesen et al. 2004) researchers investigated how training impacted the WM-related brain activity in children with epilepsy as measured by EEG and the effect of WM training on cognitive functions such as working memory capacity, attention, inhibition and reasoning.

Seventeen children aged 8 to 14 years with a diagnosis of epilepsy were randomly assigned to either an adaptive WM training group (n = 9) or a non-adaptive placebo training group (n = 8). Each participant trained for about 30 minutes, 5 days a week for 5 weeks. The adaptive training automatically adjusted the difficulty of visuo-spatial and verbal WM exercises to push the participant to their WM capacity. The placebo-control training included the same exercises but did not adapt in difficulty level to meet the performance of the participant and thus, served as a control for non-specific test-retest effects. The transient effects of IEA were measured with EEG during each training session. Participants were assessed for reaction time, verbal WM, visuo-spatial WM, inhibition (Stroop interference task) and reasoning and problem solving (Raven’s Progressive Matrices) prior to training, immediately after training and at a 3 month follow up.
Results: The adaptive WM training group evidenced improvement from baseline to post training compared to the placebo group on WM and reaction time tests. There was no significant difference between the groups in interference control or reasoning post training. The treatment group differed significantly from the placebo group on reaction time measures of vigilance, attention and speed of cognitive processing. Although the difference between the groups was no longer significant at 3 month follow up, 60% of the training effect remained.

This study was the first to examine the efficacy of Cogmed Working Memory training in a population of children with epilepsy. Not only was it possible to train this group but also, children who performed adaptive training improved in WM and attention.

Although analysis of the EEG data had not yet been performed at the time of this presentation, a subsequent manuscript from this study revealed an increase in frontal and parietal alpha band power in the adaptive group after visuo-spatial WM training exercises. It was posited that this increased band power may have been due to increased activity in the fronto-parietal visuo-spatial WM network.