There is a newer study from 2010 with a bigger sample (N>500) consisting of children (around twelve years old) that aimed to reproduce the cognitive improvements from previous findings. The authors also tried to eliminate methological problems from the study mentioned in the article.