Schellenberg (2011) reports the results of an experiment to determine the cause of the association between music training and intelligence quotient (IQ) in children. There was no difference between trained and untrained groups in a set of executive function tasks and no evidence that executive function mediated the effects of music training on IQ, since music training and IQ remained related after controlling for executive function performance. Therefore, he concludes that children with high IQs are more likely to take music lessons, assigning responsibility for the reported correlation to genetic differences in intelligence. The present article challenges that conclusion by discussing the nature of the experience represented by children in the two groups, the validity of the tasks used to measure executive functioning, and the logic of the conclusion from the reported data.