Complexities related to cognitive slow wave experiments: a reply to Rösler and Heil

Rösler and Heil (1991) were unable to replicate one of the results of our previous study of slow wave activity and mental arithmetic (Ruchkin, Johnson, Mahaffey, & Sutton, 1988). Thorough examination of the error rates indicates that the greater complexity of their response selection procedure caused crucial differences between how the task was performed in the two studies. Thus Rösler and Heil's study emphasizes the need to avoid using a response that is so complex that it produces a superfluous task that obscures the results.