A previous study showed that pairs of students interacting in a second language produced more words when they were assigned a fictitious expert position in a specific competence dimension than when they were assigned a nonexpert position. It has also been shown that the usual level of expertise has an impact on the assigned fictitious expertise effect. The present study was designed to determine whether the processing capacity allotted to the current task could partly determine performance. A given position of expertise may demand a large or small attentional capacity. Two experiments were conducted using a dual-task paradigm. As expected, the different expertise positions led to different reaction times on the secondary task. The second experiment showed that the impact of assigning a position of expertise to students depends on their usual academic standing. This study supports the idea that in interactive situations, performance variations as a function of the expertise position can be partially explained by differences in the processing resources allocated to the task. Implications for teaching are discussed. Copyright 2001 Academic Press.