Two studies [Ivry, R. B., Franz, E. A., Kingstone, A., & Johnston, J. C. The psychological refractory period effect following callosotomy: Uncoupling of lateralized response codes. Journal of Experimental Psychology: Human Perception and Performance, 24, 463–480, 1998; Pashler, H., Luck, S., Hillyard, S. A., Mangun, G. R., O'Brien, S., & Gazzaniga, M. S. Sequential operation of disconnected hemispheres in split-brain patients. NeuroReport, 5, 2381–2384, 1994] reported robust dual-task costs in split-brain patients even when the two tasks were associated with separate cerebral hemispheres. Although the patients failed to demonstrate specific forms of interference observed in control participants, the timing of the two responses suggested that performance was constrained such that the responses could not be initiated independently. Alternatively, the split-brain participants may have adopted a strategy in which the second response was withheld until the first was initiated. The present study revisits this phenomenon using a procedure in which the stimuli for both tasks are presented simultaneously and neither is given priority over the other. Under these conditions, neurologically intact participants show robust dual-task costs that are mediated by compatibility effects between the responses of the two hands. In contrast, the split-brain participants show greatly reduced dual-task costs and compatibility effects. The minimal dual-task costs observed in the current study indicate that previous dual-task costs in split-brain patients may be strategic, reflecting experimental instructions to prioritize one task, rather than reflect fundamental constraints of the cognitive architecture.