task, the n-back 3 task. This task was performed as a single task and concomitantly with the dual task. Performing multitask impaired cognitive and postural task performance. More errors and greater mean time to answer to the cognitive task were observed. At the postural task we found differences on motor strategies on the cognitive task condition: greater joint flexion at hip and shoulder, greater trunk linear displacement, greater trunk and tray velocity, as well as a reduction on gastrocnemius magnitude. At this condition we also found that the participants took longer to revert the anterior oscilation to return to a more stable position. Therefore the integration between tasks was successful, given the interference between the attentional processes and/or necessary resources to accomplish the tasks. It is possible that may have been withdrawal of attentional resources form the sensorial inputs processing associated with the postural control. That would affect central set, producing an inaccurate influence on postural answers to perturbation. Manual task produced positive effect on postural task. On the higher restriction condition it was found a more efficient postural answer, with less joint range on shoulder, hip and ankle. Additionally, the trunk variables: linear displacement, velocity and verticality, as well as tray velocity, also were inferior. This study indicates that central set is elaborated based on functional restrictions imposed by suprapostural tasks and it can influence postural task, even with a parallel processing in order to execute a concomitant cognitive task.Competition between processes and/or resources to accomplish the proposed tasks affects balance, with negative results when a cognitive task is present, and with positive results when there is a manual restriction.