In our modern society, complex technological systems typically require people to perform several tasks in a limited period of time. In order to be able to optimize such technological systems, the present thesis aims at explicating some difficulties with regard to the fundamentals of multiple-task performance theory - especially in relation to aging- and at providing some basic theoretical improvements. First, the fundamental assumptions of the main recent theories concerning multiple-task performance. In brief, these theories are based on the notion that human performers possess one or a few 'pools' or supplies of central limited-capacity resources. A theoretical analysis shows that these current frameworks offer rather trivial explanations, lack neurobiological support, efficiently account for behavioral plasticity with training. The prevalent conception of attention as 'central supervisory control' does not provide an explanation for what has to be explained, i.e., the control process itself. In this connection, a cognitive neuroscience framework is invoked, which starts with current knowledge concern-ing the basic principles of brain function-ing in combination with the nature of capacity limitations in human performance. According to this framework, the problem of limited behavioral capacity is explained by the way biological systems satisfy two conflicting requirements: massive associative processing power and flexibility against coherent and goal-directed action control. This control conflict is not solved by a mysteri-ous supervisory attentional system. Behavior control basically emerges from elementary, self-regulating, characteris-tics of neuronal information processing. In addition, neurobiology and psychologicanervous system is well-suited for integrated information processing. Hence, atten-tional limitations generally occur when perceptual, cognitive, or motor opera-tions have to be segregated in task performance, whereas the potential efficien-cy of information processing and action increases with the degree to which dual-task elements are related or coherent, such that subtasks can be performed as a whole (skill integration). According to the cognitive neuroscience framework, presented in this thesis, aging-related neuronal de-cline may provide an appropriate, and a neurobio-logi-cally well-founded, basis for explanati-ons for aging-related functional problems. Decreased neuronal connectivity and plasticity may produce several phenomena, such as dealing with emerging task characteristics that affect the integration or segregation of skills (brain programs), to suppress the activation of irrelevant (dominant) routines or to modify these, or to process informa-tion quickly. The second, experimental part, of this thesis reports four experiments aimed at demonstrating that these mecha-nisms may explain dual-task deficits encountered in old age.