Multiple demands on cognitive control in dual-task processing

We frequently perform two tasks in temporal overlap, even though this ordinarily leads to impaired performance of the individual tasks. Such dual-task-related performance costs have been extensively studied in cognitive psychology in order to get insights into the cognitive architecture and the mechanisms of information processing. The findings have sparked a decades-long theoretical controversy about the role of cognitive control in dual-tasking. Cognitive control enables us to handle complex task demands in a flexible and context-dependent manner and thus to act in a goal-directed way. In dual-task contexts, cognitive control might be required, inter alia, for shielding the processing of each task from interfering influences resulting from the simultaneous processing of another task and for the temporal coordination of multiple task processing. Until today, it is disputed whether dual-task-related performance costs are behavioral markers of cognitive control, or whether they are solely attributable to structural limitations of the cognitive system. These limitations prohibit simultaneous response selection and lead to the situation that the responses for the two individual tasks performed within a dual task are selected on a ‘first-come, first-served’ basis. Against this background, the main objective of the present cumulative dissertation was to re-examine the role of cognitive control in dual-tasking, thereby contributing to a better understanding of the mechanisms of cognitive control that underlie our ability to perform in an adaptive and goal-directed manner. To this end, two levels of dual-task processing on which cognitive control might act, namely the local level and the global level of dual-tasking, were investigated based on two lines of research. The first research line focused on the local level of dual-task processing. This level refers to the cognitive processing of the individual tasks performed within a dual task. To ensure the correct processing of an individual task, cognitive control might act upon the mental representations of individual tasks, which are denoted as task sets. To gain insights into the underlying mechanisms of such local task-set control, dual-task processing was related to cognitive processing in task-switching contexts. The performance costs that arise in task-switching contexts are commonly interpreted as markers of cognitive control. Based on the dual-task literature and the findings of Study I (Hirsch, Schwarzkopp, Declerck, Reese, & Koch, 2016), parallel task-set activation in dual-task and task-switching contexts was identified as a common ground for related mechanisms of cognitive control at the local level of task sets. To specify these related mechanisms of cognitive control, correlations between the performance costs arising in dual-task situations and the performance costs occurring in task-switching contexts were analyzed in Study II (Hirsch, Nolden, Declerck, & Koch, in press). The findings indicate for the first time that, in addition to or instead of the structural limitations that prohibit parallel response selection, dual-task-related performance costs reflect, at least partly, those cognitive control processes that are also measured by switching-related performance costs. The second line of research addressed the global level of dual-task processing, which refers to cognitive processing that is linked not only to one of the two individual tasks performed within a dual task but to both individual tasks at the same time. Using a novel empirical approach, the findings of Study III (Hirsch, Nolden, Philipp, & Koch, 2017) provide evidence for the first time that the identity of the individual tasks performed in a dual task is jointly represented in a single mental representation, termed task-pair set. This suggests that the two individual tasks performed in a dual task represent the subtasks of a more complex single task (e.g., starting the car, with the temporally overlapping subtasks of releasing the clutch and accelerating the gas). Consequently, the mental task representation needed for performing a dual task consists of multiple components, including a task-pair set organized at a hierarchically higher-level than the subtask-specific task sets. In Study IV (Hirsch, Nolden, & Koch, 2017), persisting task-pair set activation that results in residual positive priming of task-pair sets was identified as a crucial selection mechanism at the global level of dual-task processing.In summary, the findings of the present dissertation indicate that there are multiple requirements for cognitive control in dual-tasking. Cognitive control is not only required at the local level of dual-task processing to ensure the correct processing of the individual subtasks but also at the global level to enable the selection of an appropriate task-pair set. The task-pair set, in turn, ensures the specification of the individual subtasks to be performed in a dual task. These novel findings highlight the importance of considering cognitive control processes in future dual-task studies and models