Hierarchical task organization in dual tasks: evidence for higher level task representations

Abstract

To examine whether hierarchical higher level task representations comprising the task sets of Task 1 (T1) and Task 2 (T2) are activated within each trial in dual-task situations, we combined the psychological refractory period (PRP) paradigm with the task-pair switching logic (Hirsch et al. 2017). In Experiment 1, in which subjects switched between task-pairs including a varying T1 and a constant T2, we found a PRP effect (i.e., worse performance with short stimulus onset asynchrony [SOA] than with long SOA) and task-pair switch costs in T1 and T2 (impaired performance in task-pair switches compared to task-pair repetitions). However, since in Experiment 1 there were no forward and backward response-response compatibility effects that indicated interference between T1 and T2, we could not exclude that the activation of T1 persisted into the next trial despite the intervening T2, and hence, that task-pair switch costs are due to repetition-priming effects of T1 across task-pairs rather than due to persisting activation of task-pair representations. In Experiment 2, we used a modified task-pair switching logic with a constant T1 and a varying T2, and replicated task-pair switch costs under conditions that not only rule out repetition-priming effects of T1 across task-pairs as the source of task-pair switch costs but also disentangle the effects of switching task-pairs from those of switching T1. These effects were associated in previous studies using the original task-pair switching logic. Thus, the findings of the present study strongly suggest that hierarchical higher level task representations are activated during dual-task processing.

Evidence for a multicomponent hierarchical representation of dual tasks

“…Task-pair switching logic Further evidence for cognitive processing at the global level of dual-tasking comes from studies on hierarchical task organization in dual tasks by Hirsch and colleagues (Hirsch et al, 2017(Hirsch et al, , 2018. They developed a novel empirical approach, called "task-pair switching logic".…”

“…Performance is typically worse in task-pair switch trials than in task-pair repetition trials, leading to task-pair switch costs in both T1 and T2. The task-pair switch cost is a robust effect that has been found with distinct task types for T1 (i.e., T1 with speeded manual responses or deferred nonspeeded vocal responses; e.g., Hirsch et al, 2017;Koch & Rumiati, 2006), varying extent of overlap in the response sets of T1 and T2 (i.e., conceptual and physical response-set overlap), and several numbers of task pairs (i.e., two and three; Hirsch et al, 2017Hirsch et al, , 2018.…”

Adjustment of cognitive control to the frequency of dual-task interference

“…However, as far as the current data are concerned, it seems that adapting to a single T1 during frequent T2 does not entail switching costs related to accuracy of performance. On the other hand, we must note that among a series of experiments on hierarchical task processing published by Hirsch and colleagues Hirsch, Nolden, Philipp, & Koch, 2017), there was only one that included a similar condition, i.e. adaptation to changes of only secondary task settings.…”

“…adaptation to changes of only secondary task settings. In that study, the authors used the same T1 for all blocks of trials and presented T2 or T3 interchangeably as a secondary task (Experiment 2, Hirsch, Nolden, Philipp, & Koch, 2017).…”

“…This suggests that at a global level, cognitive control mechanisms can group dual-task settings together to create a higher-level single task, with participants forming a mental representation of both task sets. Such hierarchical representation has sustained activation, thus when the task pair changes, an update of both tasks is needed, despite the response setting for one of them being already available at the level of cognitive control (Hirsch, Nolden, Philipp, & Koch, 2017). Interestingly, other studies have also shown performance costs being a consequence of a previous task requiring a response to either primary or both tasks (Olszanowski, Bajo, & Szmalec, 2015).…”