Working memory load effects of different tasks during multitasking may differ: an FMRI study

Abstract

Multiple demand (MD) regions are well-known for their activation pattern in task-related events and are especially sensitive to working memory load (WM) conditions. Different WM loads deplete varying amounts of WM capacity and may lead to interference. However, whether MD regions are sensitive to precision in addition to cognitive load still remains unclear. This thesis investigates whether challenging multitasking induces interference and whether brain responses to demanding tasks are modulated by precision. Thirty healthy participants (M = 27.1) underwent two fMRI runs and completed a hierarchically nested visual WM task, where task A’s items were maintained during task B under alternating high and low WM loads. Parametric modulation, whole-brain, and region-of-interest (ROI) analyses were performed. Results showed that MD regions were activated during the encoding phase of task B, whereas task A exhibited a deactivation pattern. However, tasks were not differentiated during their corresponding retrieval steps. Brain regions correlating with performance accuracy also did not differ. The Default Mode Network (DMN) showed sustained deactivation during recall, regardless of task type, indicating domain-general suppression mechanisms. These findings support hybrid WM models that allow flexible control mechanisms and discrete capacity limits. The MD network appears to adjust engagement depending on task context and phase, aligning with hierarchical models. Taskspecific encoding but shared recall dynamics suggest that while storage may be selectively distributed, retrieval relies on generalized control. This enhances understanding of cognitive control–memory interactions and highlights the need to examine individual differences to inform interventions for WM-related disorders.