Abstract
In this paper, we explore the possibility of applying multimodal feedback to improve multitasking performance. For this purpose, we have devised a general multitasking test application, called the MSP-Blocks, which includes many basic elements of multitasking and can be used to carry out a variety of multimodal multitasking experiments. An experiment was run to study the effects of two factors (the number of jobs and types of multimodal feedback) to user task performance, specifically, interaction effort, concurrency, fairness and output quality. The results indicated that multimodal feedback did influence multitasking performance, and moreover, non-redundant multimodal feedback was more effective than no multimodality or redundant multimodality for tasks with reasonable difficulty, e.g. when the number of jobs was more than four.
Daniel (Danny) Hyeongcheol Kim
Human-Centered AI
Researcher & Builder
HCAI researcher and builder, Postdoctoral Research Fellow at KAIST and co-founder of KEONIX Labs.