Community Research and Development Information Service - CORDIS


MULTITASK Report Summary

Project ID: 283597
Funded under: FP7-IDEAS-ERC
Country: Netherlands

Final Report Summary - MULTITASK (Towards safe and productive human multitasking)

How do people choose and prioritize the tasks they take on in a multitasking situation? The popular press often lets us believe that multitasking consists of rapid switching between single tasks. We conducted fMRI studies that show that this idea is not consistent with the data, and other experiments have shown that in many situation people are actually quite good at multitasking.
From several experiments we have concluded that decisions in multitasking are driven by the availability of cognitive resources (vision, manual, working memory, etc.). This means that if someone performs a task that requires only vision and manual control, like driving a car on an empty road, they will seek out secondary tasks that do not require vision or manual control, like listening to the radio or carrying out a conversation. In other words, people optimize their multitasking on the basis of local criteria instead of global criteria. In many cases this works well. We found that driving quality actually improves when people listen to the radio. There are, however, many situations where local criteria lead to suboptimal or even dangerous global behavior. In the case of multitasking during driving, the traffic situation can suddenly change, with the result that driving suddenly needs working memory for planning. If the secondary task people are doing also requires working memory, this leads to a conflict and a potential dangerous situation.
Apart from dangerous multitasking behavior during driving, a second example of suboptimal multitasking is if people interrupt themselves during work. For example, they are writing a report, and interrupt themselves to check email. Too many self-interruptions lead to fragmented work patterns, which, as a result, produces suboptimal performance and decreased work satisfaction. A major cause of these interruptions is pauses or impasses in the main task. During a pause, many resources are unused, enabling other tasks to intrude and grab these resources, resulting in a self-interruption.
Although the availability of resources is a major factor in decisions about multitasking, this does not mean there is no means of controlling or changing it. The project has produced a new cognitive architecture (PRIMs) that can be used to investigate the consequences of multitasking, and how training can help to become better at multitasking. In addition, we have developed a methodology to determine what the best moment is to interrupt someone based on the size of the pupil.

Reported by

Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top