While it is often said that our 'modern world is a multitasking world' it has become more and more clear that multitasking is not necessarily a good thing. Many studies have shown that while we spend much of our time performing multiple tasks at the same time, in general this leads to a decrease in performance. The topic of this dissertation is human multitasking, and in particular the so-called problem state bottleneck: one of the reasons why multitasking is often counter-productive. While in general humans are extremely good at multitasking – when do we truly do one task at a time? – in certain situations our ability to multitask breaks down. On the one hand, multitasking is obviously limited by physical constraints. We simply cannot look at two things at the same time, as texting cyclists prove daily during the authors ride to work. More interestingly, there are also limitations in our cognitive system that hinder multitasking. One of these is the problem state bottleneck: a limitation in processing intermediate representations that are necessary for a task. To support the idea of a problem state bottleneck, the author presents several experiments and a computational theory of how intermediate representations are processed in our minds. As support for this theory, he not only looks at behavioral data, but also relates the theory to neuroimaging data. (Author/publisher)
Abstract