Multitasking not always efficient and demands choices

Our working memory can only contain one element at a time. If we have to remember more elements we store these in our long-term memory, with the consequence that the retrieval of this information takes more time and leads to more mistakes. This is what Jelmer Borst concludes in his research on multitasking. Borst will be awarded a PhD on 20 April 2012 by the University of Groningen.

It was already known that multitasking works well as long as the same areas of the brain are not needed for the different tasks. Borst studied how our memory deals with different tasks. In his tests he got subjects to keep switching between two tasks. They had, for example, to add up two large numbers while in the meantime – with constant interruptions, ‘now go back to the sum’ – typing in a ten-letter word that was not visible on the screen.

Alternating between tasks

Alternating between these tasks was sometimes simple: the subjects only needed to remember one element at the most. Other times the alternation between tasks was less simple: the subjects then had to remember different elements at the same time. The latter is what usually happens during multitasking.

Model

Borst saw that the subjects were slower at these complex tasks and made more mistakes. As a result of his tests Borst postulated a model of the functioning of our working memory, the basis of which was the fact that the working memory only retains one element at a time.

Borst used this to develop a computer model that enables him to predict which tasks will go well. Furthermore, it makes it clearer which parts of the brain are involved in this. When he tested his predictions with a functional MRI they proved to be surprisingly correct. Borst has shown here that multitasking can only be efficient as long as multiple elements do not have to be remembered at the same time.

Follow-up question

Borst’s PhD research is part of the research by his supervisor, Niels Taatgen, Professor of Cognitive Modelling at the University of Groningen. Taatgen’s research began with the primary question of when multitasking goes well and when it does not. This question has now been answered: multitasking is not always efficient and it demands choices. Taatgen has therefore set to work on a follow-up question: how do people make choices when multitasking and what makes a choice good or not?

Support in multitasking

Taatgen believes that answering this question could help us to come up with ways to help people make better choices. Their working environment could be set up in such a way that they decide on the right moment to read their e-mail or actually to be open to interruptions from colleagues. Taatgen says, ‘This would not just allow people to work more efficiently but would also give them the feeling that they were getting their work done instead being at the mercy of their environment.’

CV

Jelmer Borst (Oenkerk, 1983) studied Artificial Intelligence and was awarded a Master’s degree in Behavioural and Cognitive Neurosciences by the University of Groningen. He is currently working as a postdoc at the Carnegie Mellon University in Pittsburgh, USA.

Note for the press

Further information:

Jelmer Borst, e-mail: jelmerborst gmail.com
Prof. Niels Taatgen