Effects of training on the parameter adaptation in man-machine-interfaces in medical engineering

Abstract

Man-machine-interfaces in medical assistance systems often require an adaptation of its parameters to the user. However, there is no consistent approach regarding the way and the frequency of the adaptation. Simple systems are calibrated by hand, more complex systems use online-adaption algorithms. Only little is known about the impact of changing parameters on the humanability to operate the system reliably. Here, we present a study comparing three ways of parameter adaptation by measuring the performance of 20 human subjects by means of gamification. We provide evidence, that an adaptation is necessary for intuitive operability and that the human ability to adapt to the system has a great impact on the performance. Therefore, parameter adaptations must be too extensive.