Cognitive Neuroscience Robotics A

pp 153-171

System Evaluation and User Interfaces

  • Hideyuki NakanishiAffiliated withGraduate School of Engineering, Osaka University Email author 

* Final gross prices may vary according to local VAT.

Get Access


It is essential for developing a useful system to evaluate its user interface. The interface should be evaluated by potential users instead of experts, if unbiased results are to be obtained. It is more recommendable to observe users’ reactions to an actual working prototype of the interface than to simply ask them about what kind of interface is preferable. The observation in a laboratory room can produce more general and scientific results than that in a real-world situation does. Thus, this chapter describes how to conduct a laboratory study. It collects users’ reactions to the prototype of a user interface in a controlled situation. There are many items to consider in order to conduct a laboratory study: goals, hypotheses, factors, conditions, experimental design, tasks, subjects, data collection, and analysis. Each of these items is explained in detail. An example of system evaluation that is conducted in the past study on telepresence robots is discussed.


Human-computer interaction (HCI) Usability User study Iterative development Order effect Between-subjects experiment Ceiling effect Human-robot interaction (HRI) Questionnaire ANOVA