EMU in the Car: Evaluating Multimodal Usability of a Satellite Navigation System

Abstract

Abstract. The design and evaluation of multimodal systems has traditionally been a craft skill. There are some well established heuristics, guidelines and frameworks for assessing multimodal interactions, but no established methodologies that focus on the design of the interaction between user and system in context. In this paper, we present EMU, a systematic evaluation methodology for reasoning about the usability of an interactive system in terms of the modalities of interaction. We illustrate its application using an example of in-car navigation. EMU fills a niche in the repertoire of analytical evaluation approaches by focusing on the quality of interaction in terms of the modalities of interaction, how modalities are integrated, and where there may be interaction breakdowns due to modality clashes, synchronisation difficulties or distractions.

Using Formal Verification to Evaluate Human-Automation Interaction: A Review

“…Many erroneous human behaviors have predictable cognitive causes [12] related to human working memory, human knowledge, human perception, or human physical coordination. It is possible to prevent some erroneous human behaviors through changes to the behavior of the device's automation and human-automation interface [32]- [35].…”