Why option generation matters for the design of autonomous e-coaching systems

• DOI: 10.1007/s00146-013-0532-5
• Publication Date: 2014-01-16T11:10:32Z
• Authors: Bart Kamphorst; Annemarie Kalis
• Keywords: Human-Computer Interaction; Artificial Intelligence; 0202 electrical engineering, electronic engineering, information engineering; Scientific; 02 engineering and technology

Autonomous e-coaching systems offer their users suggestions for action, thereby affecting the user's decision-making process. More specifically, the suggestions that these systems make influence the options for action that people actually consider. Surprisingly though, options and the corresponding process of option generation--a decision-making stage preceding intention formation and action selection--have received very little attention in the various disciplines studying decision making. We argue that this neglect is unjustified and that it is important, particularly for designers of autonomous e-coaching systems, to understand how human option generation works. The aims of this paper are threefold. The first aim is to generate awareness with designers of autonomous e-coaching systems that these systems do in fact influence their users' options. The second is to show that understanding the interplay between a person's options and the e-coaching system's suggestions is important for improving the effectiveness of the system. The third is that the very same interplay is also crucial for designing e-coaching systems that respect people's autonomy.