Implementation bias in HCI – and escaping it

Abstract

HCI, whether for research or for specific usability case studies, tries to align the user task with the user interface design, with its computer implementation. For research, HCI tries to identify the principles; for specific case studies, HCI tries to improve the user experience or the product delivered. In all approaches, the user task is central, yet the user task may be an artifact of previous implementations (it may suffer implementation bias), and users — and user studies — may therefore accidentally focus on problems with their current task as implemented rather than opportunities for a new HCI approach. In this paper, we use presentations (i.e., giving talks or lectures) as a detailed stand-alone case study, and we show that implementation bias is a serious impediment to thinking clearly about improving the task of delivering quality presentations. We then argue, using a very different example (healthcare IT) that implementation bias is, in fact, a common and serious problem across HCI generally.