Human-computer interaction as science

‘Human-computer interaction as science’ presents an interesting view of the longstanding debates surrounding HCI in relation to science and HCI as a science. Reading this paper as one of my first ventures into HCI without a computer science background felt like a daunting task at first but I feel like I’ve gained an oversight into the origins of HCI and some of the challenges the field faces as a growing discipline.

In my understanding, Reeves presents the idea that rather than focusing on theoretical debates, we should ‘stop worrying’ about how scientific HCI is or what the structure of the discipline is and simply apply rigour in our work and engage with the interdisciplinary nature of HCI. Before reaching this conclusion, Reeves focuses on the disciplinary shape of HCI and how this shape relates to science.

Reeves outlines two main problems for HCI which he would later reflect on throughout the paper: incoherence and inadequacy. The issue of incoherence centres on the lack of agreed upon laws, problems or approaches in the field of HCI and it could be suggested that the lack of rules means the discipline is, therefore, unruly. In addition, the issue of inadequacy focuses on the need to compare HCI to other scientific disciplines leading to the comparison of the ‘hard’ and ‘soft’ sciences. Reeves also crucially introduces the origins of HCI as way of background to explain how, traditionally, design problems were viewed as construction problems whereby engineering was employed to ‘fix’ them and essentially relied on trial and error engineering.

To address these key anxieties, there have been numerous attempts to define the scientific design space with the aim of leading to a hard and fast definition proving HCI as a science and, over the years, there seems to have been varying degrees of success in this area. Key to the story of HCI was the introduction of cognitive science as a way to ‘harden’ the ‘soft science of HCI’ through the use of representational theory of the mind to explain why computer interfaces failed or succeeded.

I felt there was much discussion throughout the paper about ‘science’ and ‘hard and soft science’, particularly the attempts over the years to compare HCI to the natural sciences which are seen to be the purest form. Having come from a ‘soft’ science background in Town Planning, I could relate to some of the anxieties, however, I could not understand why you would try to compare two incomparable disciplines to the detriment of one. I was relieved to see that a discussion around ‘science’ was included toward the end of the paper and clarified this approach to be unhelpful when defining a relatively new discipline.

I found Reeves’ conclusion to ‘stop worrying’ reassuring in that debates about theories and science will never stop. In the meantime we can carry on with our research in an interesting and relatively new field providing we can be rigorous in our processes and remember the interdisciplinary nature of HCI.

The paper I have chosen for this week can be found here.