Usability & IT Systems
There is often need for improvement of the usability of IT-systems in large organisations. We develop such improvements based on the theories from the research field human-computer interaction. We also study human behaviour in complex dynamic systems, such as systems for intensive care units, traffic control, and industrial processes, using control theory as a suitable language.
We study human behaviour in complex, dynamic systems, addressing research questions on different levels and topics, and with the help of methods we develop in close connection to the tasks we study and the stakeholders we cooperate with. To us, an important scientific principle is that research shall focus where humans focus their interest, in general and in particular. Our research is therefore empirical rather than based on theoretical models. Examples of application domains range from vehicles and vessels, traffic control centres, intensive care units to surveillance of industrial processes and power plants. Our point of departure is often situations and contexts where highly skilled workers use their expertise to manage different work tasks. But rather than doing traditional task analyses, we perform work domain analyses in order to reach a more thorough understanding of how different work tasks can be performed. The work tasks we study are often critical components in reaching general objectives, such as efficiency, safety, productivity and reduced cognitive work load.
Control theory provides us with a generic language for understanding human actions in complex, dynamic systems. Its usefulness stems from the fact that it specifies the four general requirements for control of any system, regardless of whether control is exercised manually or by automation. These four control conditions are: there must be a goal; there has to be a model of the system; it must be possible to ascertain the state of the system (observability); and finally, it must be possible to affect the system (control functions). Goals and models are properties on the behalf of humans or organizations, whereas observability and control functions are characteristics of technical systems. We have formulated a general research hypothesis that guides much of our research: Humans develop mental models and formulate goals and sub-goals as a function of the observability and the action possibilities that the systems provide.
One research topic is information acquisition methods. Experts are often in possession of knowledge that we need to find a way to elicit. Without this information, we don´t know how to support them when designing new interfaces or process dialogues. Here, we have developed our own method - collegial verbalisation. It makes it possible for us to perform knowledge elicitation, resulting in data for which there is inter-subjective agreement that they correspond to the observed behaviour.
A second topic is human factors. Here we study automation issues specifically. Automation is often necessary, even in socio-technical systems. The design of the automats affects the operators´ action possibilities, however, and therefore the design needs to be evaluated thoroughly before put into practice. Here, we have developed an important design principle, that is, to design for management by awareness, rather than management by exception, which is too common in systems engineering. One intriguing topic is to design for unusual and unexpected situations. In these situations, the demand for control is irrevocable and the demand for situation awareness is unconditional.
Human Decision Making
A third research topic is human decision making. Our approach here is that judgements and decisions are contingent. Shortly, environmental constraints are imposed on the individual, affecting the deployment of strategies to the extent that the outcome of the decision is a joint effect of individual and environmental factors. The methods we use are field studies, where we specifically study the cognitive strategies and the mental models that operators use and base their decision on.
Usability is the practical application of theories from the interdisciplinary research field of human-computer interaction (HCI). Both HCI and usability have influences from computer science, engineering, psychology, as well as, sociology and pedagogics. Usability as a research discipline is often divided into sub-disciplines, such as analysis, design, and evaluation. But it is also about the process of developing usable systems.
In our research, we often focus on the process of developing and enhancing the usability of current systems. Often there is a need for improvement. These demands may be consequences of ill-functioning systems that do not deliver what organizations need in terms of efficient and effective IT-systems, or they may be associated with high cognitive work load leading to personnel not being able to perform their work tasks as they would like to. But the demands may also be consequences of new technical solutions that promise higher efficiency and productivity, often associated with loosely defined new functionality.
User-Centred Systems Design
User-centred system design is both a methodology and a philosophy. As a methodology, it is used as a way to capture the design requirements, as a way to check preliminary design suggestions and as a way to evaluate more ready design solutions. As a philosophy, it is concerned with the division of labour and responsibility, often connected to over-arching objectives of the IT-system employed.
A specific methodology developed at our department is concerned with the deployment of IT-systems in large organizations and the ability to introduce and continuously use usability principles as part of the design and construction of the IT-systems in the organization. One method we use is vision seminars. Here, different levels in an organization iteratively find out about purposes and strategies in the organization, and how the employment of IT-systems contributes to these main objectives. Another method is participatory design. Design suggestions in terms of prototypes are developed and used as a way, not only to evaluate the interface prototypes, but also as a way to help the users realize the potential in new technical solutions and platforms. In these situations, the researcher becomes more of catalyst. When we as researchers adopt such a role, we not only develop technology as tools for a group of users or an organization. More important, this is an example of how we apply action research within our department.
Finally, we also conduct research within the discipline of assistive and facilitating technology, including inclusive design. Usability is for all categories of users, and the needs and requirements of disabled and elderly are necessary to be dealt with specifically. One leading principle here is that design should be non-stigmatizing and that technology should enhance and augment the possibility of design for all.
Anders Arweström Jansson
Professor emeritus in Human-Computer Interaction
Professor in Human-Computer Interaction.