Who was Jens Rasmussen
Jens Rasmussen (1926–2018) was a Danish engineer and researcher who studied how people understand and control complex technical systems. He developed tools for analysing decisions, representing constraints and examining the interaction between work organisation and technology. [1]
Education and career at Risø
Born in Ribe, he qualified in electronic engineering in Copenhagen in 1950. He joined Risø in 1956, headed its Electronics Department from 1962 to 1987 and served as research professor of cognitive engineering from 1987 to 1992. This institutional career provided the setting for his investigations of operators, information and control. [3]
With A. Jensen, he published a 1974 study of technicians repairing electronic equipment in their normal workplace. Verbal protocols revealed search routines that could save time and mental effort despite repeating information. They also observed fixation on unsuccessful strategies, helping explain technical diagnosis and the role of professional experience. [4]
Skills, rules and knowledge
In 1983 he presented his influential distinction between skill-, rule- and knowledge-based performance. Skills support automated sequences; rules guide familiar responses; knowledge enables reasoning about unfamiliar situations. The paper connected these modes with signals, signs and symbols, providing a vocabulary for examining the information people need during an activity. [5]
That year, Design for Error Tolerance focused on two system properties: making the effects of an action observable and allowing them to be reversed before harm occurs. Rasmussen linked this proposal to adaptation and learning, processes through which people try strategies and adjust their performance. [6]
The abstraction hierarchy
His 1985 paper on hierarchical knowledge representation described systems through connected levels of abstraction. A functional representation links equipment with what it does and the purposes it serves. This organisation of knowledge mattered both for database and display design and for understanding how people reason during equipment failures. [7]
Published in 1986, Information Processing and Human-Machine Interaction brought together his approach to cognitive engineering. It addresses task analysis, diagnosis, search strategies and mental representations. The book connects a system’s physical operation with the decisions required to understand it, providing a broader account than his papers on individual components of the framework. [8]
Modelling tasks and responses
Another 1986 work organised cognitive analysis into five dimensions: problem domain, decision task, organisation of decision makers, mental strategies, and cognitive mechanisms or preferences. He explored this framework across process plants, emergency management, offices and libraries. The proposal placed user-task interaction before decisions about the form of the interface. [9]
Also published in 1986, Simulation of Operators’ Response in Emergencies proposed a simulation approach for process plants. It combined Brunswik’s lens model, a repertoire of normal tasks and an error taxonomy. Its scope was restricted to skill- and rule-based responses during the early phases of an accident. [10]
Error, adaptation and causality
The Role of Error in Organizing Behaviour appeared as a Risø report in 1989. Rasmussen argued that causal analysis differs according to whether its purpose is to explain an event, assign responsibility or improve a system. Where the search for causes stops determines which decisions and relationships enter the explanation. [11]
He developed this argument in Philosophical Transactions of the Royal Society in 1990, examining the ambiguity of selecting particular events as causes and the relationship between error and learning. The paper questioned the separation of erroneous behaviour into a distinct category in modern environments, where people continually adapt their practices to changing conditions. [12]
Ecological design and cognitive systems engineering
With Kim J. Vicente, he published the foundations of ecological interface design, EID, in 1992. It aimed to support all three cognitive modes without demanding more reasoning than a task required. The paper combined design principles, theoretical discussion and early evaluation results, presented as preliminary support for the framework. [13]
A 1999 paper extended the discussion to reliable systems, including technical aircraft subsystems. He advocated integrating users into functional design and making system constraints visible. The proposal connected information presented on displays with the boundaries within which the activity could take place. [14]
His collaboration with Annelise M. Pejtersen and L. P. Goodstein produced Cognitive Systems Engineering, published by Wiley in 1994. Its shared authorship reflects an interdisciplinary project bringing technical systems analysis together with the study of activities and information needs. [15]
Pejtersen and colleagues later described how Risø’s programme integrated research on fiction retrieval in libraries with studies of power plants and hospitals. Comparing domains with different degrees of coupling contributed to cognitive work analysis and to tools for designing and evaluating information systems. [16]
Dynamic risk management and AcciMaps
With Inge Svedung, he published Proactive Risk Management in a Dynamic Society for the Swedish Rescue Services Agency in 2000. The book proposed adaptive control based on observing actual safety conditions and communicating objectives across decision-making levels. Feedback was central to managing environments undergoing technological and organisational change. [17]
Svedung and Rasmussen published Graphic representation of accident scenarios in Safety Science in 2002. It addressed graphical representation of system structure and accident causation, contributing to their investigation of how relationships between decisions and events could be visualised beyond a chronological list of occurrences. [18]
Later collaborations and legacy
With Richard Cook, he published Going solid in 2005, examining hospital dynamics. They described how reduced buffers can increase dependencies between units, allowing local difficulties to affect other services. This was a conceptual model of patient safety and organisational saturation, rather than an experimental demonstration of a healthcare intervention. [20]
He joined the US National Academy of Engineering in 2013. An ODAM symposium in 2014 brought researchers together to examine his legacy, including operating boundaries, cognitive work analysis and AcciMaps. It recognised the continuing influence of work spanning engineering, psychology and organisational research. [19]
Connection with occupational safety
His preventive contribution broadens investigation beyond the last visible action. It encourages questions about how objectives, resources, information and successive decisions shape the conditions workers encounter.
