ABOUT COGNITIVE SYSTEMS ENGINEERING

by Erik Hollnagel

The domain of CSE comprises natural and artificial cognitive systems and how they perform in their physical and social environments. The study involves the analysis, modelling, design and evaluation of cognitive systems (CS) and joint cognitive systems (JCS).

CSE emphasises a functional rather than a structural point of view, which means that the performance of the joint cognitive system (human-machine ensemble) as a whole is more important than the interaction between the parts. CSE does not view performance as an emergent phenomenon that can be explained solely in terms of the characteristics of system components, but rather as something with properties that requires the study of the CS/JCS as a whole. The difference between CSE and, e.g., HMI and HCI, is thus in the approach rather than in the objects of study.

What Is A Cognitive System?

In plain terms, a CS is able to perform in an orderly manner. This specifically means that a cognitive system – and therefore also a joint cognitive system (JCS) – is able to control what it does.

The set of cognitive systems conforming to the definition includes humans, higher order living organisms, organisations, and certain kinds of machines or artefacts. It makes pragmatically sense to describe cognitive systems as if they have a purpose or goal, and it is essential to understand the role of the purpose / goal in how performance is organised and controlled.

What Is A Joint Cognitive System?

A joint cognitive system is defined as two or more systems considered together, where at least one of them is itself a cognitive system. (An organisation can therefore be considered as either a cognitive system or a joint cognitive system.) An important set of JCSs is human-machine ensembles. These constitute a key field of research, partly overlapping with HMI and HCI. In practice, the purpose of a JCS is often identical to the purpose of the human part of the system, although larger entities – such as organisations – may have purposes independently of the humans that are part of them. A JCS is defined by what it does (function) rather than by what it is (structure).

Background of CSE

CSE was formulated to provide a consistent conceptual and methodological basis for research on human-machine ensembles, with design and evaluation as the two major activities. The domination of the information processing paradigm emphasised the physical separateness of humans and machines and turned the focus inward to "cognition in the mind". Yet rather than being isolated in the mind of a thoughtful individual, cognition at work typically involves several people and artefacts distributed in space or time, with control, co-operation, and co-ordination control as primary characteristics. In order to cope with this complexity, it is necessary to have concepts that can be applied to systems at many levels of decomposition/aggregation. This led to the notion of a cognitive system as defined above, and to the set of methods that gradually have matured since.

Practically all activities in the industrialised societies involve joint cognitive systems. The field of CSE is therefore cross-disciplinary and draws on contributions from cybernetics, behavioural sciences, computer science, and systems engineering – with close links to cognitive science and human factors / cognitive ergonomics. A special concern is the way in which emerging technologies affect system functions and work conditions. An important aim is therefore to develop techniques and methods that can enhance the safety and efficiency of real-world applications such as industrial process control, commerce, manufacturing, transportation and communication systems, contingency management, etc.

Subject Matter

In practice, CSE has mostly studied JCSs that are human-machine ensembles, and which have one or more of the following characteristics: