Rasmussen
(1983)
Copyright Notice: This material was
written and published in Wales by Derek J. Smith (Chartered Engineer). It forms
part of a multifile e-learning resource, and subject only to acknowledging
Derek J. Smith's rights under international copyright law to be identified as
author may be freely downloaded and printed off in single complete copies
solely for the purposes of private study and/or review. Commercial exploitation
rights are reserved. The remote hyperlinks have been selected for the academic
appropriacy of their contents; they were free of offensive and litigious
content when selected, and will be periodically checked to have remained so. Copyright © 2018, Derek J. Smith.
First published online 11:21 GMT 11th March 2002,
Copyright Derek J. Smith (Chartered Engineer). This
version [2.1 - link to graphic] dated 09:00 BST 2nd July 2018
Rasmussen's (1983)
"Typical Levels of Performance" Omega
The Danish engineer Jens Rasmussen
produced the following diagram in an attempt to apply better cognitive theory
to the design of man-machine interface systems and thus help reduce the
potential for accidents. He began by characterising
human performance in a familiar environment as goal-oriented, and
rule-controlled, and he then proposed three qualitatively different levels of
cognition in which qualitatively different types of information circulate and
qualitatively different types of decision are made. He referred to these as
"typical levels of performance" (p258). This analysis was
subsequently adopted without fundamental alteration in James Reason's (1990)
book "Human Error". Here is Rasmussen's diagram and a summary of his
proposals:
Rasmussen's (1983) Three Levels of Skilled
Performance: This is
Rasmussen's (1983) attempt to distinguish between the three levels of the
standard five-box omega, (a) by the nature of the behaviour controlled by
each hierarchical processing layer, and (b) by qualitative differences in the
nature of the control information itself. Here are the three levels he
proposes: Rasmussen describes the simplest form of behaviour
as skill-based behaviour (SBB). It is controlled from the lowest level
of the cognitive processing hierarchy, and may be characterised as
"smooth, automated, and highly integrated" (p258) and takes place
(critically) "without conscious attention or control"
(p259). Effective SBB performance relies on heavy feedforward control flows
throughout, "depends upon a very flexible and efficient dynamic internal
world model" (p259), and will usually involve rapid coordinated
movements. Examples of SBBs are bicycle riding or musical performance. As for
the nature of the information at this level, SBB is described as relying on signals,
which are defined as "representing time-space variables from a dynamical
spatial configuration in the environment" (p261). ASIDE: If unfamiliar with the difference between
"feedforward" and "feedback" information in a control
system, click here. The next level of complexity is rule-based
behaviour (RBB). It is controlled by the middle level of the processing
hierarchy, and may be characterised as consisting of "a sequence of
subroutines in a familiar work situation" (p259), where the subroutines
follow previously stored rules, again relying primarily on feedforward
control. Examples of RBBs are mathematical problem solving and system control
tasks such as the discrete manoeuvring of aircraft or cars. As for the nature
of the information at this level, RBB is described as relying on signs
to indicate the state of the environment. These are defined as "related
to certain features in the environment and the connected conditions for
action" (p261). ASIDE: Rasmussen explicitly warns that "the
boundaries between skill-based and rule-based performance is not quite
distinct" (p259), varying with both level of training and attentional
state. The acid test is this: rule-based control is ultimately based upon
"explicit know how" - the rules can be explained in words by the
person concerned; so if you cannot explain it then it must be skill-based. The highest level of complexity is knowledge-based
behaviour (KBB). It is controlled by the highest level of the processing
hierarchy, relies upon a "mental model" of the system in
question, and in general terms is to be strongly avoided
because what it achieves in terms of sophistication it loses in the time it
takes [for an example of what happens when all training fails and things have
to be consciously diagnosed and responded to, see the story of the Staines
air crash, 1972]. KBB is therefore
what you have to turn to only when SBB or RBB are momentarily not up to the
task at hand. This means that the goal of a given piece of KBB has to be
"explicitly formulated" at the time it is needed, taking into
account the nature of the problem and the overall aims of the subject.
Examples of KBBs are problem solving and fault diagnosis. As for the
nature of the information at this level, KBB is described as relying on symbols.
These are defined as "abstract constructs related to and defined by a
formal structure of relations and processes" (p261), and include
language itself and mathematical equations. If this diagram fails to load
automatically, it may be accessed separately at http://www.smithsrisca.co.uk/PICrasmussen1983.gif |
Redrawn from a black and white original in Rasmussen (1983; Figure 1). This version Copyright © 2002, Derek J. Smith. |
Slips, Lapses, and
Mistakes
If interested in the error types
associated with each of these three processing levels, see Section 2 of Unit HE2.
References:
Rasmussen, J.
(1983). Skills, rules, and knowledge: Signals, signs, and symbols, and other
distinctions in human performance models. IEEE Transactions on Systems, Man,
and Cybernetics, SMC-13(3): 257-266.
Reason,
J. (1990). Human Error.
Recommended "Human Error" Reason (1990) To see an abstract, or to order this book, click here |
|