Présentateur : W.A. Sparrow
School of Health Sciences, Deakin University, Melbourne AUSTRALIA
Date : 21 octobre 99 ŕ 14h
Lieu : CENA salle de conférences
Contact : Sylvie Athenes
|
|
Sujet : Human error - A perspective from theory in
perception and action
Abstract :
It has been said that "to err is to be human", an observation that
must stand as one of the most fundamental facts of human existence.
Understanding error might be considered the ultimate challenge for
theories of perception and action because a good theory should be able
to explain the processes of both success and failure. When, however,
errors are not so severe as to be harmful they provide the opportunity
to learn. As Cutting (1982) wrote: "When we misperceive things and the
consequences tell us that we have erred, we are forced to reevaluate
and learn more about these objects of our attention (p. 211)". Being
error-free would preclude the opportunity to learn. It is interesting
to observe, for example, how a child's motor skills are refined through
errors that lead to painful consequences. In this essay an attempt is made
to define "an error", and to identify different types of errors. You may,
for example, select the wrong action but execute it perfectly such as
in driving turning right when you should have turned left. The action is
completed perfectly but it is the wrong one! Alternatively, we may have
errors in response execution. If a gust of wind pushes my beautifully
struck backhand drive off target an "unexpected event" has intervened to
stop an otherwise perfectly chosen and executed response from attaining
the task goal. Presumably, many of the events that we commonly describe
as "accidents" are a consequence of these types of error. Finally, what
we mean by the term "accident"? Presumably accidents arise from errors
that have negative consequences, with the major research focus being on
errors that result in physical injury or death.
In addition to developing the above framework on error, limitations
of human sensory and motor processes will be discussed in terms of
our evolutionary legacy. In response to these limitations we have done
two things to the modern technological environment: (i) designed it to
accommodate the limitations of the human visual and other sensory systems.
And (ii) devised activities to make extraordinary demands on the visual
and other sensory systems via "contrived" skills such as sports. The
significance of the latter observation is that engineers and managers
tend to de-skill work tasks to reduce errors. Stress can, however,
be caused by a lack of environmental stimuli that cause boredom. The
discussion concludes, therefore, by addressing the broad question of
human-machine system design and the role of the human operator in the
post-industrial workplace.
