Copyright 1998. David Gilmore, Elizabeth Churchill, & Frank Ritter

Lecture 3

(part 2 of 4)

A closer look

Through considering anthropometrics (users defined by what size they are, what muscle strength they have, what visual acuity they will have, etc.), behavioural issues (users defined by what they will do) and cognitive (users defined by what they will know), and social factors (users defined by where they are- context, and their relationships to other people).

Anthropometric approach (Can it be used?)

Concerned with:

e.g. BALANS chair for secretaries, etc.

Optimum kitchen work surface/heights, etc.

(Ergonomics, 1973 - ergonomics of chapati making - requires a different design of kitchen)

Ergonomics of armies marching in heavy boots versus trainers.

US AIR Force ejector seats in WWII took average male statistics and built ejector seats which removed knee caps of 5-10% of pilots.

Car speedometers - questions of legibility of characters, avoidance of glare in bright sunlight, avoiding parallax problems with different heights of drivers.


Behavioural approach (How is it used?)

Problem of asserting how something is used, to see if can be made quicker/safer/more productive, and so on.

Looking at mistakes that are made, to see how they could be prevented.

Windscale fire, where important temperature gauges were at ceiling height, with ladders to be climbed in order to read them.

Making knobs and levers tactually discriminable, to enable them to be used without looking to check whether right knob being used (e.g. RAF Cottesmore accident)

e.g. design of car speedometers, such that the numbers on the dial are not obscured by the rotating needle.

e.g. dials at opposite ends of room

e.g. Three Mile Island

e.g. Car boot

Typical of this style of ergonomics are checklists of those tasks best performed by man and those best performed by machine. For example, Murrell (1965) lists the following:

Norman's forcing functions: affordances: artifacts suggest certain usage. Note that what is afforded interacts with the task at hand and the way in which the user perceives the task.


Cognitive Approach (How do they think they are using it?)

Understanding how someone decides where to look for information.

Understanding how user decides what action to take.

Recent development, since systems used not to be so complicated that user had much choice, or had to react quickly.

Car speedometer - need to consider how we best perceive relevant information - which is approximate speed, assessed rapidly without loss of attention to driving task. Dial is best.

What information do users need to plan a strategy for performing some task. Is there so much noise that they can't process information, e.g. interruptions in Kegworth crash?

What strategies are available to the user when the system goes wrong?

How can we ensure that they do not lose their ability to perform the task manually as a result of automation?

e.g. the design of displays that are not only legible but which show the sort of information that is required. Not all digital displays are used to show the actual value of whatever, e.g. change in temperature may be more important than actual temperature.

How can word processors be made more accessibly to secretaries and others? How can we ensure that computerised office systems do not bring down the business?

The subtitles of these three approaches (Can, How is, How perceived) suggest an ordering. The first is often tackled first, then the second, and finally cognitive issues. This, of course need not be the case.

Involves observation of the tool/environment in use, asking the question Why/When is it used?

Involves understanding how someone decides

Requires knowing or understanding of

Imagine if you had two dials to read, each with a different number and these numbers varied randomly. Your task is to press a button IF the difference between the two number exceeds 10. This task would require you to process the information from the two dials, make a mental calculation and evaluate whether it is over 10 or not. The existence of a third dial with the smaller number subtracted from the larger would make your task easier and quicker removing the need for the mental calculation (the cognitive effort).

Better designs can come from understanding the mental effort involved in tasks in terms of the information processing architecture that supports our thinking, that is, constraints such as how many arbitrary symbols we can remember, etc. These issues may help us to understand how complex devices with high functionality (e.g. videos, etc.) can be made more accessible to non-expert users.

Cognitive ergonomics: concerned not necessarily with looking for completely correct or accurate psychological theories of interaction, but with applicable models that are of use to designers, instructors and users. It is also concerned with the interaction of many cognitive processes: planning, language, problem-solving, learning, memory, perception, etc. In this respect it is different to the direct application of cognitive psychology, in that it does not look at cognitive processes in isolation, but their integration and how they are involved in particular activities or situations. Cognitive ergonomics also differs from cognitive psychology in focusing on theories which can predict behaviour in the real world, which may require a more detailed treatment of, for example, individual differences, than many other branches of psychology. Will consider this more when dealing with basic issues from psychology and their impact on design: Norman, Green and cognitive dimensions, etc.

Social issues (How about others when using it?)

Some people really focus on this: for example in the Monitor of the American Psychological Association recently there was a report recently to suggest that many aircraft accidents are caused by the social dynamics of the situations in which users find themselves. As above.......

"Research shows that errors have their roots in the backgrounds of participants, the dynamics of the group and the environment in which the activity occurs. Failure in inter-personal communications characterises can also cause accidents:

Lecture 3 Continued...

Return to Contents...