Introducing User-Centered Systems Design - Foundations for Designing User-Centered Systems - page 4

Fig. 1.1 a Rank order the
quality of these switch to
light mappings. b Note how
long, on average, it will take
to push a button on each
panel. (Adapted from Payne
1995 )

(a) (b)

(c) (d)

error—that everyone is the same. We know the second assumption is not true from
simply observing that the world is made up of very different people with different
motivations, different backgrounds, and different skill sets.
To illustrate how our intuitions about people may be incorrect, and why it is
always worth testing your designed system with people who will use that system,
we offer the following examples.

1.2.1 Designing Mappings Between Buttons and Lights

It is generally claimed that the better designs are those that offer simple, clearer
mappings between an action and a response. However, the question is: what is a
clearer mapping? Consider Figs. 1.1 and 1.2 taken from a study by Payne ( 1995 )
on how well naive subjects could judge the quality of interface designs for a
simple system. Payne's experiment assessed what design people predicted would
rank best to worst on a very simple interface, where a number of different map-
pings between the controls and resulting system state were compared (what is
called ''stimulus-response compatibility'' in the scientific literature). In the fol-
lowing example, study participants were able to rank order the designs in Fig. 1.1
from best to worst. They were asked two questions: (1) what is the mapping of
lights to switches that gives the fastest response time? and (2) can you give a
prediction of how long they will take on average?
Sixty out of 70 subjects got the top ranked one correct. However, only four out of
70 got the complete order correct. The results of the study suggest that, when
confronted with anything but the most obvious choices, designers without
training may make poor design choices. Before going on, you may wish to try
this task yourself. The correct order is given in the exercises at the end of this
chapter.