Introducing User-Centered Systems Design - Foundations for Designing User-Centered Systems - page 7

USA have both tried to introduce new coins (e.g., the Susan B. Anthony dollar) with
little success, partly due to the lack of care in the design of the new coin. In contrast,
when Britain got a new currency in 1971, switching to a system where one UK pound
was equal to 100 pennies, the introduction of the new coinage was a resounding
success. It turned out that one reason for the success was a substantial body of
research on how people perceived the value of coins (e.g., Bruce et al. 1983 ) as well
as attention to how the different proposed coins might be made least confusing to the
elderly or sight impaired. During the research it was recognized that many people
need to identify coins from touch alone (e.g., the coin in your pocket) and that
designing for the blind user actually meant designing for everyone. The cost of this
research was a very small component of the costs of introducing a new coinage
system (e.g., all of the new vending machines to be developed), but it helped ensure
the success of the whole enterprise. Subsequent changes to the coins have also
followed these guidelines, with the two pound coin, for example, being the same
basic shape as the one pound coin, but larger and heavier.
In these examples we see one of the first universals of human behavior—people
remember those details that they pay attention to but only in sufficient detail for the
tasks they are performing. This is universal, but it does not enable us to predict fully
what details someone will remember, because there are differences in how much
attention people have to spare, what tasks they are performing, and thus what details
they will remember. The first two problems in Figs. 1.1 and 1.2 are difficult because
the differences in performance of the tasks are not particularly available to con-
sciousness, and most people's representation of how they think they perform these
tasks in this area do not reflect how people actually perform the task. The penny
question (and the menu question) represent the difference between recognition and
recall memory. Usually identifying a penny just requires being able to discriminate
between it and other coins. With the set of alternatives provided by Nickerson and
Adams, the choice has to be based on recalling the features of a penny, which most
people have never bothered to commit to memory (why would they?).
Another classic example is remembering your new cell phone number. It takes a
long time to learn it because you, personally, never need to use it (unless you
misplace your phone, in which case calling it is a good strategy for finding it!).
However, if someone asks you for it, you either have to recall it or have to go
through the menus on your phone to find it, eventually recognizing the steps that
enabled you to find the number.
This disconnection between how we think we behave and how users really
behave is common and there are plenty of reasons for it. In most cases we are too
busy doing a task to properly observe how we are doing it. When we can observe
how we are doing it, it is rare that we can correctly and completely infer why we
are doing the task—the observation of behavior is separate from the generation of
it. Ericsson and Simon ( 1993 ) provide a full explanation of why it is hard for
people to examine their own thinking processes. Their explanation includes that
when we recognize how we behave, we rarely make written notes and thus any