Cognition: Mental Representations, Problem Solving, and Decision Making - Foundations for Designing User-Centered Systems - page 186

Table 6.4 Example problems illustrating the framing problem
Assume that you are preparing a new manufacturing process that will cost $600,000 to create.
Two alternatives have been created to help save money implementing this system. Assume
that the exact scientific estimates of the alternatives are as follows
Program A: $200,000 will be saved
Program B: there is a 1/3 chance to save $600,000 will be saved, and a 2/3 chance that no
money will be saved
Imagine that the Programs A and B are not available, but only C and D are
Program C: $400,000 will still have to be spent
Program D: There is a one in three chance that no money will have to be spent, and a two in
three chance that the original estimate will have to be spent
astounding answers. The actual population sizes differ by a factor of at least two in
all of these pairs.
6.4.4.5 Framing Effects
The way that outcomes are presented (framed) has a powerful influence on how
users choose between alternatives. The framing effect (Tverksy and Kahneman
1981 , 1986 ) notes that decision makers are quite sensitive to the decision's frame
of reference. Consider the choices shown in Table 6.4 . Which would you choose?
The most common answers are given in Exercise 6.2.
Where the outcomes are noted in positive terms, lives are saved, money gained,
and so on—people making decisions are risk aversive. They appear to act as if they
wish to lock in their savings, choosing an outcome with certainty.
Where the outcomes are noted in negative terms, lives are lost, money lost, and
so on—people making decisions are risk seeking. They make decisions based on
trying to avoid the loss.
The choices in Table 6.4 have the same expected statistical values, but how you
and your users will choose solutions will depend in some part on how the infor-
mation is presented.
For example, in an experiment, Kahneman gave students mugs in class. When
they were asked to sell them back, they wanted $7.12 for them. When they just got
to look at the mugs, and could either collect a mug or get paid, they valued the
mug at $3.50. This process had the same outcomes, mug or money, but the people
holding the mug valued it much higher than the people who could get the mug,
showing that how information is presented and the order in which it is presented
can influence cognition.
This effect of order on having and valuing also plays out for users and software
and software features. Losing a feature is seen by users as much more costly than
not adding a feature. When a subscription to an online service is dropped it is
likely to be valued higher than when adding it. This is useful to know for