If Human Beings are Rational Decision Makers, Why do so Many People Have Goats

1
If Human Beings are Rational Decision Makers, Why
do so Many People Have Goats?

• Investigating the Monty Hall Problem

2
How Do We Make Decisions?

• Assumption of Rationality People make the choice
  to maximize their gains when making a decision
• ECONOMIC MAN/WOMAN (Edwards, 1954)
• Decision makers are fully informed of all
  possible option and outcomes, are sensitive to
  the differences between options and are
  objectively rational
• SUBJECTIVE EXPECTED UTILITY
• Decision makers seek to achieve
  subjectively positive outcomes and avoid
  subjectively negative outcomes based on their
  subjective estimates of how likely it is they
  will achieve their goals. People are
  subjectively rational.

3

• Bounded Rationality Humans are not irrational,
  but are rational within limits (Simon, 1957)
• SATISFIZING (Simon, 1957)
• We consider the minimum number of options that
  would allow us to minimum criteria (i.e. we only
  look until we find what were looking for)
• 2) HEURISTICS AND BIASES (Tversky Kahnemann,
  1979)
• We make decisions based on rules of thumb of and
  personal preferences often based on preconceived
  notions (e.g. We pick Molly to play on our
  basketball team because shes tall and tall
  people are good at basketball)

4
Decision Making and Probability

• Recent theories suggest that human beings are NOT
  always rational decision makers
• People are particularly irrational decision
  makers when evaluating a combination of possible
  events

5
Lets Make a Deal In class demonstration

• Underneath one of the cups is a prize, there is
  nothing under the other two cups
• Pick a cup
• Once youve picked, I will flip over one of the
  remaining cups that has nothing under it
• You can then change your mind (SWITCH) or STAY

6
Explanation

• You originally have a 1/3 chance (33) of
  selecting the right cup/door
• You are always better over switching in the long
  run because your new pick offers you a 2/3 chance
  (66) of winning

7
That Doesnt Make Sense

• Explanation 1
• Your original pick gives you a 1/3 shot of
  winning
• This means that its more likely that the prize
  is behind the other doors (2/3)
• When Monty tells you which door does not have the
  prize, the probability is still 2/3 that one of
  them hides the Grand Prize, but now you know
  which of the two it is
• Explanation 2
• Imagine you have 1000 doors and behind one of
  them is a prize. When you pick a door, your odds
  are 1/1000 that youre correct
• Monty then opens 998 / 999 doors
• One door and your door remain Do you want to
  switch now?

8
CogLab Experiment Monty Hall

• Hypothesis Always staying will lead to winning
  33 of the time and always switching will lead to
  winning 66 of the time
• Method
• Participants Fifteen undergraduate students at
  UBC enrolled in PSYC 309 901 completed the
  self-administered experiment. Only the first ten
  were selected (N10)
• Materials Personal computer and online
  demonstration from CogLab website
  (http//coglab.wadsworth.com)

9

• Procedure
• Ps logged on the CogLab website and, once signed
  it, clicked on the lab entitled Monty Hall where
  instructions were provided
• Three doors appeared on the screen and the P
  selected a door by clicking on it
• After a door was selected a - or ? appeared above
  the other two doors (- indicated no prize and ?
  indicated there might be a prize)
• Ps could then choose to STAY with their original
  door or SWITCH to the other door. There were 29
  trials/subject.

10
Results

• Typically, if you always STAY with your original
  pick you will win 33 of the time (pop. mean
  .33)
• In our study m.30, s.0541
• t(72)-0.53, plt0.5
• Typically, if you always SWITCH you will win 66
  of the time (pop. mean .66)
• In our study m.67, s.0292
• t(259)0.31, plt0.5

11
Discussion

• The results from our experiment support our
  hypothesis Always staying lead to winning 33 of
  the time and always switching lead to winning 66
  of the time
• Our sample is statistically identical in
  performance to the population
• However, some subjects still chose to stay on
  every trial or stayed on at least some of the
  trials

12
Why do People Have Such a Hard Time with the
Monty Hall Problem?

• Most people think the odds should be 50/50, that
  it shouldnt matter if they switch or stay (Burns
  Wieth)
• Not due to lack of knowledge (Burns Wieth)
• Illusion of control (Dorr Granberg, 1998
  Langer, 1975)

13
References