Rational analysis of the selection task

1
Rational analysis of the selection task

• Oaksford and Chater (1994)
• Presented by
• Bryan C. Russell

2
Wason selection task

• Rule if there is an A on one side, then there is
  a 2 on the other side

K
2
7
A
3
Another view of the task

• Let rule be if p then q
• Four types of cards
• (p,q), (not-p,q), (p,not-q), (not-p,not-q)
• Two hypotheses
• MD p,q are dependent (rule is true)
• MI p,q are independent (rule is false)

4
Assumptions

• We can assign probabilities to the cards
• Should reflect natural statistics of if p then
  q statements in nature
• P(p MD) P(p MI) a
• P(q not-p,MD) P(q not-p,MI) b

5
Card probabilities
6
Card probabilities

• Task Select card that maximally reduces
  hypothesis uncertainty

7
Entropy/uncertainty
8
Entropy/uncertainty
9
Entropy/uncertainty
10
Another experiment

• Suppose you observe
• TTHTHHTHHHHTHHTHTHHT

11
Another experiment

• Suppose you observe
• TTHTHHTHHHHTHHTHTHHT
• HTHHTTHTHHTTTTTTTTTH

12
Another experiment

• Suppose you observe
• TTHTHHTHHHHTHHTHTHHT
• HTHHTTHTHHTTTTTTTTTH

13
Mutual information
14
Application to selection task
a Pr(p)
15
Application to selection task
a Pr(p)
16
Model behavior
17
Observations

• If Pr(q) is low, then choosing p card is
  informative
• If Pr(p) and Pr(q) is low, then choosing q card
  is informative
• If Pr(p) is high, then choosing not-q card is
  informative
• not-p card is not informative (results in zero
  information)
• P(MI) only scales information values

18
Model behavior
R
19
Rarity assumption

• For selection task, in humans Pr(p) and Pr(q) are
  low
• Expected information over region R
• choose p 0.76
• choose q 0.20
• choose not-q 0.09
• choose not-p 0

20
How do humans compare?
21
(No Transcript)
22
(No Transcript)
23
Analysis

• Both humans and model accounts for the following
  information relationship
• choose p gt choose q gt choose not-q gt choose not-p

24
Thematic selection task

• If a person is drinking beer, then they must be
  over 20 years old

drinking sprite
25 years old
16 years old
drinking beer
25
Thematic selection task

• If a person is over 20 years old, then they may
  drink a beer

drinking sprite
25 years old
16 years old
drinking beer
26
Rule types

• Obligations if action (p), then must condition
  (q)
• Permissions if condition (p), then may action (q)

27
Subject perspective for permission rule

• Enforcer
• Pretend you are a bouncer at the RD pub
• Actor/inquirer
• A guy walks into a bar

28
Utility-based model

• Focus on rule-use, not rule-testing
• Associate cost with turning over a card

29
Utility-based model
30
Performance of utility-based model
31
Comparison with humans
32
Andersons (1990) steps for rationality

• Specify precisely the goals of the cognitive
  system
• Develop a formal model of the environment to
  which the system is adapted
• Make minimal assumptions about computational
  limitations

33
Andersons (1990) steps for rationality

• Derive the optimal behavior function given the
  previous steps
• Examine the empirical evidence to see whether the
  predictions of the behavioral function are
  confirmed
• Rinse, lather, repeat, and refine the theory

34
Discussion questions

• Why use a probabilistic framework for
  rationality?
• Is the rarity assumption valid?
• Is the selection task representative of
  accounting for rational thought? Is it
  exhaustive?
• How does one learn the utility costs?
• How does one learn Pr(p) and Pr(q)?