Lecture 1 - Heuristics and Biases Flashcards

You may prefer our related Brainscape-certified flashcards:
1
Q

How do we make a rational decision?

A
  1. Define the problem
  2. Identify all decision criteria
  3. Allocate weights to criteria
  4. Identify all alternatives
  5. Evaluate the alternatives
  6. Choose the best alternative
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are concepts (beyond rational choice) that people use to make decisions?

A

1) Bounded rationality (Simon) - limited resources (time and cognitive resources) - we do not choose the best options for us because of not having enough resources (we ‘sacrifice’) - satisfactory decision among few alternatives

2) ‘‘Predictable’’ irrationality - systematically wrong decisions (non-optimal) - (Tversky and Kahneman)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is the more important model to understand how people make decisions?

A

System I and System II - Bat and ball example

System I - ‘‘Intuitive answer’’ - fast, automatic, not cognitively demanding

System II- ‘‘rational answer’’ - slow, controlled, cognitively demanding

Idea: System II questions the validity of System I choices and comes up with a rational answer.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What are systems I and system II responsible for?

A

System I - responsible for heuristics (mental shortcuts to satisfactory(tend to be correct) solutions) - efficient, work most of the time;

However, when system I does not work, it leads to biases - systematic deviations from ‘rationality’ , these biases are systematic and predictable;

NOTE: not all biases and heuristics are due to System I (they can occur in System II as well) ;

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Where can we apply probability estimation/judgment?

A

Buying an apartment (price changes in years), starting to feel ill and go on a vacation, when we buy products, visit new places etc.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What concepts do we often use when we make probability estimations?

A

Representativeness Heuristic - mental shortcuts to make judgements in an efficient way ; when we estimate that x belongs to y => the more x resembles y, the more likely is that x to be y;

Availability Heuristic - the instant recalling of certain information influences our choice

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What are the issues with the representativeness heuristic?

A
  1. How confident are we in the resemblance (of x to be y)?
    A) Stereotypes (female CEO - parent role - calling to wish ‘‘Good luck, son’’ while the father is in the car - accurate stereotypes are still stereotypes, but they can still be misleading.
    B) Insensitive to sample size - small sample = less likely to represent reality; bigger sample = more likely to represent reality
  2. How likely is Y in the first place?
    C) Conjunction fallacy - Likelihood of x being y also depends on the probability of y being y itself in the first place.

People ignore base rates (id
Bank teller - Feminist bank tells - Feminists - It feels like

Representativeness ignores base rates.

D) False positive percentage - it increases the overall likelihood of having a false positive result.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Explain conjunction fallacy

A

Based on this description, subjects are asked which of the following statements is more probable.

(1) Linda is a bank teller.

(2) Linda is a bank teller and a feminist.

Statement (2) is the intuitive response for many people (more than 80% of participants in the original formulation of the problem), who tend to associate Linda with feminism. However, we must note that (1) is more probable than (2), because, once again, statement (1) must be true in order for statement (2) to be true. To choose statement (2) is to commit the conjunction fallacy.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What concepts do we use when we make probability estimations?

A

Representativeness heuristic and availability heuristic

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is the availability heuristic?

A

The more examples of X come to mind, the more likely X is.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Why is availability heuristics challenged? Give precise factors for this.

A

Because our recalling capability is biased (easier or more difficult to recall certain examples), some of the factors are:

Familiarity - (celebrity names) more likely for something to come up to our mind and to think of its occurrence in the real world to be higher

Recency - recent terrorist attacks - increased sense of likelihood to become a victim of terrorist attacks recently (even though they were still substantially lower than rates in 70-80s - twice more likely for another terrorist attack to happen

Salience - estimates of causes of death (cancer vs. car crashes) - hear more likely on news of shark deaths or plane crashes because they’re exceptional!

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What do anchoring and insufficient adjustment mean?

A

Pieces of information that we encounter first tend to matter somewhat more than future information. (more than 65 countries in UN vs less than 10)? High number first => adjust response based on high number reference point; low number first => adjust response based on low number reference point;

NOTE: This anchoring can occur effectively even if the anchor (last digits of social security number) is not relevant to subsequent judgments (determining a price of a product);

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is the main theme reason for bias in the use of System II?

A

Choice context effect - the context in which a choice happens - how alternatives are presented to people can affect the choice a person makes

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Give examples of choice context effect?

A

1) Attraction effect - introducing an intermediate (dominant) option between two options (options becoming 3) can make people be substantially more attracted towards one of the initial two options.

2) Compromise effect - small and medium vs small, medium and large (medium becomes more attractive COMPROMISE option)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What are binary choice, attraction effect, and compromise effect?

A

Binary choice - A and B (two options)

Attraction effect - A, B and C - C being a dominator and increasing the preference towards A or B

Compromise effect - Small and medium vs Small, medium and large - medium becomes a compromise option (between small and large);

NOTE: People justify their choices with System II thinking -adding extra options which seemingly increase rationality, and that still leads to biased results.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What does base rate frequency mean?

A

A frequency or likelihood of an event occurring without intervention.

17
Q

What is the Baye’s rule?

A

People evaluate probabilities based on stereotypes rather than on prior probabilities.

18
Q

When do people use prior probabilities correctly?

A

When there is no additional information that can ‘‘blind’’ them from recalling prior probability’s crucial importance for accuracy.

19
Q

What is posterior probability?

A

The probability of event B occurring, given that event A is true.

I.e. determining probability affected by the proportions in the sample, while taking into account the crucial sample size.

20
Q

What does the Gambler’s fallacy mean?

A

Gambler’s fallacy means identifying a streak of a certain colour as the basis for bigger probability of the opposite colour in the next move.

21
Q

What does ‘‘law of small number’’ mean?

A

In a small sample, even minor random changes result have a big effect on the analysis of data.

22
Q

What does the illusion of validity represent?

A

People choose outcome based on representativeness (disregarding predictive accuracy).

23
Q

What does regression toward the mean signify?

A

Results and scores always come back to the mean and deviate around it. (Outstanding performance first => a bit worse performance next time - regression towards the mean;

24
Q

What are illusory correlations?

A

Phenomenon of perceiving relationship between variables even if no such exists.

25
Q

How do people react to conjunctive and disjunctive events?

A

People overestimate conjunctive events and underestimate disjunctive (separate) events.

26
Q

Do System I and II work separately? (Dhar and Gorlin, 2013)

A

No, they always work together.

27
Q

What makes System II (deliberate processes) decrease?

A

Time pressure, load, depletion (limited info);
The attraction effect decreases during time pressure (minimum amount of time makes System I dominant)

28
Q

How can we test whether a particular problem is due to System I? (Dhar and Gorlin, 2013)

A

By seeing if there is an augmentation of the effect with limited deliberation (System II)