Presentation debiasing m-azimi,amshirazi,hdarzi

1

DBIASiNG!BY RICHARD P.LARRICK

INSTRUCTOR:DR.ISAIEE

Presented by: Amir MohammadShirazi Hamid Darzi Mehran Azimi

2

OUTLINE Introduction(Rationality & Debiasing) Nature of Biases Motivational Strategies Cognitive Strategies Technological Strategies Adoption & Diffusion Future of Debiasing

3

INTRODUCTION

Bridge The Gap

Descriptive Behavior

Normative Ideal

Debiasing

Strategies

4

INTRODUCTION

Approaches:

1.People posses normative strategies2.Intuitive and normative strategies

5

INTRODUCTIONPrescriptive Decision Making:

1.By Status Quo and Environment2.Increasing Decision Making Skills

Two Groups:3. Meliorists4. Apologists

Robustness

6

INTRODUCTIONDebiasing requires Intervention

TWO STEPS:1.Recognition2.Correction

People Recognize and Correct their own Biases3. Feed back problem & Source of Error4. Uncertainty of Environment(Self Serving attributions

of ability)5. Compensation of Cultural Mechanism

7

Approaches for Debiasing

• Specific Bias• Source of Bias

The nature of Biases

8

AssociationisticStrategic (system 2)

Psychophysical

The nature of Biases

Second Approach: Source of Biases

9

Psychophysical based errors

Shifting reference point

Inconsistent judge

The nature of Biases

10

Association based errors

Activation of associated cognitions

Inhibition of unassociated cognitions

Narrow band information

The nature of Biases

11

Strategy based errors

intuitive repertoire

Benefit-Cost calculation

The nature of Biases

12

13

Debiasing Strategies

Technical Strategy

Motivational StrategyCognitive Strategy

14

accountability

incentive

15

Incentives

Incentives

16

LOST PILOT EFFECT

Cognitive Capital

I DON’T KNOW WHERE I AM GOING, BUT I’M MAKING GOOD TIME.

17

Incentives

Incentives

CLERICAL & MEMORIZATION TASKS

18

Accountability

Accountability

Preemptive self criticismADVANTAGES

Greater effortAccuracyBetter performancePreventing biasesNeed to look consistent

19

Potential Problems

Give the people what they wants

Justification-base Decision Bias

Accountability

Accountability

20

Consider the oppositeTraining in rulesTraining in representationTraining in biases

COGNITIVE STRATEGIES

21

CONSIDER THE OPPOSITE

Anchoring

Overconfidence

Hindsight

ASK:What are some reasons that my initial judgment might be wrong?

What about simply listing reasons?

22

TRAINING IN RULES

How inferior strategies can be replaced by the better one?

Experience Training

23

TRAINING IN RULES

The rules can be thought in the form of mathematics and logic

24

TRAINING IN RULES

The rules can be applied in everyday life decisions

25

TRAINING IN REPRESENTATIONS

People reason better about “frequencies” than about “probabilities”

ProbabilitiesFrequencies

26

• Information as “Probabilities” %0.8 of women have breast cancer The probability of “positive result” for

those who have cancer is %90The probability of “positive result” for

those who don’t have cancer is %7

TRAINING IN REPRESENTATIONS

27

• Information as “Frequencies”‘Eight’ out of every 1000 women have

breast cancerOf this ‘eight’ women ‘seven’ will have

a positive resultOf the ‘992’ women who don’t have

cancer ‘70’ will still have a positive result

TRAINING IN REPRESENTATIONS

28

TRAINING IN BIASES

Is “Awareness” Enough?

29

TECHNOLOGICAL

STRATEGIES

30

GROUP DECISION MAKING

Increase sample size of experienceSynergyError-checking system

Social cost of rejection Free riding Anchoring

Disadvantages

Benefits

31

LINEAR MODELS& STATISTICAL TECHNIQUES

Experts can be replaced by models based on their judgments

Theses models can make better prediction than experts .

U(A) = w1 X1 + w2 X2 + w3 X3

Proper & ImproperLinear Models

32

MUA (Multiattribute Utility) Analysis

Decision Analysis

Decision support system (DSS)

33

The main question is:

ADOPTION AND DIFFUSION

How to encourage people to adopt better decision

strategies?

34

People resist being debiased

ADOPTION AND DIFFUSION

Me?!Biased?!

IMPOSSIBLE!!!

35

People resist being Debiased

“Doing it Wrong” for all these years?!

Relinquish control over a decision process

Techniques are alien and complex and the benefits are noisy, delayed or small

ADOPTION AND DIFFUSION

36

Adoption and diffusion

Promote Adoption

Compliance to behavioral norm

Internalization of beliefs

Vs.Internalization of beliefs

Compliance to behavioral norm

37

Management Tools

ADOPTION AND DIFFUSION

TQM Six Sigma

38

ADOPTION AND DIFFUSION

Statistical tools

THE FIRST TO BE ABANDONED

Defect?Where is it?

39

• Senior management do not practice endorsement

• Doubtfulness of outside consultants

• Separation from coworkers

Adoption and diffusion

40

ADOPTION AND DIFFUSION

“Learn it, Teach it, Use it, Inspect it”How can it be fixed?

41

Adoption and diffusion

Simple

Domain-Specific

Social

Top-Down

Complex

Domain-General

Indivisual

Bottom-Uop

Vs.

Cognitive Reapair

42

The Most Successful

Adoption and diffusion

Simple

Domain-Specific

Social

Top-Down

Complex

Domain-General

Individual

Bottom-Up

43

Informal Debiasing techniquesADOPTION AND DIFFUSION

44

The higher you are, the harder you fall !f

45

ANY QUESTION?