JUDGEMENT UNDER UNCERTANINTY :
HEURISTICS AND BIASES
Based on paper by Amos Tversky and
Daniel Kahneman
NEERAJA PRAKASH
HEURISTICS
Heuristics are Mental shortcuts that avoid heavy thinking
 we avoid the whole lot of thinking and rather use a rule of thumb, to make
decisions pretty quickly .
 It is quick , informal and it helps to find answer to reasoning problem.
People use heuristics to assess the probability of an uncertain event for
simple judgement.
Heuristics and biases
Heuristics are practical method but it need not be optimal decision.
The decisions have potential to be an in accurate one i.e., decision does not
guarantee accuracy
 It can sometimes result in a cognitive bias, which is the tendency to
draw an incorrect conclusion in a certain circumstance based on cognitive
factor
AVALIBILITY
 Memory based judgements of frequency of a class or probability
of an event.
i.e., making a choice based on easy and immediate examples that
come to your mind when making a decision.
BIASES ; Biases due to the retrievability of instances
The impact made by famous examples, visual examples, or recent examples
causes them to have an especially strong influence on assessment. Classes
whose instances are more easily retrievable will seem larger
E.g. If I read out names of men and women from a list, asked later if the list
contain more men or women. your answer will be biased by how famous the
names were.
Do you think there are more words (three letters or more)
1. Words that start with r
or
2.words where r is the 3rd letter?
Effectiveness of a Search Set
The answer is the latter, but because your
judgement is effected by examples you can
recall.
 We often form mental “search sets” to estimate
how frequent are members of some class; the
effectiveness of the search might not relate
directly to the class frequency
Biases of imaginability
Assessing frequency of class whose instances is not stored in the memory.
The ease in which component can be imagined affect the overall likelihood.
E.g.. Imagining the risk to a Adventurous expedition
Illusory correlation
It is a phenomenon of perceiving a relationship between variables(Typically
people, events or behaviours ) even when no such relationship exists.
Frequency of occurrence of two events together will lead to a false jugement
that there is a strong bond between two events and they are paired
EXAMPLES:
• A football fan believes that every time he wears a specific jersey his team
wins, so each time they play, he will only wear that jersey.
•
• A student scores A+ when he wrote with blue pen.so ,he think that blue
pen can bring him A+ every time.
• A student fails an exam given on a Monday so he determines that he is
unlucky and unable to pass a test if it is administered on future Mondays.
REPRESENTATIVENESS
Judgement of likelihood of instances belong to a category.
Making jugdements based on the similarity of object or a person
to existing persona or prototype.
.
Is Susan a Librarian, a Teacher, or a Lawyer?
DESCRIPTION
• Susan is very shy and withdrawn, invariably helpful, but with
little interest in people, or in the world of reality.
• A meek and tidy soul, she has a need for order and structure, and
a passion for detail
Subjects were given sample of 100 professionals ,all engineers or
lawyers and they were given a brief description and asked the
probability whether the person is engineer rather than lawyer.
EXPERIMENTS
1.subjects were told that group contain 70 engineers and 30 lawyers
2.Subjects were told that groups contain 30engineers and 70
lawyers
The subjects corretly gave the response that unknown individual was an engineer at 0.7 and 0.3
in 1 and 2 respectively
Insensitivity to prior probability of outcomes
• Personal sketch
• Sam is a 30 year old man. He is married with no children. A man of high ability and high
motivation, he promises to be quite successful in his field. He is well liked by his
colleague
But when description is given ,the subjects responded that probability of sam being an
engineer is 0.5 rather than lawyer
CONCLUSION : when no specific evidence is given, prior probabilities are used properly;
when worthless evidence is given, prior probabilities are ignored.
Kahneman & Tversky concluded that when no specific evidence is given, prior probabilities
are used properly; when worthless evidence is given, prior probabilities are ignored.
Consider large hospital, 45 babies are born each day, whereas only 15 are born in the
smaller hospital. 50% of all babies are boys.(exact % varies). For 1 year each hospital
recorded the days on which more than 60% of the babies born were boys. Which hospital
do you think recorded more such days?
The larger hospital
The smaller hospital
About the same
INSENSITIVITY TO SAMPLE SIZE
Failure to appreciate the role of sample size even when it is emphasised in
problem.
 statistics tells us that we are much more likely to observe 60 percent of
male babies in a smaller sample than in a larger sample.
• A coin is to be tossed 6 times. Which sequence is more likely?
•1.H T H T T H
•2.H H H T T T
Misconceptions of chance
• They are both equally likely, but most people think the first is
more likely, because it "looks more random".
DESCRIPTION OF COMPANY PROFIT AND SHARE PRICE
FAVORABLE DESCRIPTION ??
UNFAVOURABLE DESCRIPTION ??
. Insensitivity to predictability
• This example illustrates that one does not consider the reliability and accuracy of the
descriptions since s/he only considers whether they are favorable or not. This may result
in wrong predictions about future values such as profit,.
People predict future performance mainly by similarity of description to
future results
"People express more confidence in predicting the final grade-point average of a student
whose first-year record consists entirely of B's than in predicting the gradepoint average
of a student whose firstyear record includes many A's and C's.“
But In reality, when objects are correlated the accuracy will be lower, prediction based on
independent events are more accurate
The Illusion of Validity
Internal consistency of input pattern increases confidence
good match between input information and output classification
or outcome often leads to unwarranted confidence in the
prediction
MISCONCEPTION OF REGRESSION
Regression to the mean is all about how data evens out. . Our performance
always varies around some average true performance. Extreme
performance tends to get less extreme the next time. Why? Testing
measurements can never be exact. All measurements are made up of one
true part and one random error part
People expect predicted outcomes to be as representative of the input as possible
Failure to understand regression may lead to overestimate the effects of punishments and
underestimate the effects of reward
ADJUSTMENT AND ANCHORING
It is the human tendency to rely on first piece of
information while making decisions
People make adjustments to initial value to get the final
answer.
Eg.car price negotiations
Two sets of students were given the tasks of finding out the
product of
1st set :8*7*6*5*4*3*2*1
Your best guess?
...........
Insufficient adjustment
Kaheman and tversky discovered this anchors happen
all the time even when anchors are totally arbitrary.
Linda is 31 years old, single, outspoken, and very bright. She
majored in philosophy. As a student, she was deeply concerned
with issues of discrimination and social justice, and also
participated in anti-nuclear demonstrations.
Which is more likely?
• Linda is a bank teller.
• Linda is a bank teller and is active in the feminist movement.
Biases in the evaluation of conjunctive and
disjunctive events.
judgments of probability indicate that people tend to
overestimate the probability of conjunctive events
and to underestimate the probability of disjunctive event.
I.e., Assuming multiple specific conditions are more
probable than a single general one.
• THANK YOU

HEURISTICS AND BIASES

  • 1.
    JUDGEMENT UNDER UNCERTANINTY: HEURISTICS AND BIASES Based on paper by Amos Tversky and Daniel Kahneman NEERAJA PRAKASH
  • 2.
    HEURISTICS Heuristics are Mentalshortcuts that avoid heavy thinking  we avoid the whole lot of thinking and rather use a rule of thumb, to make decisions pretty quickly .  It is quick , informal and it helps to find answer to reasoning problem. People use heuristics to assess the probability of an uncertain event for simple judgement.
  • 3.
    Heuristics and biases Heuristicsare practical method but it need not be optimal decision. The decisions have potential to be an in accurate one i.e., decision does not guarantee accuracy  It can sometimes result in a cognitive bias, which is the tendency to draw an incorrect conclusion in a certain circumstance based on cognitive factor
  • 4.
    AVALIBILITY  Memory basedjudgements of frequency of a class or probability of an event. i.e., making a choice based on easy and immediate examples that come to your mind when making a decision.
  • 5.
    BIASES ; Biasesdue to the retrievability of instances The impact made by famous examples, visual examples, or recent examples causes them to have an especially strong influence on assessment. Classes whose instances are more easily retrievable will seem larger E.g. If I read out names of men and women from a list, asked later if the list contain more men or women. your answer will be biased by how famous the names were.
  • 6.
    Do you thinkthere are more words (three letters or more) 1. Words that start with r or 2.words where r is the 3rd letter?
  • 7.
    Effectiveness of aSearch Set The answer is the latter, but because your judgement is effected by examples you can recall.  We often form mental “search sets” to estimate how frequent are members of some class; the effectiveness of the search might not relate directly to the class frequency
  • 8.
    Biases of imaginability Assessingfrequency of class whose instances is not stored in the memory. The ease in which component can be imagined affect the overall likelihood. E.g.. Imagining the risk to a Adventurous expedition
  • 9.
    Illusory correlation It isa phenomenon of perceiving a relationship between variables(Typically people, events or behaviours ) even when no such relationship exists. Frequency of occurrence of two events together will lead to a false jugement that there is a strong bond between two events and they are paired
  • 10.
    EXAMPLES: • A footballfan believes that every time he wears a specific jersey his team wins, so each time they play, he will only wear that jersey. • • A student scores A+ when he wrote with blue pen.so ,he think that blue pen can bring him A+ every time. • A student fails an exam given on a Monday so he determines that he is unlucky and unable to pass a test if it is administered on future Mondays.
  • 11.
    REPRESENTATIVENESS Judgement of likelihoodof instances belong to a category. Making jugdements based on the similarity of object or a person to existing persona or prototype. .
  • 12.
    Is Susan aLibrarian, a Teacher, or a Lawyer? DESCRIPTION • Susan is very shy and withdrawn, invariably helpful, but with little interest in people, or in the world of reality. • A meek and tidy soul, she has a need for order and structure, and a passion for detail
  • 13.
    Subjects were givensample of 100 professionals ,all engineers or lawyers and they were given a brief description and asked the probability whether the person is engineer rather than lawyer. EXPERIMENTS 1.subjects were told that group contain 70 engineers and 30 lawyers 2.Subjects were told that groups contain 30engineers and 70 lawyers The subjects corretly gave the response that unknown individual was an engineer at 0.7 and 0.3 in 1 and 2 respectively
  • 14.
    Insensitivity to priorprobability of outcomes • Personal sketch • Sam is a 30 year old man. He is married with no children. A man of high ability and high motivation, he promises to be quite successful in his field. He is well liked by his colleague But when description is given ,the subjects responded that probability of sam being an engineer is 0.5 rather than lawyer CONCLUSION : when no specific evidence is given, prior probabilities are used properly; when worthless evidence is given, prior probabilities are ignored. Kahneman & Tversky concluded that when no specific evidence is given, prior probabilities are used properly; when worthless evidence is given, prior probabilities are ignored.
  • 15.
    Consider large hospital,45 babies are born each day, whereas only 15 are born in the smaller hospital. 50% of all babies are boys.(exact % varies). For 1 year each hospital recorded the days on which more than 60% of the babies born were boys. Which hospital do you think recorded more such days? The larger hospital The smaller hospital About the same
  • 16.
    INSENSITIVITY TO SAMPLESIZE Failure to appreciate the role of sample size even when it is emphasised in problem.  statistics tells us that we are much more likely to observe 60 percent of male babies in a smaller sample than in a larger sample.
  • 17.
    • A coinis to be tossed 6 times. Which sequence is more likely? •1.H T H T T H •2.H H H T T T
  • 18.
    Misconceptions of chance •They are both equally likely, but most people think the first is more likely, because it "looks more random".
  • 19.
    DESCRIPTION OF COMPANYPROFIT AND SHARE PRICE FAVORABLE DESCRIPTION ?? UNFAVOURABLE DESCRIPTION ??
  • 20.
    . Insensitivity topredictability • This example illustrates that one does not consider the reliability and accuracy of the descriptions since s/he only considers whether they are favorable or not. This may result in wrong predictions about future values such as profit,. People predict future performance mainly by similarity of description to future results
  • 21.
    "People express moreconfidence in predicting the final grade-point average of a student whose first-year record consists entirely of B's than in predicting the gradepoint average of a student whose firstyear record includes many A's and C's.“ But In reality, when objects are correlated the accuracy will be lower, prediction based on independent events are more accurate
  • 22.
    The Illusion ofValidity Internal consistency of input pattern increases confidence good match between input information and output classification or outcome often leads to unwarranted confidence in the prediction
  • 23.
    MISCONCEPTION OF REGRESSION Regressionto the mean is all about how data evens out. . Our performance always varies around some average true performance. Extreme performance tends to get less extreme the next time. Why? Testing measurements can never be exact. All measurements are made up of one true part and one random error part People expect predicted outcomes to be as representative of the input as possible Failure to understand regression may lead to overestimate the effects of punishments and underestimate the effects of reward
  • 24.
    ADJUSTMENT AND ANCHORING Itis the human tendency to rely on first piece of information while making decisions People make adjustments to initial value to get the final answer. Eg.car price negotiations
  • 25.
    Two sets ofstudents were given the tasks of finding out the product of 1st set :8*7*6*5*4*3*2*1 Your best guess? ...........
  • 26.
    Insufficient adjustment Kaheman andtversky discovered this anchors happen all the time even when anchors are totally arbitrary.
  • 27.
    Linda is 31years old, single, outspoken, and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice, and also participated in anti-nuclear demonstrations. Which is more likely? • Linda is a bank teller. • Linda is a bank teller and is active in the feminist movement.
  • 28.
    Biases in theevaluation of conjunctive and disjunctive events. judgments of probability indicate that people tend to overestimate the probability of conjunctive events and to underestimate the probability of disjunctive event. I.e., Assuming multiple specific conditions are more probable than a single general one.
  • 29.