Chapter 13 - Judgement and Decision Making

Inductive Reasoning

Drawing a general conclusion based on specific observation and evidence

-Conclusions are probabilistic (they have uncertainty)

Inductive Reasoning is Affected By:

-Representativeness of the Observation - Are our observations representative of the truth?

-Number of Observations - The more observations, they better our conclusion

-Quality of the Evidence

Inductive reasoning can be effortful, therefore we use heuristics to make conclusions. Heuristics are useful because they often work (but not always)

Availability Heuristic

If an event comes to mind more easily it is more probably or common

Example; “Which is more prevalent in English, words that begin with the letter ‘r’ or words in which ‘r’ is the third letter?”

-Most people believe that words beginning with ‘r’ is more common, but worsd with ‘r’ in the third letter is far more common

Words that start with ‘r’ are easier to come up with, therefore we believe it is more common (availability heuristic)

Problems with the Availability Heuristic

Illusory Correlation - Perceived link between two events when there is no link

Gambler’s Fallacy - Belief that future outcomes are dependent on past outcomes when gambling (ex: person losing a lot might think this means they will win a lot later)

Stereotype - Overgeneralizations about an object or category or person based on previous encounters

Representativeness Heuristic

Making decisions about a category can be based on a representative example of that group

-Good for making decisions about categories

Representativeness Heuristic Ignores Base Rate

Experiment: We read a description of a person, participants have to determine their occupation (librarian or a farmer)

-When making judgements based on resemblance, people tend to ignore the base rate when answering the question

Conclusion - The representativeness heuristic may or may not improve our decision making

Representativeness Heuristic and the Conjunction Rule

-Using the representativeness heuristic leads us to ignore the conjunction rule

Experiment: We read a description of a person, have to determine their future occupation based on their previous experience

-Participants more likely to select the bank teller + feminist movement because of the person’s prior activist movement

-This ignores the conjunction rule - adding an additional criteria decreases the probability of that (being more specific)

So the more probable answer would be that the person is only a bank teller because it is more common

Representativeness Heuristic and Small Samples

The representativeness heuristic leads us to make bad decisions from small samples

Experiment: Two hospitals are presented, larger hospitals has more babies born per day than smaller hospital. For 1 year, each hospital recorded the days that more than 60% of the babies born were boys. Which hospital recorded more such days?

-Participants answer the larger hospital records it, but it is actually the smaller hospital because of its smaller sample size allowing for a greater variation

Effects on Attitudes

Myside Bias - Your prior beliefs and opinions influence how you process new information

-E.g., “Products made in my country are better quality compared to products made elsewhere”

Confirmation Bias - People look for information that confirms their prior beliefs

-Experiment: Participants have to determine a rule in the experiment, then have to find evidence that supports the rule. Participants tended to only look for the evidence that confirmed their rule

False Evidence - False viewpoints can actually become stronger in the face of evidence that they are wrong (backfire effect)

Deductive Reasoning

Determine whether the conclusion logically follows from statements

-E.g., Categorical syllogism, uses premises to make a conclusion

• Premise 1: All men are mortal (all things in category A have property B)

• Premise 2: Socrates is a man (individua X is in category B)

• Conclusion: Socrates is a mortal (Therefore, individual X has property B)

A premise is just a broad statement that we assume to be true

Syllogisms Depend on Validity and Truth

Syllogism can be valid but untrue, or invalid but true

Syllogism 1 - All birds are animals (true) → all animals have 4 legs (untrue) → All birds have four legs (untrue), the structure of the syllogism is valid but the premises are untrue

Syllogism 2 - All of the students are tired → some tired people are irritable → some of the students are irritable, the structure of the syllogism is invalid but the premises are true

-Some people will mistakenly think that this is valid because they focus on the final conclusion

Belief Bias and Syllogism

Belief Bias - If the conclusion seems to be true, the syllogism must be valid

Syllogism 3 (invalid/true); All of the students are tired → some tired people are irritable → some of the students are irritable, provides a believable conclusion

Syllogism 4 (invalid/true): All of the students live in Tucson → some people who live in Tucson are millionaires → some of the students are millionaires, provides an unbelievable conclusion

Between syllogism 3 and 4, they both have the same structure but are both invalid

-Participants were more likely to accept the conclusion of syllogism 3 (and therefore its validity) because the final conclusion is more believable than syllogism 4

Mental Model of Deductive Reasoning

Syllogism: None of the artists are beekeepers → all of the beekeepers are chemists → some of the chemists are not artists

Rule 1 - No artist can be a beekeeper

Rule 2 - All of the beekeepers must be chemists

We can imagine (mental model) the artist with a hat being unable to wear the beekeeper’s hat, but the beekeeper can wear the chemist’s hat

-A normal person can wear both the artist’s hat and the chemist’s hat

A conclusion is only valid if it cannot be refuted by any model of the premises, therefore the conclusion “some of the chemists are not artists” is true

Difference in Syllogism Structure Changes Accuracy

Conditional Syllogisms - “If p, then q’

Basic Premise: If I study, I’ll get a good grade

Conditional Syllogism 1 (p, therefore q); I studied (p), therefore I’ll get a good grade (q)

-This syllogism is an easy premise (it is valid). 97% of participants judged it correctly

Conditional Syllogism 2 (not q, therefore not p); I didn’t get a good grade (q) therefore I didn’t study (p)

-This syllogism is more difficult, it is a valid structure but only 60% of participants got it right

Conditional Syllogism 3 (q, therefore p); I got a good grade (q) therefore I studied (p)

-Invalid, but only 40% of participants correctly identified it

Conditional Syllogism 4 (not p, therefore not q); I didn’t study (p), therefore I didn’t get a good grade (q)

-Invalid, but only 40% of participants correctly identified it

Conditional Reasoning and the Wason Task

Conditional Reasoning - Something has to be true for something else to occur

Experiment; Wason Task

-Cards will have a letter on one side and a number on the other side

-Rule is that if there is a vowel on one side of the card, then there is an even number on the other side, participants must determine if this rule is true by examining as few cards as possible

-Participants must actively look for an example that goes against the rule (the falsification principle)

-Turning over the card with an even number is useless, because we either get a vowel (which confirms the rule) or a consonant (which doesn’t confirm or violate the rule)

-Therefore, we must turn over either the card with an odd number or a card with a vowel to determine if the rule is true or not

Real-World Scenario Applied to Wason Task

Rule: If a person is drinking beer, they must be over 19, but there is no age requirement for drinking soda.

-If a person is drinking beer, you have to check their age. If they’re drinking soda, you don’t

-If you’re unsure what they’re drinking (but know their age) you have too check what they’re drinking

Participants did better on the beer problem than the Wason Task even though they’re solved in the same way

-Explanation is that people might have a permission schema to determine whether or not someone should be doing something

Expected Utility Theory

People are rational and will make decisions in order to maximize expected utility

-Utility is the outcomes that achieve a persons goal/how valuable the outcome of the decision is

Rational decisions are therefore defined as decisions that maximize expected utility

Experiment With Utility

Task is to pull out a red jelly bean from various white jelly beans

-Two jars are presented; one that is smaller with 10 jelly beans (1 is red), and one that is larger with 100 jelly beans (7 are red)

-Participants tend to select the bowl with more red jelly beans, even though the probability in the larger bowl is less than the probability o selecting the red jelly beans in the smaller bowl

-Demonstrates that people don’t always maximize utility in decision making

Decisions are Affected by Emotions

We tend to make decisions based on expected emotions

-We think of a loss as feeling worse compared to how good a gain would feel (this leads to risk aversion)

Coin Flip Experiment

-Either participants win the coin flip and gain 5 dollars, or they lose the coin flip and lose 3 dollars

-Participants make an estimate on how positive they’d feel if they win, and how bad they’d feel if they lose

Results; Before the coin flip, participants reported feeling much worse about losing than how they actually feel when they lose

-But they’re feelings about winning before and after winning the coinflip were similar

So we tend to overestimate how bad we’ll feel if we lose

Decisions are Affected by Context

When we make decisions, we are often influenced by other environmental factors

E.g; Decisions about whether to attempt a c-section results in students being less likely to decide on doing a c-section, but showing better outcomes for c-sections increases their chance of deciding to do a c-section

Decisions Depend on Choice Presentation

Demonstrated in “opt-in” (needing to take action to make a decision) versus “opt-out” (needing to take action to not make a decision)

-For example, you have to ‘opt-in’ to be an organ donor in Canada, but in Australia you have to ‘opt-out’ to be an organ donor

Status Quo Bias - If doing nothing is an option, people would rather do nothing

Therefore in Canada, there is less organ donation (because people rather not take action to opt-in) but in Australia there is more organ donation (people rather not take action to opt-out)

-This is also why internet/cell providers can offer a lower rate in the first year, because they know customers are not likely to leave after the first year is finished

Decisions Depend on Framing

Experiment: Imaginary disease with two treatment options, Program A and B:

Program A - 200 people will be saved (guaranteed)

Program B - 1/3 chance everyone is saved, or a 2/3 chance no one is saved (probabilistic)

-Both have the same utility (1/3 of 600 is 200, so 200 could be saved), but they differ in the presence of risks

-People tend to select Program A because they are displaying risk aversion

The options of Program A and B were then reframed as Program C and D

Program C - 400 people will die (guaranteed)

Program D - 1/3 chance no one dies, 2/3 chance everyone dies

Participants were more likely to select Program D, even though it is the same as Program B

-Program D is frame in a bad connotation (people dying) so more participants are likely to risk take

-When you frame something in a positive connotation (people living), participants are more likely to display risk aversion

Demonstrates the “Framing Effect”

Neuroeconomics

The study of decision-making uses psychology, neuroscience and economics

Ultimatum Game - Two people, one person is the proposer who proses a certain split of money (ex; I get 6 dollars you get 4 dollars). The other person is the ‘accepter’, can either accept the money or can reject it, but if they say ‘no’ nobody gets anything

-Both participants have power, since one can give themselves more money in the cut, but the other can ensure nobody gets money

-In order to maximize expected utility (the rational move) is to accept any amount of money as any amount is greater than nothing

Results; When participants had a 50/50 cut, they accepted 100% of the time. At a 70/30 split, most people accepted. At most unfair splits, people tend to reject the offers

-Pariticipants are not making rational decisions (maximizing expected utility), their decisions are being influenced by their emotions (feelings of being unfair)

Pre-frontal Cortex and Anterior Insula

Anterior Insula - Region in the brain related to emotion

-In the Ultimatum Game, when participants rejected offers they had more activity in the Anterior Insula, but when they accepted offers they had less activity

-Pre-Frontal Cortex (plays a role in decision making), activity in the PFC was similar whether or not participants reject or accept the offer