Decision making + Problem solving

Question 1

What is higher-order cognition based on?

Answer 1

• Higher-order cognition is built on other aspects of cognition
o Does something with the products of perception, memory and language
• Key aspect is creating and using knowledge, rather than extracting knowledge

Question 2

What is thinking?

Answer 2

• Thinking- the systematic transformation of mental representations of knowledge to characterise actual or possible states of the world often in service of goals

Question 3

Why is thinking difficult?

Answer 3

• It is built on other aspects of cognition that are imperfect
o Perception is easy and unconscious
o Thinking is hard and conscious
o Information provided may be inaccurate or incomplete
• It is recent in terms of evolution
o Frontal lobes most distinguish the human brain
o Frontal lobes activated by all forms of thinking
 Activation of pre-frontal cortex
• We are aware we make mistakes

Question 4

What inaccuracies and thinking traps are we prone to?

Answer 4

o Confirmation bias- tendency to weigh evidence more strongly if it fits with our pre-existing beliefs our perception of facts favours the hypotheses that we already have
o Simple rules
o Motivated reasoning -emotion-biased decision making

Question 5

Describe the case study of the Piltdown man hoax

• What it is
• Why it was considered a big thing
• How it became suspicious
• How it was disproved

Answer 5

• The Piltdown man hoax which exploited our inaccuracies in thinking:
o 1908 to 1915-fossil parts found near Piltdown, UK
o Clearly human skull and ape-like jaw
 Missing link between apes and humans- modern skull but ancient jaw
 Suggested big brain developed first, and ape-like features disappeared second
o Appeared 500,000 years old
o Reconstruction suggested certain wear patterns on molars- too with that pattern discovered in Piltdown
o Piltdown man eventually became an important piece of evidence for evolution
o However, Piltdown man evidence ran into some problems
 Contradicted with other evidence found:
• In 1924, Dart found earliest hominid skull in Taung, South Africa
• Ignored for 20 years due to Piltdown man
• However, more findings suggested that the big brain came after the disappearance of ape-like features
o Eventually, the Piltdown man became the outlier finding, causing it to be re-examined as evidence:
 1950 fluorine test suggested the skull was only 50,000 not 500,000 years old
 1953- through electron microscope inspection, found the wear on the teeth was made with a metal file: the abrasion was artificial

Question 6

How did the Piltdown man exploit errors in our way of thinking?

Answer 6

o Piltdown man was eventually recognised as a hoax, after 40 years, made by one person, that exploited:
 Confirmation bias that the big brain came first and that Europe and Asia was where humans evolved
• Didn’t look at evidence as much as should have as it confirmed pre-existing bias
• African specimen were dismissed as strongly believed that humans evolved from Europe and Asia
• Uncertainty and fact that only had a few bone fragments let conformation bias take place
 Simple rule and short-cut that we should trust the experts
• Experts assumed experts in other domains had confirmed evidence from their area
• People had doubts in their area of expertise, but dismissed them because thought that other experts fully supported it
• Trusting experts may not have worked as necessary knowledge didn’t exist yet
 Motivated reasoning
• England felt national pride of having human ancestor, as had rivalry with Germany and their Neanderthals- felt competition and hence wanted this finding to be true
 Mindsets and representations-
• Treating the Piltdown man finding as legitimate made it harder to recognise it as a hoax- if it had been treated sceptically from the beginning, it would have been easier to detect its falsity.
 Overestimating value of learning from past situations-
• Piltdown man case followed same pattern as many of the previous fossil finds that advanced knowledge about evolution

Question 7

Why do errors in our thinking occur?

Answer 7

• Errors in thinking can be due to problems in different processes
o Errors because other processes are not effective: there might be misperception of misremembering, or don’t have the right information
o How we understand situations is critical to how we deal with them

Question 8

Why are cognitive illusions useful?

Answer 8

• Cognitive illusion are like optical illusions:
o Systematic
o Errors tell us something about how an extremely effective system works
 Errors tell us about how thinking works- informative

Question 9

What are vital components of our everyday thinking and decision making?

Answer 9

-We have to deal with constant uncertainty
-We use short-cuts
-We use old information in a new way
o Often apply what was learned in one situation to new situations
o Use analogies involving what is known to understand new problems
-Representation is critical
o How situations are represented determine how they are dealt with, and can change how easy it is to draw the right conclusions
o Mindset important for approach of situation can influence what is done with information

Question 10

What are 4 different aspects of thinking?

Answer 10

• Problem solving- generating route to a goal
• Decision making- evaluating alternative outcomes or making choices
• Reasoning- drawing further inferences from current knowledge and beliefs
• Expertise and skill acquisition- knowledge as routine

Question 11

How do we deal with uncertainty in thinking? What is an issue with doing this?

Answer 11

o Try to deal with uncertainty by filling gaps with logical information and deductions
o Make decisions that try to predict the future despite uncertainty
 Make choices on how we think the future will turn out: lots of ways we think try to reduce uncertainty or try to make decisions despite it such as implementing short-cuts
-Issue: problem of induction

Question 12

What is the problem of induction?

Answer 12

o	The problem of induction (Hume)- the philosophical question of whether inductive reasoning leads to knowledge understood in the classic philosophical sense, highlighting the apparent lack of justification for:
	Generalising about the properties of a class of objects based on some number of observations of particular instances of that class
	Presupposing that a sequence of events in the future will occur as it always has in the past

Question 13

Why and how do we use shortcuts? What is a problem with this?

Answer 13

o We use heuristics (short-cuts) because we lack all information
o Rules of thumb that are often effective, but are not guaranteed
o Replaces hard questions regarding the future with easier ones through simple rules
o Can lead to predictable errors
 Can give over or under-importance to things

Question 14

What is a problem?

Answer 14

• Problem- if a living organism has a goal that is not immediately available, but does not know how this goal is to be reached.

Question 15

What is a well-defined problem?

Answer 15

• Well-defined problems- ones in which all aspects (initial state, range of possible strategies, goal) of the problem are clearly specified

Question 16

What is an ill-defined problem?

Answer 16

• Ill-defined problems-aspects (initial state, range of possible strategies, goal) of the problem are unspecified

Question 17

What is a knowledge rich problem?

Answer 17

• Knowledge-rich problems- can only be solved by those having much relevant specific knowledge

Question 18

What is a knowledge lean problem?

Answer 18

• Knowledge-lean problems- most information needed to solve the problem is contained in the initial problem statement: no additional information is required

Question 19

What is problem solving?

Answer 19

o There are two states of affairs
o The agent is in one state and wants to be in another state
o It is not apparent to the agent how the gap between the two states is to be bridged
o Bridging the gap is a consciously guided multi-step process

Question 20

What are 3 major aspects to problem solving?

Answer 20

o It is purposeful
o It involves controlled processes and is not totally reliant on automatic processes
o A problem exists when someone lacks the relevant knowledge to produce an immediate solution.

Question 21

Are only humans able to problem solve? Give an example (Weir, Chappell and Kacelnik study 2002)

Answer 21

• Other species can see goals and find paths to them
o Studying animals leads to insight in both human and animal behaviour
o Weir, Chappell and Kacelnik (2002)
 New Caledonian Crow tries to reach hooked bucket with meat in it which is inside a tube, but the tube is too narrow for the head of the crow
 After a few tries with beak and straight wire, bends the wire and eventually hooks the bucket- the crow created a tool
 A female retrieved food on 10/17 trials, whilst a male did so just once with a straight wire- females had higher success and more ingenuity than male crows
 This is spontaneous learning, not stimulus-response learning
• But evidence of crows in the wild using curved twigs for getting insects, so some wild behaviour may translate to this problem

Question 22

What are 3 different approaches to problem solving?

Answer 22

• Gestalt approach
• Problem solving as search
• Knowledge and problem solving

Question 23

What is reproductive thinking?

Answer 23

o Reproductive thinking- systematic reuse of previous experiences

Question 24

What is productive thinking?

Answer 24

o Productive thinking- novel restructuring of the problem, which requires insight

Question 25

What is the Gestalt approach to problem solving?

Answer 25

o First systematic attempt to understand the psychology of problem solving o Emphasized importance of representation- changes in perspective, prior knowledge and assumptions o Approach demonstrated a number of important phenomena  Especially restructuring representations • Wertheimer 1945

Question 26

Describe Kohler (1925) observations?

Answer 26

o Kohler (1925)  Studied chimps solving problems  Chimps had problem solving mechanism  Set up problems for the chimps • Banana hung above compound above the chimp’s reach and jump range • Chimps initially jumped, then stacked boxes to reach the banana

Question 27

What is the problem with the Gestalt approach?

Answer 27

o Gestalt approach seemed largely descriptive- did not describe the process of problem solving

Question 28

What is the problem solving as a search approach and how did it come about?

Answer 28

o General problem solving methods  Use heuristic strategies • Help obtain solution, but not guaranteed • Contrast to an algorithm, which is guaranteed to produce an answer  Approach came out of early work on artificial intelligence o Newell and Simon (1970) proposed problem solving as a physical or mental search through a problem space of possibilities  Start with a situation we want to transform into something new • Do so by applying the different options we have  Can’t be exhaustive search, so need strategies • Can’t search all possible states

Question 29

What is a problem space?

Answer 29

 Problem space- consists of initial state of problem, goal state, all possible mental operators that can be applied to any state to change it into a different state, and all intermediate problem states

Question 30

What is functional fixity?

Answer 30

• Functional fixity o We develop mental sets: tend to see things/objects in a certain way, use certain solutions  Mistakenly assume that any given object has only a limited number of uses o Fail to notice novel uses of objects- need to recognise that an object that has one purpose could be used differently o Need to overcome functional fixity

Question 31

Describe Duncker's 1945 experiment

Answer 31

o Duncker’s 1945 candle problem-  Box of thumbtacks sitting on the table, candle and matches  Need to attach candle to the wall and be able to burn upright  Solution: pin box to wall with thumbtacks and put candle in it and burn it

Question 32

Describe Adamson's 1952 results involving Duncker's 1945 experiments of- what is this an example of?

Answer 32

o Adamson 1952 based on Duncker’s 1945 candle problem  Found that 86% solved when box was empty, 41% when not  Requires box to be seen in a different way (not a container, but a platform)  Hence requires a different mental representation, and easier when the presentation of the box is changed  Hence, overcoming function fixity can be a key to a solution

Question 33

What are set effects?

Answer 33

o Fixed on the way we do/think about solutions-Procedural fixity  Continuing to use a previously successful problem-solving strategy even when it is inappropriate or sub-optimal

Question 34

Describe Luchin's 1942 experiment and what t is an example of

Answer 34

o Luchin’s 1942 problems:  Three water jugs of given capacities  The goal of obtaining an exact amount of water  Gave people whole set of problems which can be solved with exact same procedure  Then gave people another problem with an easier solution to the previous ones  Found that people picked the harder solution they had gotten familiar with in first set of problems to solve second set of problems (which had an easier solution • After a sequence of similar solutions, most people fail to notice an easier solution o Set effect (or Einstellung)

Question 35

What is the duration of functional fixity and set effects? What is a case study of this?

Answer 35

• Functional fixity and set effects can be long lasting o 1846-Urbain Le Verrier applied Newton’s laws to anomalies in Uranus’s orbit to predict the location of a planet  He calculated where the planet influencing Uranus’ orbit would be  Also noticed wobble in Mercury’s orbit, so in 1859 predicted the planet Vulcan • But was wrong about Vulcan • But kept issuing new alerts until he died in 1877- Vulcan was never confirmed • Many people believed that Vulcan existed (some saw it) because it fit with current science o Einstein’s theory of relativity eventually explained Mercury’s weird orbit  Was victim of set effect- solution worked one time so thought it would work again and again

Question 36

What is a state?

Answer 36

• State-specification of situation

Question 37

What is a goal?

Answer 37

• Goal-the desired state

Question 38

What is an operator?

Answer 38

• Operator- an action that changes one state into another

Question 39

What is a solution?

Answer 39

• Solution- a sequence of operators that transforms initial state into goal state

Question 40

What are constraints?

Answer 40

• Constraints- restrictions on what can be done

Question 41

What is a method?

Answer 41

• Method-procedure for performing a search for a solution

Question 42

What is the generate-test method for problem solving as a search?List advantage and disadvantages

Answer 42

``` • Generate-test o Randomly generate solution, then test it  Advantage- requires no knowledge  Disadvantage- but sometimes ineffectual • Generation may be hard • Testing could be hard • Search space may be very large • May reach solution too slowly ```

Question 43

What are two problems that need to be overcome when problem solving according to the Gestalt approach?

Answer 43

- Functional fixity | - Set effects

Question 44

Evaluate the theory of problem solving as a search-advantages and disadvantages

Answer 44

o Evaluation of theory-  Normative theory-specifies ideal • People are often so as systematic as mean-analysis suggests • Do less planning than predicted  Provides framework for thinking about problem solving • Provides good terminology • But not refutable • Can build theories within framework, such as CSP theory  Disadvantage- underplays the role of knowledge in problem solving • Lot of influence of knowledge in solving problems in real life • Both representation and process are important

Question 45

What can knowledge and problem solving interact?

Answer 45

o Knowledge can eliminate problem solving- if know how to reach goal, then no longer have a problem o Knowledge can provide us with the most creative solutions-> use old knowledge to solve new problems o Knowledge can sometimes be a barrier to problem solving  It can provide the wrong representation

Question 46

What is insight?

Answer 46

• Insight-any sudden comprehension, realisation, or problem solution that involves a reorganisation of the elements of a person’s mental representation of a stimulus, situation or event to yield a nonobvious or nondominant interpretation

Question 47

What is a change in representation?

Answer 47

• Change in representation- overcoming initial knowledge block

Question 48

What is the impact of right or wrong representation on a solution?

Answer 48

* If have right representation of a problem then often just work our way through to the solution * If have wrong representation than no amount of effort will get us to the solution * Re-representation can instantly make a hard problem easy to solve

Question 49

What are 3 ways to change problem representation for insight occur (Ohlsson 1992)

Answer 49

 Constraint relaxation-inhibitions on what is regarded as permissible are removed  Re-encoding- some aspect of the problem representation is reinterpreted  Elaboration- new problem information is added to the representation

Question 50

Describe the re-representation process

Answer 50

o Re-representation process (Ollinger et al.2014)  Prior knowledge and perceptual aspects of a problem lead to the formation of a problem representation  Followed by a search process • If search process is repeatedly unsuccessful, there is an impasse/block • New problem representation is formed to try to overcome impasse and is followed by another search process

Question 51

What brain areas are correlated with insight?

Answer 51

• Anterior superior temporal gyrus, anterior cingulate cortex and prefrontal cortex have been correlated with insight

Question 52

What is an analogy

Answer 52

• Analogy-comparison between two objects, or systems of objects, that highlights respects in which they are thought to be similar

Question 53

Is functional fixity useful in problem solving?

Answer 53

o Need to overcome functional fixity

Question 54

What is the difference reduction aspect of problem solving as a search?

Answer 54

• Difference reduction (hill climbing) o Try to apply operator(s) to reduce difference between current state and goal state o Mostly used when problem solver has no clear understanding of the problem structure o Focus on short-term goals

Question 55

What are the advantages/disadvantages of difference reduction aspect of problem solving as a search?

Answer 55

o Advantages-  Steps can be small  Don’t have to know much, just what gets you closer to the goal • Only have to definitely know where goal is o Disadvantages-  Requires some knowledge • Measurement of distance from the goal • Knowledge of goal  May not be desirable to getting closer to goal on each step-there may be advantage of initially being farther from goal to get closer to goal

Question 56

What is means-end analysis?

Answer 56

1. Compare current state to goal state and identify differences 2. Select an operator to reduce the largest difference (includes operators that can’t presently be used) 3. If operator cannot be applied set subgoal of creating preconditions for its application 4. Return to step 1 until goal is reached

Question 57

Describe means-end analysis using the Tower of Hanoi as an example and how means-end analysis can be used to solve this problem

Answer 57

o Example-Tower of Hanoi  3 posts (A B and C) and 3 disks (1,2,3) • 1,2,3 are of different sizes, from smallest to largest  Goal-move 3 discs from post A to post C by moving one disk at a time • Constraint-must not place a larger disk onto a smaller disk  Means-end analysis could be used: • Subgoal 1a)-move disk 3 to post C o Subgoal 2 to achieve subgoal 1-Move disk 1 and disk 2 to post B  Subgoal 3- move disk 1 to post C • Operator can meet subgoal 3 so don’t set another • Subgoal 1b)-move disk 2 to post C o Subgoal 2- to move disk 1 to post A

Question 58

What is evidence for the means-end analysis

Answer 58

 Egan and Greeno (1974)- a problem becomes more difficult with more subgoals • Probability of error increased with number of subgoals necessary  Verbal protocols (Anderson, 1983) • Had people think aloud and verbalise their thoughts while working out problems o Found that people spontaneously set subgoals  Catrambone (1995)- better performance in math problem solving when instructed to set subgoals

Question 59

What is evidence against the means-end analysis

Answer 59

 Patsenko and Altmann(2010) showed that people were not using detailed planning in tower of Hanoi • Changing number of disks during task did not change actions o Would add/subtract a disc while person was solving the problem and not looking (eye saccades) o If planning ahead, would have disrupted entire problem- however, found that people were not very disrupted o Next move was mostly triggered by the current state of the problem rather than by a pre-formed plan  Omerod, MacGregor, Chronicle, Dewald and Chun (2013)- • Gave 9 ball pproblem to 26 students who suggested the following: o 9 said 5 vs 4 o 11 said 4 vs 4 o 2 said 3 vs 3 o 1 said 2 vs 2 o Found that most participants appeared not to have thought much about second weighing, instead trying to maximise number of balls used in problem

Question 60

What is the 9 ball problem?

Answer 60

• Nine metal balls that all look identical, but one is slightly heavier • Figure out which is the heavier ball • Can use a balance scale o Constraint- can only do 2 weighings • Solution- o Put 3 balls on each side of the scale-isolate the group of 3’s that has the heavier ball o Weigh 2 balls from that group of 3 and determine which is the heavier ball

Question 61

What is the Criterion of Satisfactory Progress theory or refining problem spaces? Describe a piece of proof for this theory

Answer 61

o Less systematic than means-end analysis o Three general heuristics in problem solving-  Maximisation of progress towards goal, effectively difference reduction  Minimization of search space • But sometimes this means that the search space selected doesn’t even have the solution-added constraint  Trial and error o Ormerod et al (2013) saw 9-ball problem results as consistent with their criterion-of-satisfactory-progress (CSP) theory  Heuristics make 9-ball problem harder to solve than 7-ball version

Question 62

What are different components/theories of problem solving as a search approach and where did this approach come from?

Answer 62

- Generate-test - Difference reduction (hill climbing) - Means-end analysis - CSP theory-refining problem spaces -studying artifical intelligence

Question 63

What is a consequence of not applying the right constraints in problem solving

Answer 63

o Wrong constraints can be part of incorrect representation

Question 64

Describe the importance of applying the right constraints using the nine-dot problem

Answer 64

o Nine-dot problem (Maier 1930)  Connect all nine dots set out in a square with four straight lines without lifting your pencil from the paper  Solution involves going outside the square • However, found people kept their lines in the square- each line finish on dot  Problem appears hard because unnecessary constraints are part of the representation

Question 65

What is incubation period and why is it useful?

Answer 65

• Incubation period could provide a reset- move away from one fixed representation and can embrace new representation after incubation period o Incubation-when a problem is set aside for some time: subconscious mind continues to work towards a solution during incubation and so incubation facilitates problem solution

Question 66

When are incubation effects stronger?

Answer 66

 Sio and Ormerod meta study (2009) • Incubation effects stronger with creative problems have multiple solutions • Effects larger when there is fairly long preparation time prior to incubation

Question 67

What is the importance of hints and the right representation in problems? Describe with a study involving the nine-dot problem

Answer 67

o Weisberg and Alba (1981)-25% of subjects given a hint solved the nine-dot problem  None solved problem without the hint  Right representation is not the same as solution, but may just be a pre-requisite  Having right representation doesn’t necessarily guarantee solution-might be because we try to minimise problem space • Fits with CSP theory o Minimize search space, maximise lines

Question 68

Where do new representations come from?

Answer 68

o New representations come from experience from other situations  Often see new situation as being like a familiar one

Question 69

What is transfer? Distinguish between positive and negative transfer

Answer 69

o Transfer- the application of old knowledge to a new problem  Negative transfer: functional fixedness and mental state  Positive transfer- using prior knowledge to help solve a problem

Question 70

Are analogies used in transferring information?

Answer 70

o Analogies provide one form of transfer  Analogies are relationships between two similar situations, problems or concepts • Particularly if dissimilar on the surface, but have a similar structure

Question 71

Where can analogies be seen and what can they result in? Give an example

Answer 71

 See analogies in standardised tests  Analogies can be the basis of very creative solutions- • William Gilbert in 1600- Earth is a magnet due to knowledge of magnet known at the time

Question 72

Describe the usefulness of analogies using the Gick and Holyak 1980 study

Answer 72

 Gick ad Holyak 1980- the ray problem vs fortress problem: • Ray problem- o Patient has stomach tumor that can be destroyed with high intensity ray but a ray of that intensity also destroys healthy tissues- how do you destroy the stomach tumor without destruction of the tissues? o Convergent solution- divide ray up into low intensity ray sources and make them all meet at stomach tumor at once • Fortress analogy- o Several roads lead to a fortress but on each road there are mines-but mines are set so that small numbers of men can pass over safely. Therefore, general divides army into smaller units and sends them down all of the roads so that they eventually meet at the same point unscathed • 10% of people got convergent solution without the analogy • 40% of people got convergent solution with analogy • 80% of people got convergent solution when hinted to use analogy • Therefore, can form analogies on basis of structure-lack of superficial similarities between story and problem could have resulted in people not spontaneously remembering it

Question 73

What are the 3 types of similarities?

Answer 73

o 3 types of similarities:  Superficial similarity- solution-irrelevant details are common to the two problems  Structural similarity-causal relations among some of the main components are shared by both problems  Procedural similarity-procedures for turning the solution principle into concrete operations are common to both problems

Question 74

Can analogies be transferred over long time periods? give proof

Answer 74

 Holyok and Koh (1987): some tutorials discussed ray problem in class • 3-7 days later presented with lightbulb problem in experiment o Solved by 81% (10% for controls) • Catrambone and Holyoak (1989) showed analogical transfer over 3 weeks o But had to give multiple analogies initially  Therefore, analogies can be transferred over long time periods

Question 75

What are aspects of knowledge and problem solving?

Answer 75

* Applying the right constraints | * Using knowledge-transfer

Question 76

Is there am absolute theory of problem solving? How/how not?

Answer 76

* Search is more a language for describing problem solving than a testable theory * Have theory about components of problem solving, but not overarching theory * Theories of particular types of problem solving (analogical) * Different processes may underlie phenomena called problem solving (which is a broad idea), so there may not be a single theory

Question 77

What is the fundamental problem with decision making and how do we deal with it?

Answer 77

• Problem in decision making is fundamentally uncertainty, but we have to make decisions o Tension decision making-cannot say with certainty what the right decision is, but are forced to make one o Assign probability to each outcome to deal with uncertainty-but lack of information about probability of option makes it hard to calculate expected value

Question 78

What is decision making?

Answer 78

• Decision making-choosing between options on the basis of the available information

Question 79

What is reasoning?

Answer 79

o Contrasts with reasoning, which derives new conclusions from available information

Question 80

Are decision making and reasoning linked?

Answer 80

o But reasoning is often involved in decision making

Question 81

How do we calculate expected values to options involved in decision making?

Answer 81

• Decision making requires judgement- what are the values of these options? o Need to calculate expected values of options and choose whichever option has the highest value o Calculate expected value by multiplying the value of each outcome by its probability  More likely outcomes weigh more heavily on expected value than less likely outcome

Question 82

What do we discount for when making decisions? Is this the same for everyone?Describe

Answer 82

• Tend to discount outcomes for uncertainty o People would take below expected value when uncertain o But If odds get low enough, can inflate for uncertainty rather than discount for uncertainty • Tend to discount for future o Rate of discounting the biggest for the near future o Rate of discounting decreases as look further into the future o Discount for future is a hyperbolic function o Overall, people discount the most the further into the future you get (not rate, just overall value) • Different people discount at different rates o Older adults discount for future less than kids or young adults o People who are more impulsive or from less stable backgrounds discount for future at a greater rate o Smokers discount future at higher rate than non-smokers o Men discount more after looking at pictures of hot women (Wilson and Daly, 2003) but hot men did not have the same effect on women-makes men more impulsive

Question 83

Do people like uncertainty? What do they like?

Answer 83

• Nontrivial decision-making involves uncertainty o Make a bet about the future- make judgement on what you think will happen in future o People don’t like uncertainty (especially for gains)-like guaranteed gains  Most people would rather like guaranteed outcome of lesser value than uncertain outcome of greater value

Question 84

Describe Green et al's 1999 study about discounting for uncertainty

Answer 84

o Green et al (1999) found people would take below expected value when uncertain  Gave people choices between guaranteed amount or gamble for a higher amount (with various differing odds)  Odds against winning vs subjective value graph, with line plotting expected value  But can change for every low odds  Discount for uncertainty when dealing with larger values more as dealing with bigger amounts o But If odds get low enough, can inflate for uncertainty rather than discount for uncertainty

Question 85

Are decisions more weighted when concrete or abstract?

Answer 85

• People are much more likely to make cautious decision when value is made concrete instead of abstract o Whether something is loss or gain is about how it is represented in mind

Question 86

Describe Green, Fry and Myerson's 1994 study about different people and discounting

Answer 86

o Green, Fry and Myerson (1994)  Kids (M=12.1years old) discount future more than young adults (M=20.3 years old) and then older adults (M=69.7 years) o Older adults discount for future less than kids or young adults

Question 87

Who developed the theory of expected utility?

Answer 87

* Idea goes back to Daniel Bernoulli (1738)- concerned with casinos and how people make bets * But developed systematically by von Neumann and Morgenstern (1944)

Question 88

What were the earliest studies on decision making and why?

Answer 88

• Earliest studies of decision making were by economists o Businesses are intensely interested in how people make choices between options o Economy- sum of people’s decisions

Question 89

What is the expected utility hypothesis?

Answer 89

• Assumed the expected utility hypothesis- o The utility of an agent (how pleasurable something is found) facing uncertainty is calculated by considering utility in each possible state and constructing a weighted average, where the weights are the agent’s estimate of the probability of each state o Then chose the option with highest expected utility

Question 90

How do you find expected utility?

Answer 90

o Expected utility=probability of a given outcome x utility of an outcome

Question 91

Describe the game deal or no deal (development ,rules and expected value)

Answer 91

• No skill involved in game show: only involved in decision making • Developed in 2002 in Netherlands o Shown in 48 countries, and for a long time in Australia, but not currently • Outline: o 26 boxes with 50c to $200,000 o But don’t know which box conceals which amount o Contestant puts one box on stage, never to be opened until the end o Then progressively chooses other boxes, revealing their contents (reducing uncertainty) o Periodically, bank offers to buy his/her box to stop the game for an offer  Offer depends on amounts revealed  Lower offer if boxes that have been revealed have large amounts and vice versa o Contestant then decides deal or no deal o Expected value=20,000 dollars approximately

Question 92

Describe how the quality of a decision is often measured vs how it should be measured

Answer 92

• We have an outcome bias (tend to look at good vs bad decision based on outcome of decision), but quality of decision depends on process of decision making when faced with uncertainty and limited resources o We engage in satisficing: Decision making operates under information, time and cognitive constraints o When looking at quality of decisions, need to consider the constraints someone is under • Research focuses on how people make choices between options, especially under certainty, not outcome

Question 93

Why is Deal or No Deal a good case study for decision making?

Answer 93

• Deal or no deal good stimulation- cannot raise the stakes too high in laboratories

Question 94

Describe Post et al's 2008 study on Deal or no deal episodes, and what he found

Answer 94

• Post et al (2008) studied 151 German, Dutch and US episodes o Bank offers usually below expected, but improve over rounds  Average accepted offers: 76% of expected value Dutch, 91% (German or US) of expected value • Dutch version has more net value  Losers and winners take more risks • People who have just experienced the large change (losers- large decrease, winners- large winners) o Inconsistent with expected utility theory  But seem to be influenced by history  Past should not affect choice in expected utility theory

Question 95

Are decisions uninfluenced by the past and future?

Answer 95

• Decisions are not in a vacuum- affected by previous outcomes and decisions that have occurred

Question 96

What are the problems with the expected utility theory?

Answer 96

• Theory often doesn’t fit to empirical data o Leads to various paradoxes  “Sunk cost” fallacy • Probabilities and utilities may be subjective, based on our own experience o Could change o Could represent individual beliefs  But there are biases in belief o Savage (1945) developed subjective expected utility theory • Can think of expected utility theory as a normative theory o Can set up normatively correct choice-an ideal o What people should do- people don’t necessarily always maximise their utility o Expected utility theory- each option evaluated independently in terms of utility  But options aren’t independent because can change framing of other options

Question 97

Describe the sunk-cost fallacy and why it is a problem

Answer 97

 “Sunk cost” fallacy • Tendency for people to pursue a course of action even after it has proved to be suboptimal, because resources have been invested in that course of action • Some resources put into something that we cannot get back-need to decide whether we keep investing or cut our losses • Should be influenced about future feasibility of product, not what was already put in (new situation not old situation) • But evidence that people are influenced by resources they’ve already put in-instead of evaluating new situation, need to keep going because already put so many resources into the product

Question 98

Describe Tversky and Kahneman's view on decision making

Answer 98

• Tversky and Kahneman (1974) suggested we have various heuristics and biases o Can sometimes lead to systematic errors

Question 99

What are the various biases in our judgements?

Answer 99

o Tend to be over-confident in success of decision o Tend to be loss aversive (tend to focus on losses)  Put greater value on concrete losses more than abstract losses  If represent as gains, more likely to take more risks o Framing of a problem is critical and can change choices  Loss vs gain representation  Concrete vs abstract representations-concrete representations makes decision more impactful and more likely to think in terms of loss  Framing effect- finding that decisions can be influenced by situational aspects irrelevant to good decision making o Influenced by information that may be dubious  Look for more information when don’t know what to do-grasp at information even if it is dubious

Question 100

Describe why we have biases in our judgements

Answer 100

• We have biases because they are adaptive- will move us towards better decisions

Question 101

What is the overconfidence bias?

Answer 101

o Don’t take into account real level of our knowledge-shows overconfidence in decisions

Question 102

Describe Kahneman's 2011 study on overconfidence

Answer 102

o Kahneman 2011  For a number of years CFOs of large corporations were asked to predict the S&P index over the next year (11,600 estimates)  No correlation between estimate and actual S&P index movement  Also asked to estimate a value they were 90% sure S&P would not be higher than, and 90% sure it would not be lower • But their estimates only fell in 90% intervals 33% of the time

Question 103

Is overconfidence needed? When is it the greatest?

Answer 103

o But overconfidence can be needed  Confidence in decision climbs as more information is obtained, even if information is dubious  However, a under-confidence bias may even be more problematic • May never make any decisions • But sometimes choice has to be made  Overconfidence can spur us to make a decision and start projects o Overconfidence bias is greater in more difficult tasks  Estimating our potential productivity

Question 104

What are heuristics?

Answer 104

o Heuristics are strategies that can be applied easily to wide variety of situations and often lead to reasonable decisions o Substitute answerable for unanswerable questions o Heuristic- strategies that ignore part of the information, with the goal of making decisions more quickly, frugally and/or accurately than more complex methods

Question 105

Do heuristics always work?

Answer 105

• But heuristics are not guaranteed to work o Provide plausible conjectures, but not irrefutable conclusions  Errors are informative

Question 106

What is the availability heuristic?

Answer 106

o Availability heuristic- frequencies of events can be estimated on the basis of how easy or hard it is subjectively to retrieve them from long-term memory o Judgement based on ease with which relevant instances be retrieved from memory o Tversky & Kahneman (1974) found that people are pretty good at doing this o Good assumption if we assume that memories are strengthened by repetitions o But can cause a few salient cases that are not representative of the field to bias decision and not give enough weight to overall data that is there

Question 107

How can the availability heuristic be overridden?

Answer 107

o Availability heuristic can be overridden by deliberate thought

Question 108

Describe Shwarz et al's 1991 study on the availability heuristic and his follow-up on this

Answer 108

• First asked to list 6 or 12(depending on assigned group) instances in which you behaved assertively • Then asked to evaluate how assertive you are • Those asked to retrieve 12 retrieved more than those asked for 6 but found it hard o Participant in 6 condition rated themselves more assertive o This is because people who were asked to generate 12 found it harder than those asked to generate 6, and therefore gave themselves a lower rating of assertiveness – what is critical is ease • In another study, then removed effect when participants told that background music would reduce fluency o Played background music during retrieval and told participants that background music would make it harder for them to retrieve cases o Hence eventually found that the effect of ease vs amount was removed when participants could attribute difficulties to background music, not to themselves

Question 109

Is the ease of retrieval or amount of case studies the most important in availability heuristic?

Answer 109

-Ease of retrieval  Ease of case retrieval is more effective on judgement than amount of cases • So ease of retrieval seems to be the main issue

Question 110

Describe Slovic, Fischhoff and Lichtenstein's 1979 study on availability heuristic

Answer 110

o Slovic, Fischhoff and Lichtenstein (1979)  Participants rated which of a pair of causes of death was more likely  Extravagant causes of death such as airplane crashes and cancer were rated more likely than causes that kill many more people • Most likely due to media coverage, so more ease of retrieval o Overestimate things that get more media coverage, not because they’re more common but because they’re more sensational  Has real consequences in predicting behaviour of people

Question 111

Describe the 3 ways of explaining people's judged probabilities of various causes of death

Answer 111

o Pachur et al (2012)-three ways of explaining people’s judged probabilities of various causes of death  People may use an availability heuristic based on their own direct experiences • This is the best predictor of judged frequencies of different causes of death  People may use an availability heuristic based on media coverage of causes of death as well as their own experience  May use affect heuristic-events of higher dread are more prevalent

Question 112

Describe Gigerenzer's 2004 study on the consequences of the availability heuristic and media coverage

Answer 112

o Gigerenzer (2004)  Analysed US road fatalities for Oct-Dec 2001  After events of 2001 (9/11), road fatalities increased as more people were driving  Estimates 350 extra people died due to perception of risk  Terrorism exploits availability

Question 113

What are the limitations of the heuristic approach to decision making?

Answer 113

• Limitations of this approach- o Heuristics identified are vaguely defined o Theorising based on heuristics and biases approach has been limited-imprecise conditions and relationships o Unfair to conclude people’s judgements are biased and error prone o Much research is detached from the realities of everyday life

Question 114

Describe Kahneman and Tversky's prospect theory

Answer 114

Prospect theory-Kahneman and Tversky (1979,1984) • Assumptions o Individuals identify a reference point generally representing their current state o Individuals are more sensitive to potential losses than to potential gains-loss aversion  Positive value associated with gains increases relatively slowly as gains become greater  Negative value associated with losses increases relatively rapidly as losses become greater • Graph is steeper for losses than for gains- losses are given greater value (loss aversion) • Inflection point between gain vs loss is critical but arbitrary • Implies that what is a loss and what is a gain can depend on perspective

Question 115

What is the representativeness heuristic?

Answer 115

* Representativeness heuristic- the assumption that an object or individual belongs to a specified category because it is representative (typical) of that category * Like availability, representativeness relies on basic cognitive process (similarity assessment)

Question 116

Are base rates used preferentially over the representativeness heuristic?

Answer 116

No-• People tend to use representativeness rather than base-rates, even if it is explicit

Question 117

Describe Tversky and Kahneman's 1974 experiment on representation heuristic

Answer 117

• Tversky and Kahneman (1974)-party situation with differing amounts of lawyers vs engineers o Describes Bob in a stereotypical way as an engineer, and group 1 is described as having a party with more lawyers than engineers, whilst it is the opposite with group 2 o People usually disregard base rate of lawyer: engineer ratio at party in favour for the representativeness heuristic  Most said Bob was an engineer with high probability, regardless of base-rate o If description is neutral (equally like engineer or lawyer), then there is 50-50 change of engineer or lawyer (which contradicts base rates)-people still try to apply representativeness heuristic

Question 118

What is base-rate information?

Answer 118

• Base-rate information- the relative frequency with which an event occurs within a population

Question 119

Is judgement performance presented better in frequencies or probabilities? What is a confounder to this conclusion?

Answer 119

• Judgement performance is often much better when problems are presented in frequencies rather than probabilities or percentages o However, the frequency version of a problem is generally not easier than the probability version when steps are taken to avoid underlying problem structure (Manktelow 2012) much more obvious in the former version- thus the benefits of the frequency formats probably do not occur because people are naturally equipped to think about frequencies rather than probabilities

Question 120

When do people use base-rate information?

Answer 120

• Use base-rate information when we possess relevant causal knowledge to solve a problem

Question 121

Describe Kaheman, Knetsh and Thaler's 1990 experiment on loss aversion

Answer 121

• Kaheman, Knetsch and Thaler (1990) had two groups: o Sellers are given decorated mug to keep and asked how much they are willing to sell it for o Choosers are asked how much they think the mug is worth o Afterwards, a market price will be set o Same situation for both (same mug) but perspectives differ-  Sellers lose their mug, but choosers gain a mug- gain vs loss changes pperspective on mug o Median prices-  Sellers=$7.12  Choosers=$3.12

Question 122

Do we place more values on losses or gains? Why?

Answer 122

• Place a higher value on what is ours, and giving what is ours up represents a loss o Same loss has greater impact than equivalent gain o Loss aversion • Bias may be adaptive because losses could threaten survival

Question 123

Describe Tversky and Kahneman's 1981 experiment on bias-loss aversion

Answer 123

• Bias-loss aversion- o Tversky and Kahneman (1981) found many people reject a 50-50 bet in which they can win $100 if it comes up heads but lose $100 if it comes up tails  Found that more people took bet but still a whole lot of people didn’t if proposed that you can win $200 it it comes up heads but lose $100 if it comes up tails o Weight prospect of losses more heavily

Question 124

Describe Odean's 1988 findings on investors

Answer 124

• Investors sell gains, hold onto their losses o Should be making decision on current situation of stock-but often remembered what paid for it which affects people’s decisions o Odean (1998) analysed over 1 million stock trades  Found investors had strong preference for stocks that were above the purchase price than those below the purchase price  Demonstrated that they were losing money due to this

Question 125

What is the framing effect and when can it be eliminated?

Answer 125

• Framing effect-the finding that decisions can be influenced by situational aspects irrelevant to good decision making o Framing effect can be eliminated when individuals think carefully about the available options

Question 126

Describe the difference in behaviour when something is framed as a gain vs a loss

Answer 126

• The way a problem is framed can change the way the options are evaluated o When something is framed as a loss, more likely to gamble to avoid that loss o When something is framed as a gain, more likely to take the sure gain

Question 127

Describe Shafir and Thaler's 2006 experiment on loss-gain representations

Answer 127

• Shafir and Thaler (2006) o Bought wine for $20, but wine now sells at auction for $75 and you drank your bottle. What best captures the feeling of the cost of drinking the bottle?  30% of people responded $0 (already paid for it)  18% responded $20 (what was paid for it)  7% responded $20 plus interest  20% responded $75- what could have gotten if sold the bottle • This is the opportunity cost- answer from economist point of view  25% responded -$55: save money by drinking the bottle o But if reframed question so that dropped bottle instead of drinking it, there is different pattern of responses-  8% of people responded $0 (what was already payed for it)  24% of people responded $20 (what was paid for it)  11% chose $20 plus interest  55% chose $75, what I could get if I sold that bottle • Greater value to loss  2% chose -$55 o Representation of problem changes, it now seems like a loss, so valued higher

Question 128

What is the impact of irrelevant information in a problem?

Answer 128

• Irrelevant information makes a difference by changing frame

Question 129

Describe Simonson and Tversky's 1992 experiment on framing questions and options

Answer 129

o Simonson and Tversky (1992) offered two possible payments for an experiment  Take $6 or a nice pen • 36% nice pen, 64% take cash  Take $6, nice pen or cheap nasty pen • When include cheap pen as options, shifted some people from taking the cash to taking the nice pen o Cheap pen changes framing of nice pen as there is now a comparison of the nice pen • 46% take nice pen, 52% take cash

Question 130

What is anchoring and adjustement?

Answer 130

Anchoring and adjustment is a psychological heuristic that influences the way people intuitively assess probabilities. According to this heuristic, people start with an implicitly suggested reference point (the "anchor") and make adjustments to it to reach their estimate

Question 131

Describe Tversky and Kahneman's 1974 experiment on anchoring and adjustment

Answer 131

o Tversky and Kahneman (1974) suggested our estimates can be influenced by an initial number  What percentage of the UN are African countries  First had them spin a wheel and asked if it was less or more than the number they just spun  Then told what real answer was  If wheel said 10, mean response was 25%  If wheel said 65, mean response was 45%  So even if knew anchor was random, having to think about that anchor had dragged final response  Arbitrary number affected responses

Question 132

What are implications of anchoring and adjustment

Answer 132

o Implications of anchors |  Anchors can influence outcome even when arbitrary, even for experts

Question 133

Describe Northcraft and Neale's 1987 study on the implications of anchoring and adjustment

Answer 133

 Northcraft and Neale’s (1987) study of real estate agents shown the same house • One group told the asking price was $65,900 o Average appraisal $66,755 • Other group told asking price was $83,900 o Average appraisal $73,000

Question 134

Describe Englich et al's 2006 study on the implications of anchoring and adjustment

Answer 134

 Englich et al (2006) gave dice role of 3 or 9 to experienced German judges who had to give sentence to criminals • Asked if they thought sentence should be more or less than that number • If roll 3, then mean sentence recommendation was 5 months, if roll 9 then mean was 8 months Adaptivity-

Question 135

Why do we use fast and frugal heuristics?

Answer 135

• Use fast and frugal heuristics to adapt to environment o One cue gives information for decision making, so spend no time looking for more information o Works because recognition is an ecologically valid cue o We have heuristics because they fit in the environment and help us make better decisions in a more efficient amount of time

Question 136

What is an example of a fast and frugal heuristic and a study done on it?

Answer 136

• Use fast and frugal heuristics to adapt to environment (Gigernzer and Todd 1999) o E.g. recognition heuristic- involves selecting the object that is recognised in preference to the one not recognised  If recognise one city but not the other, say the one we recognise is bigger • Most of the time, they’re right-> more likely to recognise bigger cities because we hear more about them

Question 137

Describe the 3 components of fast and frugal heuristics

Answer 137

o Has 3 components-  Search rule: search cues in order of validity  Stopping rule- stop after finding a discriminatory cue  Decision rule-choose outcome

Question 138

What is bounded rationality and when do we use it?

Answer 138

Bounded rationality is the idea that rationality is limited when individuals make decisions: by the tractability of the decision problem, the cognitive limitations of the mind, and the time available to make the decision. Decision-makers, in this view, act as satisficers, seeking a satisfactory solution rather than an optimal one.