Thinking and Decision Making

Thinking and decision making Mental representations  Analogical representation : qn idea that shares some of the actual characteristics of the object it represents  Symbolic representation : a mental representation that stands for some content without sharing any characteristics with the thing it represents Analogical representations Method 1. Participants were shown a map of a fictitious island containing various landmarks. 2. After memorizing this map the participants were asked to form a mental image of the island. 3. Participants were timed while they imagined a black speck zipping from one landmark on the island to another. When the speck “reached” the target, the participant pressed a button, stopping a clock. Results The time needed for the speck to “travel” between two points on the mental image was proportional to the distance between those points on the map. CONCLUSION: Mental images accurately represent the spatial relationships inside a scene. Mental images Brain scan of what happens in the brain when you see vs when you imagine a tree First colum is when you see or ‘perceive’ a tree Second column when you imagine Third is the difference The most activation is happening in the back of the head Biggest difference is in the back of the head Symbolic representations - Do not resemble the items they stand for Symbols are more flexible, we can link any idea to them Example all “cat” Propositions  A statement relating a subject and a claim about that subject.  Semantic association networks, useful for problem solving  Network structures o nodes – connecting individual symbols o associative links – connection between nodes The spread of activation - are both words real? faster decision when words are semantically related Processes of thought Direct thinking  thinking aimed at a particular goal  process of our thoughts depends on what we are trying to achieve o o o o reasoning judgement decision making problem solving Reasoning Deductive reasoning  deduction: dervice new assertions from assertions already in place  provides a means of testing beliefs example  premise 1: skl closes when at least 4 inches of snow falls  premise 2: it snowed 6 inches last night  product of reasoning (conclusion): no skl today example  Premise 1: You believe that you will succeed in the exam without reading the book  Premise 2: You sit the exam  Product of reasoning (Conclusion): If you pass, you were right (chances are you were wrong)     Syllogism: conclusion follows two premises The conclusion Is valid if it logically follow the premises Fpr the conclusion to be valid, it doesn’t matter whether it is plausible or not What matters is the FORM, not the CONTENT Example  Premise 1: All artwork is made of wood  Premise 2: All wooden things can be turned into a clock  Conclusion: Therefore, all artwork can be turned into a clock  False premises, but valid syllogism (i.e., a true conclusion) Confirmation bias  tendency to take evidence that is consistent with our beliefs more seriously than evidence that is inconsistent  your own bias toward the evidence toward the results that you want  tend to disregard or devalue evidence against ur belief example  capital punishment as a deterrent to crime (Lord, Ross & Lepper, 1979)  presentation of evidence o pro death penalty evidence o against death penalty evidence  Gambler’s fallacy o Winning strategy o Focus on past wins and past loses are reinterpreted o Near wins – “The team I bet on would have won if it wasn’t for the ref’s bad decision” o Chance events – “It was just bad luck” Faulty logic & confirmation bias  Wason’s Selection Task  If a card has a vowel on one side, it must have an even number on the other side. Which card do you flip to verify this?  46% of people choose the “4” card. Does this test “If vowel, then even”? No! More of a test of “If even, then vowel”  Attempt to confirm the rule. Not a valid test  Correct: “7” card. “If vowel, then even”  “not even, not vowel”. Yes! Attempt to falsify the rule. Valid test.  Only 4% of people choose “7”. Failure to choose “7” shows faulty reasoning and confirmation bias  Context matters! o If a person is drinking a beer, then the person must be over 19 years of age. Which card do you flip to verify this? People perform much better at this one! Why Does Context Matter?      Cheng and Holyoak (1986) Permission schemas help us reason about the situation Daily life: if-then situations related to the permission Well-honed reasoning skills and strategies that have been practiced “If under 19 then no beer” is a well-rehearsed schema and leads to selection of the correct card Judgement Induction  Going from one or more specific cases to a more general conclusion  Example:  Fact: Jane has enjoyed many trips to the beach  Conclusion: Jane will always enjoy trips to the beach How Do We Determine the Strength of These Conclusions?     Number of observations The more, the better Representativeness Is the current situation similar to past situations encountered? (any difference from past beach experiences?) Examples of Inductive Reasoning  Science o Infer theories from experimental results  Daily life o e.g., I’ve had good experiences selling things on ebay in the past  I’ll try it again to sell my dinning table Heuristics     A strategy for making judgements quickly, at the price of occasional mistakes Term taken from computer science A shortcut method for getting a result NOT guaranteed to be correct, but usually correct (or pretty good) and quick/efficient (trade-off)  Daniel Kahneman & Amos Tversky the Availability Heuristic  Conclusions based on patterns of observations, not a single observation  To get an estimate of frequency, people think about specific cases o Availability as the basis for assessing frequency  Problem? It often works, but not always  “Considering all the words in the language, does R occur more frequently in the first position of the word (e.g., rose, robot) or the third letter of the word (e.g., carve, strive)?” Why Do People Do This?  Categories words by their starting letter, thus its easier to think of words that begin with “R” o Words beginning with “R” are more available o Use mental availability as a proxy for the actual frequency in the world The Representativeness Heuristic  Strategy to judge whether an individual, object or event belongs to a certain category based on how typical of the category it seems  Uniformity within categories plays an important role o Allows us to extrapolate from our experiences, so we know what to expect next time  Fundamental to detecting irregularities, but over used. People assume that each case is representative of its class.  people watching an interview with a prison guard with extreme views will draw conclusions about the entire prison system based on this one interview -- even if they were told in advance that his views were extreme and unusual, and that most guards were quite different.  The prison guard was believed to be representative of all guards, and thus the prior information indicating that his views were really quite rare was disregarded. Dual-Process Theory  The use of heuristics (shortcuts) is not inevitable  sometimes we reject the conclusion drawn from the heuristic  System 1: Fast, automatic type of thinking  System 2: Slower, more effortful type of thinking  Some form of education make System 2 thinking more likely Decision-making Utility Theory  Each decision has costs and benefits based on our personal goals. Costs move us farther from our own goals; benefits move us closer to our goals.  We decide by finding the course of action with the most favourable balance of benefits and costs  We estimate a subjective utility for each item: the value of the course of action for a given person Example  Trip to the USA o Benefit of trip to USA = + 200 utility points o Cost of trip to USA = -50 utility points o Utility of USA trip = 150 points  Trip to South America o Benefit of trip to South America = +200 utility points o Cost of South America trip = -100 utility points o Utility of South America trip = 100 points  Decision: Go to USA this year Examples Utility Theory & Uncertainty  Sometimes there is a risk involved in making a decision  Example 1:  I’ve never been to USA. Perhaps I won’t like it. Based on the reviews I have received, 80% have been positive  Expected Utility (EU) includes information about uncertainty: EU = 0.8 * 200 utility points (from benefits) = 160 expected utility points  Example 2:  Buy a £200 lottery ticket with 1/100 chance of winning. What is the expected utility? Predictions - Utility Theory  People are always seeking to maximise their utility in their choices  People seek options with the greatest expected utility Problems – utility theory  People do not do utility calculations very often!  Our decisions are often influenced by things which seem irrelevant, and potentially should not affect the outcome, if the outcome is maximisation. Framing effect  Decisions are influenced by the way question are phrased and options are described Example  USA is preparing for an outbreak of an unusual disease. It is expected to kill 600 people. There are two proposals to combat the disease. Scientists estimate the following consequences of each  Programme A: 200 people will be saved  Programme B: 1/3 probability all 600 will be saved, 2/3 probability that no one will be saved  Which one will you adopt? Or  The USA is preparing for the outbreak of a unusual disease. It is expected to kill 600 people. There are two proposals to combat the disease/ Scientists estimate the following as the consequences of each. o Programme C: 400 people will die o Programme D: 2/3 probability all 600 will die. 1/3 probability that no one will die.  Which one will you adopt?  The only difference between the problems is in how the alternatives are phrased  Programme A (200 will be saved) = Programme C (400 will die)  Programme B (1/3 prob of 600 saved, 2/3 prob 0 saved ) = Programme D (1/3 prob of 0 deaths, 2/3 prob of 600 deaths) Loss frame - Loss aversion Avoid sure losses Gain frame - Risk aversion Avoid risk when dealing with gain Tversky and Kahneman (1987) Affective forecasting  Can we forecast our feelings?  Not very good at predicting our future feelings o We can say “sad” or “happy” but not how strongly or for how long  We tend to ignore important but less salient factors (e.g. support of friends, meeting other people…)  For the utility approach to work, all of these factors must be considered in estimating utility Decision-making: other effects  Is there such a thing as too many options? Yes.  We prefer to have many choices (we pick more numerous sets over less numerous sets)  However, we are more likely to make a choice when we have less options  The grocery store experiment (Iyengar & Lepper, 2000)  People showed more satisfaction afterwards when they were given 6 options, as compared to 24 or 30 options. Choice evaluation  What makes us happy is the way we made our choice, rather than the actual outcome  If our choice is easy to justify, we are happy  We can’t pick the best possible outcome (evaluation of all the option would be too time consuming!): we pick the option that is good enough for us Problem-solving  Well-defined problem: o Clear goal o Knowledge of the available options  Ill-defined problem: o Hazy sense of the goal o Many things are unknown Obstacles  Automaticity : Ability to do a task without paying attention to it. Useful, but might be problematic!  e.g., Stroop Task  Mental Set  Specific perspective taken in approaching the problem  It creates a “box” making it difficult to think outside of a mental set 