Judgment: Categories PDF
Document Details
Uploaded by PrivilegedPun4107
McMaster University
Tags
Summary
This document provides lecture notes on judgment and reasoning, including topics such as inductive and deductive reasoning, framing effects, and decision-making. It explores cognitive biases and heuristics that can influence these processes.
Full Transcript
Judgment & Reasoning and Concepts & Categories Key Concepts/Ideas Judgement & Reasoning - Lecture Notes Judgement Idealizing reasoning: formal logic? ○ Inductive reasoning ○ Observations → general conclusion ○ Bottom up...
Judgment & Reasoning and Concepts & Categories Key Concepts/Ideas Judgement & Reasoning - Lecture Notes Judgement Idealizing reasoning: formal logic? ○ Inductive reasoning ○ Observations → general conclusion ○ Bottom up Observation → pattern→ tentative hypothesis → theory Ie. a bee, wasp and a fire ant are all hymenopteran and all sting THEREFORE ALL HYMENOPTERANS have stingers Deductive reasoning ○ General premise(s) → leads to conclusion or decision about specific case ○ Top down ○ Theory → hypothesis → observation → confirmation Ie. all wasps have stingers, this thing in my hand is a wasp THEREFORE THING CAN STING ME Framing Effect D ecisions can be influenced by how information is presented (framed). People tend to be: ○ Risk-averse when presented with potential gains (e.g., choosing a guaranteed smaller gain over a larger but risky gain) ○ Risk-seeking when presented with potential losses (e.g., choosing a risky option to avoid a certain loss, even if the risky option has a potentially larger loss) Framing of Outcomes I dentical choices can be presented in a positive or negative frame, leading to different decisions. ○ Example: Choosing between two medical programmes with identical outcomes but framed in terms of lives saved (positive) or lives lost (negative). Decision Making: Costs and Benefits P rinciple of Utility Maximisation: People aim to choose the option with the greatest expected value. Involves balancing costs and benefits. However, decisions are often influenced by factors beyond utility maximisation. Framing of Questions and Evidence How evidence is presented can influence our judgments. ○ Medical treatments: People are more likely to endorse a treatment with a "50% success rate" than a "50% failure rate." ○ Advertising and health communications often use framing to influence consumer choices. Emotion and Decision-Making motions play a significant role in decision-making. E Somatic Markers: Bodily sensations associated with emotions can influence choices. "Gut Feelings": May lead to choosing options that trigger positive feelings. Predicting Emotions (Affective Forecasting) Ability to predict future emotions. People are generally accurate in predicting whether their reaction to an event will be positive or negative. However, they often misjudge the duration of these feelings. Dual-Process Model Type 1 Thinking: ○ Fast, automatic, intuitive ○ Relies on heuristics (mental shortcuts) ○ Minimal cognitive effort ○ Useful for rapid assessments ○ Can be biased or inaccurate ○ Processing Speed: Instantaneous Type 2 Thinking: ○ Slower, effortful, analytical ○ More likely to be accurate and well-reasoned ○ Used when accuracy is important Examples of Type 1 Judgment (Heuristics) A vailability Heuristic:Judging frequency or likelihood based on how easily examples come to mind. For judgement, need to be able to judge FREQUENCIES of events (probabilities) MEMORY is crucial ○ Do more words in English start with R or K, or are there more that have R or K as the third letter? ○ Do more words in English end with the pattern??? _n_ or with -ing ??? ○ Kahneman and Tversky People use AVAILABILITY to judge FREQUENCY ○ Heuristics are GOOD, except when they’re NOT Lottery tickets - nuclear power - Homeless are mentally ill - genetically modified foods - Stereotypes - horse meat in food Terrorism - vaccination- Car vs plane crash R epresentativeness Heuristic: Judging based on resemblance to a typical example, neglecting actual probabilities. ○ Assume homogeneity (that all members of a category are the same) ○ Assume that each member of a category is representative of that category Category homogeneity If you've seen only a few examples of a category, assume that all of the category members are like that Stereotypes Notes that representativeness isn't completely invalid reasoning strategy, just that it will lead to some mistakes Conjunction fallacy:Ppl use similarityto a prototypical example, rather than probability, as a basis for judgement Linda is 31 yrs old, single, outspoken, and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice, andalso 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 How likely is it that sometime in the next year, a massive fire in North America will kill more than 1000 people? How likely is it that sometime in the next year, there will be an earthquake in california causing a massive fire in which more than 1000 people will be killed? AKA: support theory Gambler's Fallacy: Belief that past events influence the probability of future events in random sequences. ○ Reasoning from the population to an instance The belief that prior outcomes can influence the outcomes of probabilistic events Similarity, the belief that you shouldn’t fly on the airline with the best safety record because it is “due for a crash” ○ Law of large numbers – things do tend to even out in the end, proportionally, over a LOT of trials. BUT, this does not extend to small samples ○ “In a small town nearby, there are two hospitals.: (Kahneman & Tversky, 1972) Hospital A has an average of 45 births per day; Hospital B has an average of 15 per day Overall, the ratio of females to males born is 50/50 Each hospital recorded the number of days in which, on that day, at least 60% of the babies born were male Which hospital most likely recorded more such days? Hospital A, hospital B, or equal? Reasoning from a single case to an entire population○ AKA “the man who” argument ○ “My brother in law eats nothing but pizza and beer and he’s healthy” Smoking is not that bad, I have an uncle who smoked two packs a day all his life and he died in his sleep at 85” But the whole category (smokers) doesn’t have to look like this one example -assumption of representativeness (or homogeneity) Q uickly deciding to cross a street when no cars are visible Examples of Type 2 Judgment valuating Evidence: Reviewing multiple sources before reaching a conclusion. E Considering Base Rates: Using statistical probabilities rather than stereotypes. Cost-Benefit Analysis: Weighing pros and cons of different options. Carefully analyzing investment options before making a financial decision Base Rate and Base-Rate Neglect B ase-Rate Information: How frequently something occurs in a general population. Diagnostic Information: Information about a specific case. Base-Rate Neglect: Ignoring base-rate information in favour of diagnostic details. ○ Often due to the representativeness heuristic. ○ Can lead to biased judgments. The taxicab problem ○ A cab was involved in a hit and run accident at night. Two cab companies, the green and the blue,operate in the city. You are given the following data: ○ 85% of the city cabs are green and 15% are blue ○ A witness identified the cab as a blue cab. His vision was tested (appropriate visibility conditions). Presented with a sample (half blue, half green cabs) 80% correct identifications 20% errors ○ What is the probability that the cab involved in the accident was blue rather than green? ○ IT'S NOT 80% actors Influencing Type 1 vs. F Type 2 Thinking T ime Pressure: Type 1 thinking is more likely under time constraints. istraction: Type 1 thinking is more likely when distracted. D Effort Required: Type 2 thinking is used when effort is possible. Focus: Type 2 thinking requires focused attention. Information Format: ○ Base-rate neglect is less likely with frequencies (e.g., "12 out of 1,000"). ○ Base-rate neglect is more likely with probabilities or proportions (e.g., "1.2%"). Education and Reasoning E ducation can increase the use of Type 2 thinking. Training can improve understanding of statistical reasoning and the importance of large sample sizes. Cognitive Reflection Test (CRT): Assesses the tendency to override intuitive (Type 1) responses with analytical (Type 2) thinking. Cognitive Reflection Test (CRT) CRT Problem Analysis ○ Ball and Bat Problem Intuitive (Incorrect) Answer: 10 cents Analytical (Correct) Solution: Ball = 5 cents Bat = $1.05 ○ Machine Productivity Puzzle Intuitive Response: Complex calculation Correct Answer: 5 minutes (constant time) ○ Lily Pad Growth Problem Intuitive Solution: 24 days Analytical Solution: 47 days Confirmation Bias T endency to favour information that confirms existing beliefs. Five Forms of Confirmation Bias ○ Selective Evidence Seeking Preferentially searching for confirming information Deliberately ignoring contradictory evidence ○ Belief Rigidity Failing to adjust beliefs when contradictory evidence emerges Maintaining original hypothesis despite new information ○ Evidence Interpretation Minimizing impact of disconfirming evidence Rationalizing contradictory observations ○ Selective Memory Better recall of confirming information Distorted memory of disconfirming experiences ○ Hypothesis Limitation Neglecting alternative explanations Failing to consider multiple variables Represents a failure of logic, but is very common. Reasoning about Syllogisms Categorical Syllogisms: Logical arguments with two premises and a conclusion. V alidity: Whether the conclusion logically follows from the premises. Evaluating Syllogisms: ○ Consider concrete examples to test the logic. ○ Be aware of common reasoning errors. Active Recall Questions 1. What are the two primary psychological responses to potential gains and losses? 2. How does the presentation of information impact decision-making? 3. Describe the principle of utility maximization in decision-making. 4. What are the key characteristics of Type 1 thinking? 5. How does Type 2 thinking differ from Type 1 thinking? 6. What factors influence the shift between Type 1 and Type 2 thinking? 7. What is the availability heuristic? 8. Explain the representativeness heuristic with an example. 9. What is the gambler's fallacy? 10.What makes CRT problems unique in testing cognitive processing? 11.Solve the ball and bat problem: How can you determine the correct price? 12.Explain the lily pad growth problem's counterintuitive solution 13.List the five forms of confirmation bias. 14.How does confirmation bias impact rational decision-making? 15.Provide an example of selective evidence seeking 16.Think about everyday examples of framing effects. How does the way information is presented influence your choices? 17.Consider a decision you recently made. Did you rely more on Type 1 or Type 2 thinking? What factors influenced your thinking style? Concepts and Categories - Lecture Notes Understanding Concepts and Cat Concepts allow us to: ○ Apply general knowledge to new situations. ○ Draw broad conclusions from experiences. Category ○ The set of entities or examples described by the concept Functions of concepts 1. Classification 2. Understanding 3. Prediction 4. Reasoning 5. Communication The classical view: ○ Concepts have defining features or attributes Easy to categorize! But hard to come up with a good definition ○ Hard to explain HOW you are categorizing ○ Not that we can sort this way – there are whole research programs based on taxonomy. We could even revert to DNA testing. But this is not we do in day today categorization We have a concept of furniture, even without a perfect definition We have a concept of what a dog or a bird is, without having to undertake DNA testing We know what a bachelor is, even without an elaborate definition Most of our knowledge (as we use it day to day) isn’t based on definitions Problem with Definitional Approach D efining features often fail to capture the variability within categories. Exceptions to definitions are always possible. Example: "Tables are flat surfaces with four legs" doesn't apply to all tables. Family Resemblance F ocuses on overlapping features (characteristic features) among category members. There are no defining features that all members must possess. Example: Members of the Smith family may typically have dark hair and wear glasses, but not all do. Wittgenstein: category members have a family resemblance to each other ○ Some features in common, but not every member has to have those features ○ Different members have different features in common This leads to a PROBABILISTIC definition ○ If you have these features, then you have X likelihood of belonging to category Y Probabilistic view ○ No necessary conditions for belonging to a category, no sufficient definition either ○ Does not mean that categories have no structure Prototype Theory Prototype: The central, most typical member of a category. C ategory membership is judged based on similarity to the prototype. Typicality: The degree to which an object resembles the prototype. Graded Membership: Objects closer to the prototype are "better" members of the category. Prototype theory captures the resemblance idea Typicality ○ The ‘centre’ of a category ○ The ‘ ideal’ or ‘average’ for the category ○ How similar or dissimilar things are to that Graded membership ○ There are very ‘doggy’ dogs and dogs that aren't very much like that ○ We do this with all things ○ Ie. birds Robin most similar, then eagles, then penguin least similar to the prototype Prototypicality is based on what is common to individuals Is the prototype the thing with the most in common with the category? Is it the average of all the things in the category? ○ Not necessarily the “average” - often the best one Testing Prototype Theory Sentence Verification Task: ○ Participants judge whether sentences like "Robins are birds" or "Penguins are birds" are true or false. ○ Responses are faster for objects closer to the prototype (e.g., robins). Production Task: ○ Participants name as many members of a category as possible (e.g., fruits, birds). ○ More typical members are usually named first. Rating Tasks: ○ Participants rate objects on how typical they are of a category. ○ Objects closer to the prototype receive higher typicality ratings. Example: Fruit and Bird Typicality Ratings (Refer to the provided table in source for specific examples) Test prototype theory with typicality ratings ○ Eg. rate the following for a certain category.. 1= not typical at all 7= highly typical Test with sentence verification tasks ○ Verify for truth ○ Higher grades/ more prototypical questions produce faster responses Is a robin a bird vs an emu Test with a production task ○ Come up w as many examples as possible ○ The things that come up most are more prototypical Test with picture verification ○ Show picture and ask if it belongs to a category Testing depends on where you grow up...many cultural differences ○ Ie. fruit is diff in diff places ○ A lot of it is identifying so if you don't know what it is bc not as common response will be slower/inaccurate Test by generating simple sentences about a certain category Then ask which substitute is rejected as implausible or silly Ie. the bird vs. penguin in the tree Basic-Level Categories I ntermediate level of categorization. Neither too general (superordinate) nor too specific (subordinate). Often single words. Examples: "Chair" is a basic-level category, while "Furniture" is superordinate and "Wooden desk chair" is subordinate. Vertical hierarchy of hierarchy Superordinate level ○ Vehicle ○ fruit Basic level ○ Car ○ apple Subordinate level ○ Audi ○ Pink gala However, it really does matter how much you know about the topic ○ Otherwise you would change what counts as your basic level category Can change with new experiences Exemplar-Based Reasoning A lternative to prototype theory. Draws on knowledge of specific category members (exemplars) rather than a general prototype. Exemplars: Specific remembered instances of a category. Exemplars vs. Prototypes Prototypes: ○ Provide an economical summary of a category. ○ Good for comparison. Exemplars: ○ Provide information about category variability. ○ Easier to adjust categories based on new exemplars. ombination of Exemplars and C Prototypes C onceptual knowledge is thought to involve both exemplars and prototypes. Early learning often involves exemplars. E xperience leads to averaging exemplars to form prototypes. Both are used for categorisation and object recognition. Judgement of Category Membership ypicality influences category judgments but is not always decisive. T Essential Features: Depend on beliefs about the category. Atypical features do not necessarily exclude category membership. Example: An "abused lemon" (painted, sugared, run over) is still a lemon. Judgement of resemblance is influenced by knowledge and beliefs. Example: A counterfeit bill resembles real money but is not considered currency. Inferences and Categorisation Categorisation allows us to: ○ Apply general knowledge to new cases. ○ Draw broad conclusions from prior experiences. Inferences can be guided by: ○ Typicality ○ Theories and broader beliefs Attractiveness can influence typicality and categorisation. (This point was not explicitly mentioned in the slides but was discussed in the lecture) Reasoning about Natural Kinds vs. Artifacts N atural Kinds: Objects that exist naturally (e.g., skunks, raccoons). Artifacts: Man-made objects (e.g., toasters, coffeepots). People reason about them differently: ○ Natural kinds are seen as having more stable properties. ○ Artifacts are seen as more easily changeable. Example: Children believe a skunk cannot be turned into a raccoon but a toaster can be turned into a coffee pot. Concepts and the Brain Different brain regions are activated when thinking about: ○ Natural kinds vs. artifacts ○ Living vs. nonliving things Recognition of living things may rely more on perceptual properties. Recognition of non living things may rely more on functional properties. Embodied Cognition P roposes that concepts include representations of perceptual properties and motor sequences. Sensory and motor areas are active when thinking about certain concepts. Example: Thinking about "kick" activates areas that control leg movement. Travelling through the Network to Retrieve Knowledge C oncepts are interconnected in memory networks. Sentence verification tasks demonstrate that retrieving information requires traversing links in these networks. Responses are faster when fewer links need to be traversed. Example: Verifying "A canary is a canary" is faster than verifying "A canary can fly." Propositional Networks ropositions: The smallest units of knowledge that can be true or false. P Nodes represent concepts, and links represent relationships between concepts. Propositional knowledge is stored as links between concepts. Example: The proposition "Drake is a rapper from Toronto" can be represented as interconnected nodes for "Drake," "Rapper," and "Toronto." Hub-and-Spoke Model C entral "hub" integrates different types of knowledge from "spokes" distributed in specialized brain regions (e.g., visual or motor areas). Active Recall Questions 1. hat are the five primary functions of concepts? W 2. Explain how concepts allow us to apply general knowledge to new situations. 3. What is the classical view of concept definition, and why is it problematic? 4. Describe Wittgenstein's concept of family resemblance in categorization. 5. How does the probabilistic view of categorization differ from classical definition approaches? 6. Define a prototype in the context of category formation. 7. What is typicality, and how does it relate to category membership? 8. Explain the concept of graded membership in prototype theory. 9. Describe the sentence verification task and its purpose in studying prototype theory. 10.How does the production task help researchers understand prototype formation? 11.What defines a basic-level category? 12.Provide an example of a vertical hierarchy of categorization from superordinate to subordinate levels. 13.How does exemplar-based reasoning differ from prototype theory? 14.Explain how early learning involves exemplar-based reasoning. 15.How do typicality and broader beliefs influence categorization? 16.Provide an example of how atypical features do not necessarily exclude category membership. 17.Explain the difference in how people reason about natural kinds versus artifacts. 18.Describe a child's perspective on transforming a natural kind versus an artifact. 19.How do brain regions differ when processing natural kinds versus artifacts? 20.Explain the concept of embodied cognition with a specific example. 21.Define a proposition in the context of cognitive processing? 22.How do nodes and links work in a propositional network?
