PSYC 253 Midterm Review PDF

Preview

Summary

PSYC 253 midterm exam review (Prof Royzman)...

Description

PSYC 253 Midterm Review Lecture 1: Humans are bi-mental: System 1 vs System 2 (slow and fast thinking/processing) - System 1: reactionary, unconscious o Quick, effortless, automatic - Can automate number of complex tasks without compromising performance (e.g. driving car and paying attention to number of things) - Quick vs slow - Effortless vs effortful - Implicit vs explicit - Automatic vs controlled - What do cows drink – first thought may be milk due to intuitive thinking (system 1) Normative vs descriptive: - Normative: how things ought to be (systematic ideas/set of principles) o E.g. law – tells you how to behave; mathematics – tells you how to think - Descriptive: how things actually are - Where do normative and descriptive claims/models come from? Two types of research design: - Non-experimental (observational) - Experimental Operational definition: - Defining a construct (such as violence) in a way that may be precisely, inter-subjectively measured or instantiated Why psychologists make an effort to be unobtrusive: - Reactivity: a tendency for organisms who are objects of observation to alter their behavior by virtue of being observed (Concealment is one approach to solving issue) Correlation coefficient: - A measure of linear association between 2 variables; can range from -1.00 to 1.00 - The closer the value is to -1.00 or 1.00, the stronger the relation between the 2 variables True experiment: - One (independent) variable is manipulated - Another (dependent) variable gauges the effects of this manipulation - Participants are randomly assigned to different “levels” of the independent variable Lecture 2: Remote Associates Test (RAT, Mednick): Association (priming) - Measures creativity Heuristic: - Quick (and often sub-conscious) rule of thumb that allows us to simplify and streamline an otherwise complex cognitive task, saving time and energy in the process Recognition heuristic: - If one of two items is recognized and the other is not, infer that the recognized item has higher value with respect to some dimension of interest (e.g. more famous person) Heuristics used to manipulate us and sell stuff:

- Scarcity heuristic (“only three left in stock”) - Value heuristic (“expensive = good”) - Social reference heuristic (“everyone is buying”) Attribute substitution: - When faced with task of assessing a hard-to-access attribute, people often prefer to assess a related easy-to-access attribute and use the outcome of that assessment as a proxy for evaluating the hard-to-access attribute, without being aware of the switch The paper folding problem - Folding paper in half over and over again 100 times  how thick will paper be? - Answer: 2400 trillion x distance between earth and the sun! At what age is a woman most at risk for breast cancer? - Answer: in her 80s; risk increases with age (people use heuristics to reach wrong conclusion) Anchoring and adjustment - Tendency to rely too heavily on, and fail to sufficiently adjust for, the initial impression or piece of information (the “anchor”) when making a judgment - Anchoring and adjustment seen in (mock juror) awards Availability: - Tendency to judge the frequency/probability of an event by ease with which an instance of the event comes to mind - E.g. being scared of plane crashes because of frequency on news Media exposure effect:  increases availability heuristic effect - Availability + media sensationalism - To the degree that people judge frequency based on availability (ease with which it comes to mind) - To the degree that media exposure of an event makes the event more “available”, - To the degree that level of media exposure is governed by sensationalism (rare and dramatic events get more exposure) … People may end up with a distorted frequency representation of world (and a warped idea of what they should and shouldn’t be afraid of as they go about their daily lives) Media exposure effect may explain why: - People fear crime out of proportion to its actual occurrence - People fear the wrong type of crime - People’s perception of crime varies more with their self-perceived media-manufactured victim type than their actual victim type Availability and teacher evaluations - Research showing that experienced difficulty of recall can influence evaluative judgments – extended to a study of university students rating a course - Students completed a mid-course evaluation form where they were asked to list either 2 ways course could be improved (relatively easy) or 10 ways in which the course could be improved (relatively difficult task) - Respondents who had been asked for 10 critical comments subsequently rated course more favorably than respondents who had been asked for 2 critical comments Death from shark attack versus noodle salad:

- More likely to die from noodle salad (allergy, choking, food poisoning) than shark What’s Janet more likely to be? - Answer: Janet is more likely to be a bank teller than a feminist bank teller (people assume feminist bank teller, just because of schema/representativeness) - An event can never be less likely than the subset of the event – otherwise committing conjunction fallacy Representativeness: - Tendency to judge category membership based on similarity, i.e. based on how good a match the target description is to a category-related schema Lecture 3: Logic: the oldest normative model of all - Logic = systematic study of valid inference such that conclusions follow necessarily (with absolute certainty) from the premises that are themselves presumed to be necessarily true - Makes no room for degrees of belief: things are either valid or invalid - Human beings are mortal (major premise)  Matt Damon is human (minor premise)  Matt Damon is mortal (conclusion) Why are (some) categorical syllogisms so difficult? - Building a Euler circle model of a categorical syllogism:

mammals

cows

spotted cows

-

All A are B All B are C

What, if anything, follows from: - No A are B - All B are C

C

B

A

No A are B, all B are C:

C

A

B

Or:

A C

B

Or: A

C

B

Johnson-Laird et al.’s mental models theory: - Johnson-Laird: when presented with a syllogistic construction, people: a) Attempt to come up with a mental model/models corresponding to the premises b) Don’t do thorough enough a search o Draw an incomplete “list” of mental models o STOP RIGHT THERE o Fail to consider models inconsistent with their initial model Evidence for Johnson-Laird’s theory: - Subjects made most errors when syllogisms require multiple alternative (mutually inconsistent) models - “Good reasoners” (high scores on test of syllogistic reasoning) took more time to (correctly) adjust their answers when under no time pressure - Think-aloud reports indicate that those best able to correct for their initial response were more self-critical/looking for models inconsistent with their initial assumptions - Good at syllogistic reasoning = good at fully disjunctive reasoning - Good at syllogistic reasoning = higher CRT scores Fully disjunctive reasoning: - Jen, who is married, is looking at Greg - Greg, who could be either married or unmarried, is looking at Pauline, who is unmarried - In this situation, MUST there be one married person looking at one unmarried person? Cognitive Reflection Test (CRT): - Bat and ball cost $1.10 in total. The bat costs $1 more than the ball. How much does the ball cost? - Patch of lily pads. Every day, patch doubles in size. If it takes 48 days for patch to cover entire lake, how long would it take for patch to cover half of the lake? - CRT-Reflection: o Joan’s father has five daughters: Lala, Lele, Lili, Lolo, and ___?  Answer: Joan - There seems to be a quality of the mind in Western world in which syllogistic reasoning is rooted where you must inhibit the immediate response to a question and reflect/have cognitive control to find the correct answer

Reflection/Cognitive Control:

-

Alcohol intoxication facilitates creative problem solving Time of day affects problem solving abilities There is another part of cognitive/clear thinking which is not only underrated but complete opposite of this (the reason being is the 99 problem) Lecture 4: Logic and rationality Lifting self-imposed constraints:

Gino and Ariely (2012): - High on divergent thinking = more likely to cheat on a variety of tasks - Dispositional creativity was a better predictor of unethical behavior than other traits (intelligence) - Individuals primed to think creatively were more likely to cheat The beheading debate: - When she got back to the Cheshire Cat, she was surprised to find quite a large crowd collected round it: there was a dispute going on between the executioner, the King, and the Queen, who were all talking at once, while all the rest were quite silent, and looked very uncomfortable. The moment Alice appeared, she was appealed to by all three to settle the question, and they repeated their arguments to her, as they all spoke at once, so she found it very hard to make out exactly what they said.

-

The executioner's argument was, that you couldn't cut off a head unless there was a body to cut it off from: that he had never had to do such a thing before, and he wasn't going to begin at this time of life. - The King's argument was, that anything that had a head could be beheaded, and that you weren't to talk nonsense. - The Queen's argument was, that if something wasn't done about it in less than no time she'd have everybody executed, all round. (It was this last remark that had made the whole party look so grave and anxious.) Executioner’s argument: - In order to be a candidate for beheading, there must be a body and head to be severed - Cat lacks the body - Therefore, he cannot be beheaded Propositional logic: conditional syllogisms - If A (to be beheaded), then B (must have a body) - Not B (no body) - Therefore, not A (not to be beheaded) Modus Tollens - If A, then B - A - Therefore, B Modus Ponens - If A, then B - B - Therefore, A Affirming the consequent Watson’s 4 card problem: - Every card with a letter on it has a number on the other side, and every card with a number on it has a letter on the other side - Thinking of a rule: if there’s a K on one side, there must be a 4 on the other side - Which cards must you turn over to find out if rule is true or false? Rationality: - Rational thinking: relies on methods that are generally best in achieving the thinker’s goals, taking into account the practical limitations of decision-making  Paradox of rationality 'Tis not contrary to reason to prefer the destruction of the whole world to the scratching of my finger. 'Tis not contrary to reason for me to choose my total ruin, to prevent the least uneasiness of… [a] person wholly unknown to me. --Hume, Treatise of Human Nature, Book II, Of the passions, Sect. III, 'Of the influencing motives of the will' The cannibal of Rothenberg: is he irrational?

Lecture 5: Probability: - A numerical measure of the strength of a belief in a certain proposition - Ranges from 0-1 o 0 = certainly false o 1 = certainly true o 0.5 = equally likely, e.g. rain or not rain - 0.75 = 75% = ¾ = three to one odds o P/(1-P) = 0.75/0.25 = 3/1 Frequency: - Coin flips: HHHHHTHHTT o According to frequency theory, probability that the next coin is heads is 70% Pierre-Simon Laplace (1749-1827): originator of classical/logical account of probability - Theory of chance consists in reducing all events of same kind to a certain number of cases equally possible; that is to say, to such as we may be equally undecided about in regard to their existence, and in determining the number of cases favorable to event whose probability is sought. The ratio of this number to that of all the cases possible is the measure of this probability, which is thus simply a fraction whose numerator is the number of favorable cases and whose denominator is the number of all cases possible - P (a coin rises heads) = heads/(heads + tails) = 0.5 - P (a die yields odd numbers) = 1,3,5/(1,2,3,4,5,6) = 0.5 - Can account for judgments involving vehicles of chance, but hard to apply beyond that Theories of probability: try to account for how a well-constructed probability judgment is made - Frequency  mathematical inference - Logical  mathematical inference - Personal  subjective judgment Desired number of sexual partners: - USC (Miller & Fishkin 1987): how many sex partners would you ideally like to have over a lifetime? o Average (arithmetic mean):  Women: 2.7  Men: 64 Clark and Hatfield: - 5 college women and 4 college men approached subjects who were alone at 5 different locations on (FSU) campus; all begin conversation with “I have been noticing you around campus. I find you very attractive. Would you…” o Go out with me tonight? (date) o Come over to my apartment tonight? (apartment) o Go to bed with me tonight? (sex)

% men who said yes

% women who said yes

date

56

50

apartment

69

6

sex

75

0

p men would say yes

p women would say yes

date

0.56

0.5

apartment

0.69

0.06

sex

0.75

0

1. Completeness - Any proposition will either be true or false: P(A) + P(not A) = 1 A and (~A) “complete each other” (mutually exclusive/jointly exhaustive) - Extensions of the rule: P(~A) = 1 – P(A) P(A) = 1 – P(~A) - Mutually exclusive = cannot co-occur; union is an empty set (e.g. being male and female) P(A and B) = 0 P(AnB) = 0 P(A|B) = 0 - Jointly exhaustive = one or the other must occur P(A or B) = 1 P(AUB) = 1 2. Additivity - Mutually exclusive or otherwise: P(F or C) = P(F) + P(C) – P(FnC) = 0.5 + 0.6 – 0.3 = 1.1 – 0.3 = 0.8 Mutually exclusive:

P(F or C) = P(F) + P(C) F = female (0.5) C = catholic (0.6) 3. Multiplication rule (for independent events) - Being nice = 0.7 - Being good looking = 0.8 - Being clever = 0.7 - P (nice and good looking and clever) = 0.39 P(AnB) = P(A) x P(B) Independent and dependent events: - 1st draw: P(K) = 4/52 - 2nd draw: P(K) = ? - with replacement: P(K) = 4/52 - without replacement: P(K) = 3/51 - P(A and B) = P(A) x P(B|A)

Bayes’ Theorem:

P (D | H) × P (H) P (H |D) =

--------------------------------------------P(D | H) × P (H) + P (D | ~ H ) × P (~H)

Bayes’ Problem: - How strongly should I believe in B in light of some evidence E and the strength of my prior (pre-E) belief in B? - Bayes’ theorem gives us a mathematical tool (formula) to answer this question in a normatively appropriate way - It is “normatively appropriate” because it derives from the axioms of probability

Bayes’ theorem and infidelity: - My girlfriend and Milton are having an affair = Hypothesis (H) - Letter from Milton = Datum (D) - Need to find: P(having an affair|the content of the letter) = P(H|D) - P(D|H) = P(letter or evidence such as this given an affair) = 0.6 - P(D|~H) = P(letter or evidence such as this if no affair) = - P(H) = P(affair prior to the letter) = 0.001

Principle of prior dominance: - When any initially improbably outcome is further updated with a somewhat unreliable piece of information (such as that supplied by a mammogram test or a drug test), the absolute difference between the adjusted, post-test probability and the initial pre-test probability is likely to be small - To wit, if your initial belief in X is low/high, it should remain low/high even in face of further (somewhat imperfect) evidence for X - Keep things in perspective, do not judge rashly, or overact!

“It is Sunday morning at 7 A.M., and I must decide whether to trek down to the bottom of my driveway to get the newspaper. On the basis of past experience, I judge that there is an 80% chance that the paper has been delivered by now. Looking out of the living room window, I can see exactly half of the bottom of the driveway, and the paper is not in the half that I can see. (If the paper has been delivered, there is an equal chance that it will fall in each half of the driveway.) What is the probability that the paper has been delivered?” (T&D, pp.126)

Is this man a Princeton English professor or cab driver?

Two kinds of accuracy:

Humans are inveterate pattern-seekers - Especially patterns that signify agency (human or otherwise) The pattern wasn’t there: illusory correlation Brain tumor present

Brain tumor absent

Total

Dizziness present

160 (80%)

40

200

Dizziness absent

40 (80%)

10

50

Jan Smedslund (1963): - Gave nurses 100 cards supposedly representing excerpts from files of 100 patients - Each card indicated whether symptom and disease were present/absent - Is there a relationship between the two?  NO relationship between the symptom and the disease Disease present

Disease absent

Total

Symptom present

37 (69%)

17

54

Symptom absent

33 (72%)

13

46

 85% of the nurses “saw” a relationship between the symptoms and the disease ++ bias (which leads to illusory correlation)

Another Bayesian puzzle: - It is 2399 and humanity has been infiltrated by scores of human-looking androids thought to be part of an advance recognizance crew sent by a technologically superior civilization intent on colonizing the earth. The androids, known as Sentels, look and feel human and are undetectable by any means except for intracranial autopsy. Desperate for a solution, the world’s governments institute a large cash prize to be awarded to the first person or entity to develop a Sentel-detection technology that (a) can be administered without killing its subject and (b) guarantees at least 95 % Sentel detection capability upon a positive result—P (H|D) = at least 95 percent or P (Sentel|positive result) = at least .95. - You have just developed what you believe to be a viable Sentel detection method that is non-lethal in its application and has the following characteristics: given 100 Sentels, it will correctly identify 80 of them as Sentels, will deliver inconclusive results in another 2% of the cases, and will completely miss the remaining 18%. Given 100 true humans, it will rule out all of them as non-Sentels. Knowing what you know, and assuming that the overall Sentel prevalence in the general population is yet to established (P (H) = ?), are in a good position to claim the prize?...