Understanding Framing Effect, Sunk Costs, and Confirmation Bias
School
University of Pennsylvania**We aren't endorsed by this school
Course
PSYC 1
Subject
Philosophy
Date
Dec 12, 2024
Pages
4
Uploaded by AgentIronMoose57
Study QuestionsLecture 15, Nov 7Q1: What is the framing effect? What is the classic study that shows this effect?Framing effects occur when the way a judgment or decision problem is phrased, or the way thevarious response options are described, influences participants’ responses (i.e., their choices orjudgments). In most cases, framing effects occur when different formulations of the question athand change a salient point of reference.For example, in a classic study by Kahneman & Tversky (1981), participants were presentedwith the following question:Imagine that the United States is preparing for the outbreak of an unusual disease, which isexpected to kill 600 people. We have two plans. (options framed as losses)· Plan A: 400 people will die· Plan B: 1/3 chance no one dies; 2/3 chance all 600 dieAnother group of participants received the same problem with options framed in a different way.The plans were framed as gains rather than losses:Imagine that the United States is preparing for the outbreak of an unusual disease, which isexpected to kill 600 people. We have two plans. (options framed as gains)· Plan A: 200 will be saved· Plan B: 1/3 chance will 600 saved; 2/3 chance nobody savedNote that in both cases, Plan A and Plan B have the same outcomes. However, when optionswere framed as losses, most people choose Plan B; when options were framed as gains, mostpeople choose Plan A. Kahneman & Tversky (1981) interpreted the results as indicating thatpeople are risk-seeking (i.e., prefer risky options) when they are considering losses, and risk-averse (i.e., prefer certain options) when they are considering gains, even if the outcomes theyare considering are logically or economically equivalent.Q2: What are sunk costs and the endowment effect?Sunk costsrefer to the tendency for individuals to continue investing in a decision based onpreviously invested resources (e.g., time, money, effort) that cannot be recovered, rather thanfuture potential gains. For example, if someone bought a ticket to a show and then lost it, theymight decide to skip the show because they don’t want to “waste” additional money, eventhough the initial cost is unrecoverable. Rationally, since the original expenditure is a sunk cost,it should not affect the decision to purchase a new ticket.
Endowment effectis the phenomenon where people place a higher value on objects they owncompared to objects they do not own, even if there is no difference in objective value. Forexample, when people plan to buy a ticket but find they have less money than expected, theymay still decide to purchase it even if the cost is right around the maximum they’d want to pay.However, once they own the ticket, their perceived value of it increases. If an opportunity arisesto sell the ticket or skip the concert, they might set an irrationally high price or refuse to sell italtogether, believing it to be more valuable simply because it is now theirs. This change inperceived value illustrates the endowment effect, where ownership influences decision-makingand creates an emotional attachment that affects choices.These two effects highlight how our mental reasoning can be trapped in a box, leading toirrational choices that prioritize past investments or ownership over objective value and logicaldecision-making.Q3: What is confirmation bias? How might confirmation bias have a negative impact onreasoning?Confirmation bias is the tendency to favor information that supports one’s existing beliefs (andto disregard information that is contrary to their existing beliefs), regardless of whether thatinformation is true or diagnostic. As a result of confirmation bias, people will often spend moretime gathering favorable or confirmatory information than unfavorable/disconfirmatoryinformation when forming judgments or opinions (e.g., watching only liberal or conservativenews when forming opinions about divisive political issues); may recall favorable informationselectively; and may interpret ambiguous information as evidence in favor of their existingposition.While we might debate whether it is okay to be biased in interpreting information, confirmationbias can have a negative impact on reasoning because it often drives people to draw (perhapsincorrect) inductive inferences from skewed or biased samples of information. In addition, theconfirmation bias can undermine critical thinking by allowing individuals to interpret neutral,ambiguous, or false information as evidence in support of their existing (and perhapsunsubstantiated) beliefsQ4: What is the availability heuristic?The availability heuristic is a mental shortcut by which individuals use the ease with which theycan recall a particular piece of information as a cue for how likely that piece of information is tobe true (or, in some cases, how likely an event is to occur). This is often a good heuristic: thingsthat are more available in memory usuallydo occur more frequently in the world. However, thisheuristic sometimes leads us astray. For example, horses hurt more Americans than sharks do(horses: about 12000 hospital admissions per year and about 100 deaths; sharks: fewer than100 attacks p.a. and <5 deaths per year). However, people appear to think the opposite,because it is much easier to recall examples of dangerous sharks (e.g., news reports ofAustralian Great White attacks) than examples of dangerous horses. In keeping with theavailability heuristic, this is largely true because horse-related injuries are not frequently
reported on the news (and thus, are less likely to be “available” in memory), unless the rider isfamous.Q5. Describe base rate neglect and provide an example of how it works.Base rate neglect describes the tendency for people to ignore relevant population statistics infavor of case-specific information. The example described in class (the made-up ailment bogitis)had a disease with a prevalence of 1/1000, and a test that would always detect a positive, butwould result in a false positive 5% of the time. If someone tested positive using this test,individuals showing base rate neglect would predict that this person would have a 5% chance ofactually having the disease, ignoring the actual prevalence and relying only on the informationabout the test. (In reality, the likelihood this individual has bogitis is about 1.96%.)Another example might describe Mary, a woman who is very shy, very reserved, likes to solvecrossword puzzles, and owns several cats. Would you be more likely to say she is asalesperson or a librarian? Even though there are many more salespeople in the world thanlibrarians, we are more likely to say that Mary is a librarian, because her personality aligns withwhat we expect from librarians. We are ignoring the base rate (the prevalence of salespeople inthe population compared to librarians) in favor of case-specific information (Mary’s personality).Q6. Describe the findings from Lord, Ross, and Lepper regarding individuals’ opinionsabout the death penalty and which heuristic described in lecture it illustrates. Lord, Ross, and Lepper presented individuals who had strong opinions on the death penalty(either in favor or against) with two pieces of “scientific evidence.” (actually made up, butimitating real scientific analyses.) One piece argued that the death penalty did deter people fromcommitting murder, and one piece argued that the death penalty did not serve as a deterrent.They showed that people tended to accept the piece of evidence that aligned with their beliefs,and discard the other. This resulted in a stronger polarization of the two groups relative to thestart of the experiment, with each group showing stronger beliefs in their original viewpoint. Thediscarding of evidence that disagreed with their viewpoint and acceptance of information thatsupported it is a prime example of confirmation bias at work.Q7. What is Kahneman’s dual-process theory? What is the difference between system 1and system 2 thinking? Dual-process theory is the proposal that judgment involves two types of thinking: a fast,efficient, but sometimes faulty set of strategies (system 1), and a slower, more laborious, butless risky set of strategies (system 2). System 1 is automatic thinking. A lot of heuristics we have such as the availability heuristic andthe representativeness heuristic result from system 1 thinking, as well as many facets ofperception.
System 2 is effortful thinking. Evidence shows that people are more likely to use system 2thinking when the problem is easily quantified. For example, some argue that people tend to berelatively sophisticated when thinking about sporting events. In such cases, each player’sperformance is easily assessed via the game’s score or a race’s outcome, and each contest isimmediately understood as a “sample” that may or may not be a good indicator of a player’s (orteam’s) overall quality.Q8. Summarize Henrich’s argument that humans have an innate capacity for culture,including both content and context based mechanisms for learning this culture. While other apes can engage in social learning, humans do it better, are more motivated to doso, and rely more on information they gain from others than their own instincts. Content-basedmechanisms drive this social learning in different domains, like food, gossip, and social norms,and allow us to quickly and efficiently learn and store information within these domains. Context-based mechanisms drive who we seek out to learn from: age, sex, skill, success, ethnicity, etc.We use this information to prepare for future social situations to ensure we conform to our socialgroup.Q9. Henrich argues that human learners don’t just try to learn from the largest, mostphysically dominant individual. Who are learners paying attention to?Henrich theorizes that cultural evolution builds up a body of skills and knowledge that contributeto survival. If some individuals have more knowledge or more accurate know-how than others,those individuals will accumulate prestige, which is like dominance but is not based on threats offorce. Q10. Henrich makes the anthropological observation that as societies become larger,their religions include gods that are more powerful, less humanlike, and omniscient. Whymight this be?Henrich argues that this might be a consequence of larger societies’ need for a widespreadbelief that norms should be followed even outside our own clan. If I depend on people whoaren’t in my family and not personally known to me, which tends to happen more as societiesget larger, I will have more confidence if we all have a belief that the rules have to be followed,because there is a god that can exact punishments on cheaters and norm-breakers.