1 / 45

Cognitive Bias and Human Decision

Cognitive Bias and Human Decision. Joseph Hellige Professor of Psychology. After Sapolsky, 2017. Drug Testing.

nathanial
Télécharger la présentation

Cognitive Bias and Human Decision

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Cognitive Bias and Human Decision

  2. Joseph Hellige Professor of Psychology

  3. After Sapolsky, 2017

  4. Drug Testing As you know, drug abuse has become a serious problem in contemporary society. In fact, a recent study indicates that 1 out of every 100 working adults is a regular user of at least one harmful (and illegal) drug. In the interest of weeding out regular drug users from his large corporation, an employer institutes mandatory drug testing (that is, testing of every employee). The test he is using is not perfect, but it is 95% accurate. That is, it will produce a positive test 95% of the time for users and only 5% of the time for nonusers. Assume that the employees of this corporation are representative of society as a whole, so that in reality 1 out of every 100 employees is, in fact, a drug user. Now, suppose an employee, call him Arnold, tests positive for regular drug use. Under the conditions I have outlined, what is the probability that Arnold is, in fact, a regular user?

  5. Probability that Arnold is a Drug User 1.00 .40 .90 .30 .80 .20 .70 .10 .60 .00 .50

  6. Class Responses Correct Response = .161

  7. Drug Use Problem Solved: Bayes Theorem Suppose we test 10,000 randomly chosen people. Of these, 1% or 100 will be users. The remaining 9,900 will be non-users. Of the 100 users, 95% or 95 will test positive. Of the 9,900 non-users, 5% or 495 will test positive. Of the 590 people who tested positive (95 + 495), 95 or 16.1% are really users. Thus, the probability of being a user, given a positive test result, is .161.

  8. Neglecting Base Rates • Neglecting base rate is equivalent to treating two events as equally likely (p = .5) • Factors that increase use of base rate • Making causal relationship clear (e.g., Tversky & Kahneman, 1982) • Using frequencies instead of probabilities (e.g., Cosmides & Tooby, 1996) • Natural Sampling • Encountering instances sequentially • Evolutionary emphasis

  9. The Monty Hall Problem(with cups instead of doors) Suppose you’re on a game show and you’re given the choice of three cups: Under one cup is a diamond; there is nothing under the others. You pick a cup, say No. 1, and the host, who knows what is under each cup, turns over one of the other cups, say No. 3, which is empty. He then says to you, “Do you want to stick with your original choice (No. 1) or switch to cup No. 2?” Is it to your advantage to switch?

  10. Representativeness Heuristic • Kellogg (1995, p. 385) • When people use this heuristic, “events that are representative or typical of a class are assigned a high probability of occurrence. If an event is highly similar to most of the others in a population or class of events, then it is considered representative” • Conjunction fallacy • The mistaken belief that the conjunction or combination of two events (A and B) is more likely than one of the two events alone

  11. If you toss an unbiased coin four times, which sequence is more likely, HHHH or HTTH?

  12. The Gambler’s Fallacy You toss an unbiased coin 12 times and it comes up heads each time. Which is more likely to come up on the 13th toss, Head or Tail?

  13. Meet Linda W. Linda is 31 years old, single, outspoken and very bright. She majored in philosophy. As a student she was deeply concerned with issues of discrimination and social justice and also participated in demonstrations advocating environmental sustainability. What is the likelihood of each of the following statements?

  14. Linda is a teacher in an elementary school. Linda works in a bookstore and takes yoga classes. Linda is active in the feminist movement. Linda is a psychiatric social worker. Linda is a member of the Sierra Club. Linda is a bank teller. Linda is an insurance salesperson. Linda is a bank teller who is active in the feminist movement. Linda is a captain in the air force. Linda is a pro-life activist.

  15. Linda is a teacher in an elementary school. Linda works in a bookstore and takes yoga classes. Linda is active in the feminist movement. Linda is a psychiatric social worker. Linda is a member of the Sierra Club. Linda is a bank teller. Linda is an insurance salesperson. Linda is a bank teller who is active in the feminist movement. Linda is a captain in the air force. Linda is a pro-life activist.

  16. Conjunction Fallacy BANK TELLERS ACTIVE IN THE FEMINIST MOVEMENT

  17. Availability Heuristic • Judgments of the frequency of events are based on what events come easily to mind • Tversky and Kahneman (1974) • If a word of three letters or more is sampled at random from an English text, is it more likely that the word starts with “k” or has “k” as its third letter? • Causes of death

  18. Imagine you randomly picked someone in the US. Will that person be more likely to die next year from cause A or cause B? Cause A Cause B Homicide Appendicitis Auto-train collision Drowning Measles Smallpox Botulism Asthma Asthma Tornado Appendicitis Pregnancy

  19. Imagine you randomly picked someone in the US. Will that person be more likely to die next year from cause A or cause B? Cause A Cause B Homicide (20) Appendicitis Auto-train collision Drowning (5) Measles (infinity) Smallpox Botulism Asthma (920) Asthma (20) Tornado Appendicitis (2) Pregnancy

  20. Framing Effects • Decisions are influenced by how the alternatives are presented • Tversky and Kahneman (1987) • The rare disease problem • Cash or credit? • Losses loom larger than gains

  21. Imagine that the U.S. is preparing for the outbreak of an unusual and rare disease, which is expected to kill 60,000 people if nothing is done. Two alternative programs to combat the disease have been proposed. Assume the exact scientific estimate of the consequences of the programs are as follows. If Program A is implemented, 20,000 of the 60,000 people will be saved. If Program B is implemented, there is a one-third probability that all 60,000 people will be saved, and a two-thirds probability that none of the 60,000 people will be saved. Which program would you choose? A: 73% B: 27%

  22. Imagine that the U.S. is preparing for the outbreak of an unusual and rare disease, which is expected to kill 60,000 people if nothing is done. Two alternative programs to combat the disease have been proposed. Assume the exact scientific estimate of the consequences of the programs are as follows. If Program C is implemented, 40,000 of the 60,000 people will die from the disease. If Program D is implemented, there is a one-third probability that no one will die from the disease and a two-thirds probability that all 60,000 people will die from the disease. Which program would you choose? C: 37% D: 63%

  23. Emotion and Decision Making • Decision making has historically been treated as almost purely “rational” • In fact, decision making (especially in the face of uncertainty) is heavily inflenced by emotion and motivation • Somatic-marker hypothesis

  24. Somatic-Marker Hypothesis(Antonio Damasio, Antoine Bechara and Colleagues) • Emotions, in the form of bodily states, bias decision making toward choices that maximize reward and minimize punishment • Iowa Gambling Task • Evidence from cognitive neuroscience • Brain damaged patients • Psychophysiology • Functional imaging

  25. Iowa Gambling Task A B C D Gain/card $100 $100 $50 $50 Loss/10 cards $1250 $1250 $250 $250 Net/10 cards -$250 -$250 +$250 +$250 Note that A and B are “bad” decks and C and D are “good” decks

  26. Limbic System • Central to emotion • Mediates aggression • Some key structures • Hypothalamus: Interface between • core regulatory and • emotion areas • Amygdala • Involved in emotion • Fear and anxiety • Mediates aggression • Social/emotional aspects • of decisions Limbic System

  27. Pre-Frontal Cortex (PFC) • Executive function • Decision Making • Two Important Subareas • dlPFC • “the decider” • Cognitive; ”Mr. Spock” • vmPFC • Many connections to the • Limbic system • Impact of emotion on • decision making; • “Captain Kirk”

  28. Insular Cortex • Strong connections to the amygdala • Connections to PFC • Involved in ”disgust,” • both physical and metaphorical

  29. Cognitive Neuroscience and the Role of Emotion • vmPFC and anticipating the emotional impact of future rewards and punishments • Such patients do poorly on the Iowa Gambling Task • May actually do better on some tasks where emotion “gets in the way” • Have problems in anticipating rewards and punishments

  30. Amygdala and registering the emotional impact of rewards and punishments caused by specific behaviors • Such patients have trouble registering rewards and punishments • Insular cortex and assessing risk and guiding behavior based on anticipation of emotional consequences • Insular cortex may serve to integrate and assess risk

  31. A runaway trolley is hurtling down the track unable to stop. Immediately ahead are five people with their backs turned to the trolley and who are certain to be killed if the trolley stays on its present course. By chance, you happen to be standing next to a lever which, if you push it, will divert the trolley onto a side track where it will certainly kill the one person walking on that track. Do you push the lever or not? Why? How do you think you would feel?

  32. A runaway trolley is hurtling down the track unable to stop. Immediately ahead are five people with their backs turned to the trolley and who are certain to be killed if the trolley cannot be stopped. By chance, you are viewing this from a footbridge above the track and standing next to a large man. The only way to stop the trolley is to push the large man down onto the track where he is certain to be killed but is also certain to stop the trolley. Do you push him or not? Why? How do you think you would feel?

  33. What’s the difference? In both of these scenarios you must decide whether to sacrifice one life to save five. But to most of us, they somehow seem different. Why?

  34. Giving an Ultimatum Imagine that you are paired with a person in another room. I give you $100. You must decide how to split that $100 between yourself and the other person, who knows that you have been given $100. You can divide it any way you choose and make an offer to your partner. If your partner accepts your offer, you each get to keep how ever much you proposed. BUT, if your partner rejects your offer, neither of you gets to keep anything—you both walk away empty handed. How much would you offer your partner?

  35. Accepting an Ultimatum??? Imagine that you are paired with a person in another room. I give your partner $100. Your partner must decide how to split that $100 with you. Your partner can divide it anyway they choose and make an offer to you. You can either accept your partner’s proposal or reject it. If you accept the proposal, you each get the amounts proposed by your partner. BUT if you reject the offer, neither of you gets anything. What is the minimum offer that you would accept?

  36. The Curious Case of Unconscious Cueing

  37. A few examples of priming…… • EAT, WASH, SO_P • The “Florida Effect” (Bargh et al., 1996) • What a difference a pencil makes….(Strack et al., 1988) • The “Lady Macbeth effect” (Zhong & Liljenquist, 2006) • The eyes have it! (Bateson et al., 2006)

  38. From Bateson et al., 2006

  39. Proportion of rulings in favor of prisoners by ordinal position. Dotted line denotes food break. Circled points indicate the first decision in each session.

  40. What marvels we humans are and what silly mistakes we are programmed to make

More Related