9.2 Biases

Some of the psychological biases that make our probability judgments go awry.

Useful Links

• Three Column Overview of the Week
• Lesson Slides (2023 Master)
• Website Page

Readings and Assignments

Learning Goals

After this lesson, students should

1. Be wary of underestimating the influence of context or circumstance on human behavior (their own and others'). E.g. Circumstance excuses my bad behavior more than it excuses the bad behavior of others.
2. Be aware of the pressure to conform, in oneself and others, and the difficulty of resisting this pressure.
3. Be able to sort situations by the difficulty of resisting conformity, according to:
   1. Number of people conforming
   2. Identification with the group
   3. Confidence that the group is mistaken
   4. Presence or absence of a dissenter
4. Be aware of the pressure to obey authority, in oneself and others.
5. Be able to sort situations by the difficulty of resisting obedience, according to:
   1. Perceived authority of the "authority" giving instructions
   2. Proximity of the authority
   3. Proximity of the "victim"
   4. Presence of another disobeying
   5. Availability or salience of alternative, disobedient behaviors

Overflow Learning Goals

There's not enough time for these topics. But, we would like to include them if we could.

After this lesson, students should

1. Predict that people will engage in temporal discounting when making decisions between something now vs. something later. E.g. People make decisions based on how good they think the outcomes will be, not based on when those outcomes will happen.
2. Be wary of their own temptation to excessive temporal discounting when making decisions, choosing immediate rewards over longer term benefits. E.g. If I choose to get something less good that I can get sooner, it's always because I need it sooner.
3. Be aware and wary of status quo bias. E.g. The way things are is the best way for things to be. Keeping things the way they are is safer.
4. Understand not to assume the status quo is necessarily stable, even if we don't try to change it. E.g. It is possible to keep things just the way they are, and that is the most stable and predictable approach.

Definitions

• Fundamental Attribution Error

  Also called lay dispositionism, this is the tendency to underestimate the importance of context and circumstance and overestimate the importance of individual differences in explaining and predicting behaviors, especially behaviors of other people.

• Conformity

  The tendency to behave similarly to others in proximity or in one's in-group, sometimes despite good reasons to behave differently from others.

• Obedience/Authority Bias

  The tendency to follow the instructions or commands of others perceived as having some authority, even when one has good reason to disobey.

• Temporal Discounting

  Evidence shows that present rewards are weighted more heavily than future ones. As rewards become more distant in time, they are perceived as less and less valuable.

  Hyperbolic discounting is a somewhat more nuanced version of this theory where people's valuations fall rapidly for earlier delay periods but more fall more slowly for longer ones. This means the difference between one and two days may be treated as more significant than between one and two months (for example).

Examples

• FAE: If someone cuts us off while driving, our first thought might be “What a jerk!” instead of considering the possibility that the driver is rushing someone to the airport. On the flip side, when we cut someone off in traffic, we tend to convince ourselves that we had good reason to do so. We focus on situational factors, like being late to a meeting, and ignore what our behavior might say about our own character. (source)
• Conformity Bias: Studies show that people are more likely to act in a prosocial manner, such as contributing to charity or conserving water, if they see or hear that others are doing it too. (source) However, it also works in the other direction, e.g. "cheating is contagious."
• Obedience Bias: A physician ordered ear drops to be administered to the right ear of a patient suffering from pain and infection. Instead of writing out completely “Right ear” on the prescription, the doctor abbreviated it, “place in R ear.” The duty nurse misread “R ear” to be “Rear.” Upon receiving the prescription, she promptly put the required number of ear drops into the patient's anus without considering whether this was a sensible treatment for ear infection. (source) <- This story came up on many sites and is supposedly from this book but has not been verified.
• Temporal Discounting: When asked if they'll take one cookie now or 3 cookies in 4 days people are more likely to take the cookie now. (note: this is generally true for short time scales and small rewards but gets more nuanced otherwise. If the reward later is sufficiently bigger, people are generally willing to wait. E.g. people will take $10,000 in 2 months over $100 right now. ) There are also strong individual differences in temporal discounting.
• "It is easy for people to disobey when they think orders are unjust or wrong."

  It is often quite difficult for people to disobey perceived authority, even when one feels strongly that the orders are unjust or foolish.

• "It is easy to avoid conforming, I rarely conform."

  It is often quite difficult not to conform, and indeed much of our conformity is entirely unnoticed and reflexive. Much of this is unproblematic, e.g. what time of day we eat or sleep, what language we speak, which side of the road we drive on, etc. are all easier if we coordinate with the people around us. However, other patterns of conformity are quite unnecessary and can be oppressive to those with other preferences - e.g. what to wear, who to love, etc. Even nonconformists typically conform to patterns given by countercultural subgroups like goths, nerds, etc.

Context

This lesson is in the middle our "heuristics and biases" series and tends to be one of the more popular lessons in the course. It focusses on people's "hot" biases that are more emotionally or socially driven. Unfortunately, people that know about these biases are generally not actually less likely to fall into them. We choose to teach the students about them anyways.

Before

8.1 Orders of Understanding

Orders of Understanding is about how each event arises from multiple influences, some more important than others. Human behavior, for instance, is affected by both individual differences and by circumstance or context, including pressures to conform and obey, and we often overestimate the former and underestimate the latter (FAE).

9.1 Heuristics

The first lesson on H&B is on "cold" biases around numbers and frequencies. The second lesson (this one) is on "hot" biases, more emotion and value-laden thinking. Both involve heuristics, strategies for making quick judgments that can be useful in some contexts but can also go awry.

After

10.1 Confirmation Bias

Confirmation bias is a broader form of bias in which people tend to look for and believe evidence or arguments supporting what they already believe or expect, and to neglect or dismiss evidence and arguments against what they believe or expect. It can exacerbate other biases, including status quo bias, temporal discounting, and/or the fundamental attribution error.

12.1 Wisdom of Crowds and Herd Thinking

Herd thinking is what happens when people in a group conform too much to each other's ideas, neglecting problems with their ideas and convincing each other to become more confident in their shared ideas than is warranted. This tendency is undergirded by the pressure to conform to the group and also the pressure to go along with anyone in the group perceived as an authority (obedience).

14.1 Scenario Planning

Scenario Planning involves considering four different possible future scenarios. It is one strategy that may reduce status quo bias and temporal discounting by making more salient a variety of possible futures, and how they may differ from the present in both desirable and undesirable ways.

Recommended Outline

Before Class

• Prepare a seating chart.
• Review PlayPosit and discussion questions and ask faculty, Gabriel, or Emlen any questions you have.
• Prepare Google Docs and Forms (and possibly slides) as detailed in the activities below.
• Prepare enough sticky notes for your class for the fundamental attribution error activity.
• (Optional) Prepare a presentation.

During Class

• (25 min) Lead the students through the conformity discussion.
• (24 min) Lead the students through the obedience discussion.
• (11 min) Play the fundamental attribution error game.
• (12 min) Lead the students through the status quo bias discussion. Skip this if pressed for time.
• (8 min) Lead the students through the temporal discounting discussion.

Lesson Content

Conformity

1. (4 min) Watch Asch experiment video.

Discussion Questions

1. (8 min) Why do you think the participants in the Asch experiment conformed, even though they could see the other people were not correct? Can you think of a time you conformed without really thinking about it? Why did you conform then? Can you think of a time you didn't conform? Why didn't you? Was it difficult? How did others react?
2. (5 min) Discuss as a whole group. What surprised you? Are there sometimes good reasons to conform? What makes it hard to resist conforming? What enables people to resist conformity?

Obedience

This activity introduces students to the unexpected power of the pressure to obey authority, showing them the classic Milgram study and inviting them to discuss its implications and ramifications.

Instructions

1. Warn students that the video is disturbing, and assure them that no one was actually physically injured. Show them this video up to 5:27, and stop.

Discussion Questions

Discuss in small groups:

1. Why did so many people continue to push the button, even after the "learner" cried out in pain and begged them to stop?
2. What situational factors do you think contributed to people feeling pressure to push the button?
3. What situational factors might increase people's willingness to push the button? What situational factors might decrease their willingness? What questions do you have for plenary?
4. What would prevent you from pushing the button all the way through? How can you be sure you wouldn't? (How similar would the situation have to be for you to recognize the similarity to Milgram's study?)
5. W hat could we do at the level of institutions that might reduce the pressure to conform and obey to do harmful acts? (You can draw on strategies you came up with at the end of the PlayPosit).

Return to the whole class. GSI explains: Milgram did many variations on this experiment, and found that the further away the experimenter, the less people pushed the button; and the closer the victim, the less people pushed the button.

1. Why do you think experimenter distance made it easier to disobey?
2. Why do you think victim proximity made it easier to disobey?
3. What implications does this study have for our society? For the increasingly technological nature of war (e.g. drones piloted at a distance)?

Fundamental Attribution Error

In this activity students are introduced to the broader phenomenon of the fundamental attribution error. The FAE describes people's surprise at both the Milgram and Asch experiments, as well as other situations where people underestimate the power of the situation. In other words, a lot of bad behavior comes not from bad people doing bad things, but from regular people in difficult situations. We can take this optimistically; it's not that people are intrinsically bad, it's that it's really hard to resist situations. So if we can change the situations—if we can make systemic change happen—then maybe we can drastically reduce harmful behavior. Students will practice generating alternative situational explanations to help them recognize the power of situation, and to reduce the natural tendency to underestimate it.

The activity is inspired by this game.

Preparation

• Enough sticky notes for your students.

Instructions

Give the students an example and then the following instructions.

1. Write some strange past behavior of yours that could be misinterpreted on the sticky note.
2. Once everyone in your group has written them, trade the notes with another group and try to come up with creative explanations for them.
3. Once you have explanations for the other group, get back together and share your explanations and the real ones.

Afterwards, the GSI should call the class back together and see if anyone has any particularly amusing past behaviors and misinterpretations to share.

Discussion Questions

1. Why is it so easy to jump to conclusions?
2. Why is it useful to consider multiple possible explanations for a behavior?
3. How can we improve our institutions to reduce situational pressures that often cause harmful behavior (e.g. through obedience and conformity, as in the Milgram and Asch experiments)?

Status Quo Bias

Students are introduced to status quo bias, and discuss its justifications, injustices, and ramifications for society.

1. (5 min) GSI presents briefly on status quo bias.
   1. Status quo is Latin for "the state in which." It means the current way things are. The status quo bias is thus the tendency to want to keep things the way they are, an attitude of "if it ain't broke don't fix it" and sometimes "changing it will only make things worse" or "if you can't make it perfect why bother to change it at all."

Discussion Questions

1. (5 min) Small group discussion:
   1. What are some reasons that the status quo bias might be rational in some circumstances?
   2. What are some reasons that the status quo bias might be irrational in some circumstances?
   3. How do we distinguish what aspects of the status quo are worth maintaining, and which are worth changing?
2. (5 minutes) Whole group discussion
3. What insights rose in your small group discussion?

Temporal Discounting

Students are introduced to temporal discounting and discuss its implications for personal and societal decision-making.

Have students watch this intro to temporal discounting (4 min) OR GSI explains it with a slide or two.

Discussion Questions

1. Why would we value the same amount of money more now than in the future?
2. Temporal discounting preferences are subjective. Under what conditions, if any, can we say they are irrational?
3. How might the tendency for steep temporal discounting play out in policymaking?
4. How does the tendency to value present goods over future equal goods exacerbate climate change?
5. Extension/Stretch/Bonus: Because of the hyperbolic shape of the temporal discounting curve, it's easier to give up soon-but-future rewards for much later and larger rewards than immediate-but-smaller rewards. This is one possible explanation for the movie-choosing pattern:
   1. When told to choose a movie to watch tonight vs. next week, people are more likely to choose an action flick or a chick flick for tonight, and more likely to choose a documentary or something more artistic for next week. Why do you think this is?
   2. Given this pattern, how does the transition from ordering, borrowing, or buying videos to watching everything streaming online change our video-watching habits? Is there any way to counteract this?

---

⟨

Previous

9.1 Heuristics

⟩

Next

10.1 Confirmation Bias