What do you pick when you see a sequence of options?

Some choices are made when all of the options are right there in front of you.  For example, if you go to the supermarket, you see the wall of tomato sauce, and you can stand in front of it and compare all of the different brands and the different flavors of sauce for each brand.

For other choices, though, you get information about the options one at a time, and then you have to make a decision.  For example, if you are buying a new car, you have to go from one dealership to the next, test driving cars and talking to salespeople.  Students who apply for graduate schools may visit the schools to which they have been accepted in a sequence until they have to make a decision about where to go.

What happens in those cases?

A paper by Antonia Mantonakis, Pauline Rodero, Isabelle Lesschaeve, and Reid Hastie in the November, 2009 issue of Psychological Science looks at this question.  They had people sample a number of wines in a sequence.  After sampling the wines, they had to decide which one they preferred.  Participants were assigned to try 2, 3, 4, or 5 wines, and the sequence was determined randomly.

The choices people made depended both on the number of wines they sampled and their expertise about wines.  For people who did not have much expertise, they were most likely to choose the first wine they sampled.  For example, when there were two wines, people selected the first wine they tried 70% of the time.  For a sequence of five wines, they still picked the first one 50% of the time (but if they had chosen a wine at random, they would only have chosen this wine 20% of the time). 

For people who had more expertise, there was a combination of two effects on choice.  There was still a tendency for experts to select the first wine they tried.  When sampling only 2 wines, they picked the first one 80% of the time.  When sampling 3 wines, they picked the first one about 65% of the time.  But, when they sampled 5 wines, they picked the first one only about 25% of the time.  In this case, they picked the last wine they tasted about 40% of the time. 

Why is there a difference between experts and non-experts?  The wine experts have a more sophisticated ability to taste differences among wines.  So, they keep comparing each new wine they sample to their favorite.  There is a tendency for the first wine to be a favorite, because it sets the standard for comparison.  However, each new wine that is sampled will have some new flavors, and so they provide a reason to pick the last wine tasted.  The experts taste these new flavors and so they tend to stick with the most recently tasted wines, particularly when they taste a long sequence of wines.  The non-experts can’t taste much difference among the wines, so they often stick with the first one they sample.

One thing that is interesting about these results is that the wines in the middle of the sequence are not chosen very often.  There is an overwhelming tendency for people to pick either the first or the last of the items in the sequence.  For example, for experts choosing from among 5 wines, the first and last items accounted for about 65% of their choices, while the remaining 3 items accounted for only about 35% of the choices.

What does this mean for you?

If you are in a situation in which you have to sample the items in a sequence, and if the choice is important to you (say a car or a graduate school), you should try to write down your criteria for making a choice in advance.  Try to evaluate each option after seeing it by using these criteria.  If you discover a new dimension for evaluating options after seeing the third or fourth item, go back and evaluate the earlier items along that dimension as well.  Do what you can to give each of the options (even the ones in the middle of the sequence) the best chance to be the one that gets chosen.   

More information makes you more confident, if not more accurate.

Confidence matters.  We are much more likely to act in situations when we are confident.  We make purchases based on confidence.  We are also persuaded by others based on their confidence.  A statement made confidently and forcefully is much more likely to sway our opinion than a statement that is hedged. 

Presumably, the power of confidence lies in the belief that when people are more confident in an outcome they are more likely to be correct in their predictions.

There have been many studies over the years that demonstrate that people tend to be overconfident in their judgments about how likely they are to be correct about a prediction or an answer to a question.  But what causes this overconfidence?

A 2008 paper by Claire Tsai, Joshua Klayman, and Reid Hastie in Organizational Behavior and Human Decision Processes suggests one factor that makes people overconfident.  They find that as people get more information about a judgment they are making, it increases their confidence, even if it does not increase the accuracy of their judgment.

In one study in this paper, they found experts in college football and asked them to predict the outcomes of a number of games.  The names of the schools playing the games were not given.  Instead, people were given information about how the teams had performed to that point in the season on a variety of aspects of performance that are useful for predicting the outcome of a game (like the average number of yards that the teams had gained in their games so far that season). 

For each game, the participants were initially given six measures of performance for each team and were asked to predict which team would win the game and the point spread between the teams.  They were also asked to rate their confidence in their judgment.  Based on this initial information, people predicted the outcome of the game correctly about 63% of the time, and their average confidence was about 68%.  That is, they were overconfident, but not very overconfident.

Then, people were given 6 more measures of performance and they repeated the judgment, again predicting the outcome of the game and the point spread, and giving their confidence.  This was repeated 3 more times, so that people ultimately made 5 judgments and saw 30 measures of performance for each game. 

As people got more information, their overall accuracy did not change much.  After seeing all 30 cues, people were only correct in predicting the winners of the game 67% of the time.  However, by the end of the study, they were not 79% confident in their responses.  That is, their confidence went way up as they got more information, even though their accuracy stayed the same.

One thing that seems to happen as people get more information is that they start to go from specific information to more general evaluations of aspects of team performance.  So, based on a few different measures of performance, a judge might now make an evaluation of the effectiveness of a team’s run defense or passing attack. 

Of course, all information is ambiguous.  Not every measure of performance tells the same story about teams.  As you get more information, though, you become more free to pick and choose the information that is consistent with the story you are trying to tell.  In this way, you find information that helps you to confirm what you already believe to be true.  This tendency to seek information that supports a conclusion you have already drawn is called confirmation bias.  This confirmation bias will increase confidence in the judgment you have made.

It is important to realize that confidence and accuracy are not that highly related.  We are often asked to make decisions on the basis of expert opinion.  There is a temptation to rely on the confidence of the expert to decide how much of that expert’s opinion we should trust.  Perhaps we are better off looking at that expert’s past performance to make up our own mind about how confident we should be in the accuracy of their judgments.

Vice, virtue, and the box office.

It is summertime, and the movie studios trot out their summer blockbusters and comedies.  That’s what summer is all about.  In Texas, when it gets to be 102 outside, sitting in an air-conditioned movie theater is not such a bad thing.  But why are people flocking to movies that are clearly so awful.  The poster-child for awful movies this year is Transformers: Dark of the Moon.  The reviews of this movie are uniformly terrible, but the film has already done over $300 million in sales in the United States alone.  What gives?

We could assume that the critics are all just wrong.  But that doesn’t seem to be a complete explanation.  Everyone I know who saw this movie says it was a confused mass of metal.  So the hordes going to the movie must reflect something more basic about the way people make decisions.

The success of summer movies rests on what the behavioral economist George Loewenstein called vices  and virtues.  He’s not talking about the themes of movies here, but about the way people make their choice about the movie they want to see.  The vices in movies are all about the enjoyment.  The virtues are all about the value of the experience of seeing the movie.

Some of the vices in movies are thrills.  Car chases, explosions, and action heroes mowing down endless rows of evildoers are all vices.  Neatly-tied emotional experiences are vices as well.  The standard boy-meets-girl, boy-loses-girl, boy-gets-girl plot of the standard romantic comedy is a vice.  We know by the closing credits the two beautiful stars will be locked in a kiss.  Familiarity is also a vice.  We are far more comfortable with remakes, sequels, and movies that come from familiar TV shows and toys than with things that are new.

Virtues are things that add value to the movie-going experience.  Learning something new is a virtue.  (Documentaries—even well-made documentaries—are chock-full of virtue.)  A complex emotional experience is a virtue.  Grappling with real-world issues is a virtue.  Movies with virtues are satisfying both in the theater and also long afterward.

Loewenstein suggests that people recognize the value of virtues, but when they are actually at the theater ready to buy a ticket, they invariably fall back on vices.  In one study, he gave people a coupon good for a free rental of a movie that they could pick up either that night or the next week.  They had to select the movie they wanted when they selected the coupon. They were given the choice of a virtue movie (Schindler’s List—a movie about the holocaust) or a vice movie (a low-brow comedy).  When people were choosing the movie they would watch next week, they tended to pick the virtue movie.  When they were choosing the movie they would watch that day, however, they would pick the vice movie.  In the moment of choice, people were worried that they would be bored, or sad, or generally not entertained.

Interestingly, movie delivery services like Netflix provide an interesting test of this explanation.  With Netflix, you set up a queue of movies and they get mailed to you.  So, you are choosing movies that you are going to watch some time in the future.  I find that I will put a documentary or a complex heavy “virtue” movie on my queue, and it will get mailed.  But then, it will sit at the house for a few weeks because the time never seems right to watch it.  Eventually we watch it, of course, and typically we really enjoy it and value the experience.  But in the moment of truth, it is hard to actually decide to watch it.

The same thing happens at the movie line over the summer.  In the end, we just want a pleasant experience at the movies.  So, even if the reviews are horrible, we end up valuing the potential thrills, comfort, and familiarity over the virtues of depth, complexity, and knowledge.

When will this streak end?

People have two very different responses to streaks.  For example, when watching a basketball game, we see a player hit two three-point shots in successive trips down the court, and we expect him to make a shot on the next trip down the court.  The idea that a streak is highly likely to continue is a belief in a hot hand. 

However, if you are playing roulette at a casino, and the lands on a red number 4 times in a row, you expect that there is a good chance that it will land on black the next time.  The idea that the odds that a streak will end go up as the streak gets longer is called the gambler’s fallacy.  

Just looking at these two reactions, it is clear that sometimes we expect streaks to continue and other times we expect streaks to end.  What determines which of these reactions we are going to have?

Eugene Caruso, Adam Waytz, and Nicholas Epley looked at this issue in a 2010 paper in the journal Cognition.  They find that you are most likely to assume that a streak will continue (the hot hand) when the cause of the streak is seen to be someone’s intentions.  You believe the basketball player will keep hitting shots, because he intends to make the shots.

When the streak is caused by some kind of mechanical device, then people believe the streak will end (the gambler’s fallacy).  The roulette wheel is a mechanical device that does not intend for the ball to land on either black or red numbers.

These authors tested this idea with a series of clever experiments and studies.  In one experiment, people watched a video of someone flipping a coin.  One group was told to focus on the intentions of the coin flipper to understand “what he is trying to accomplish with his tosses.”  A second group was told to focus on his actions, “the specific movements of his hands and fingers.” 

At various points, the groups made predictions for what the next coin flip would be.  For one prediction, the previous 8 tosses involved a random-looking sequence of 4 heads and 4 tails.  For a second prediction, 6 of the previous 8 tosses had been heads including a streak of 4 heads in a row.  When the sequence was random, people in both groups predicted that the coin would come up heads about half the time.  When the sequence ended with a streak, though, people who focused on the person’s intentions predicted that the coin would come up heads 68% of the time, while those who focused on the person’s actions predicted the coin would come up heads 28% of the time.  That is, thinking about intentions led people to think the streak would continue, while thinking about the mechanism of the flip led people to think it would end.

In another study, people made predictions about stock prices.  They were shown graphs of the performance of some stocks over a two-week period.  The critical items in this study were graphs that had trends in the stock price.  In one graph, the price of the stock went up consistently over the two week period.  Another second graph had a decreasing trend.  Participants were asked to predict the next day’s stock price.  Unsurprisingly, people shown the increasing trend assumed that the stock would keep going up.  People shown the decreasing trend assumed that the stock would keep going down. 

However, the authors had participants complete a questionnaire about an individual difference in anthropomorphism.  That is, some people have a tendency to give human traits and mental characteristics to inanimate objects, while others do not.  Those people who were most likely to think that the stock market has a mind of its own were the ones who were most likely to think that the streak in stock prices would continue.  Those people who were most likely to think that the stock market has no intentions were most likely to think that the streak in stock prices would end.

This last study has important implications.  In daily life, we must often make predictions about what will happen in complex situations.  It is important to recognize those situations in which we have a good causal understanding of the situation and know whether the data we observe are good predictors of future performance and those situations in which the most recent observations are just the outcome of a generally random process.  In these situations, we must take some care not to ascribe intentions to processes that are really random. 

It is more difficult to avoid anthropomorphizing than it looks.  We talk about many institutions using language often associated with the actions of people.  Reports in the business sections of newspapers talk about markets “reacting” to news.  They describe companies “striving” to overcome poor performance.  This language reinforces the idea that organizations may have intentions. 

At times, it may make sense to treat a company like a person.  Companies have managers who make decisions and these decisions may properly be cast as intentions for the strategy that will be followed in the future.  At other times, though, this language may get in the way.  There is a tremendous amount of randomness in the stock market.  When we talk about it in human terms, we gloss over all of that randomness.

In the end, we should try to be a bit more mindful about when we assume that streaks are the results of an agent’s intentions.

What is your pain and suffering worth?

The civil courts are one of the most humane systems society has invented.  We take property damage, pain, suffering, and lost wages that have been caused by the negligence or even the willful actions of others and translate them into a dollar amount.  This system allows people to settle their disputes peacefully.  But it does rely on our ability to translate things like pain and suffering into a monetary value.

What factors influence the way we judge the value of intangible elements like pain and suffering?

There was an  interesting paper exploring this issue in the August, 2008 issue of Psychological Science by Eugene Caruso, Dan Gilbert, and Tim Wilson.  They looked at the value that people give to things in the past and in the future.  Consider, for example, an auto accident.  A woman in her car is struck head-on by another car driven by a man who didn’t pay attention to a stop-sign.  The man is clearly at fault.  The woman is injured and it will take 6 months for her to heal.  How much should she be awarded by the insurance company for her pain and suffering?  The researchers asked this question in two ways.  In one situation, the accident was 6 months ago, and the woman is now completely healed.  In the other, the accident just happened, and she is beginning her recovery period.  They found that people were willing to award the woman twice as much when the pain and suffering was yet to happen than when it was now over.  So, future pain and suffering was more valuable than past pain and suffering. 

One interesting side-note, people do not believe that they ought to value the present and future differently.  The researchers did some studies in which they asked both the past and future questions to the same people.  In this case, people gave the same value to past and future events.  However, if they saw the past question first, then the values they gave to both events was lower than if they saw the future question first. 

It is also worth pointing out that this effect was not limited to negative events.  In another study, people were asked how much they would want to be paid for 10 days worth of boring work.  People asked for more money if they were asking for work they had yet to do than if they were being asked for payment for work they had already completed.

So, what does this mean?  I guess that depends on whether you are the insurance company or the person who has suffered damage…Insurance companies ought to wait to place any value on pain and suffering until they are over.  They will seem less valuable when they are finished than while they are ongoing.  On the other hand, if you have suffered damage, your pain and suffering will have the most value while you are going through it.   

Why is it so hard to quit smoking?

If you take even a quick look at the newspaper, you can see that there are massive public policy efforts aimed to curb smoking.  States are raising their cigarette taxes.  Cities around the world are banning smoking in restaurants.  Even Holland (where it seems like almost every vice is legal) no longer allows cigarette smoking in restaurants and public places.  What makes it so hard to quit smoking?

That would, of course, be a topic for a whole book, and not just one blog entry.  It does provide a good excuse for me to talk a bit about some research that I have done with Miguel Brendl, Claude Messner, and Kyungil Kim.  The question of interest to us is how a goal or a need affects what you like.  For now, let’s think about smoking.

The need to smoke is affected by many factors.  Some of them involve a physiological addiction to nicotine.  However, the situation also affects the need to smoke.  Smoking is often associated with drinking coffee, for example, so a cup of coffee will often trigger the need to smoke in a smoker.  So, the strength of a goal will grow and shrink.  Sometimes, the need will seem desperate, but at other times (say right after having a cigarette), it won’t seem so strong at all.

When the need to smoke is triggered, it affects what you like and dislike at that moment.  The idea that likes and dislikes can change over short periods of time shouldn’t be surprising.  After all, you may grab a CD to listen to in the car in the morning, only to find that you no longer feel like listening to it in the evening.  Foods that were quite appealing on one day may be much less appealing on another day.

One factor that changes your preferences is the goals that are engaged at any given moment. Most of us have the intuition that if you have any strong goal, then your preference for things that would help you satisfy that goal will increase.  If you really need to smoke, then suddenly, cigarettes will look really good to you.  We call this effect of goals valuation.  That is, having an active goal makes things related to that goal more valuable.  There is some experimental evidence for this kind of valuation.

What is potentially more interesting, though, is that in a 2003 paper in the Journal of Consumer Behavior, and a 2007 paper in Emotion, we find strong evidence for devaluation.  That is, having an active goal will decrease your liking for things that are unrelated to that goal.  So, a smoker with a high need to smoke is less interested in things like DVD players, french fries, and camping tents than that same person is when the need to smoke is low. 

We found evidence for this in many ways.  In one study with smokers, we had smokers participate in a study after sitting in a long lecture class.  They could not smoke during class.  One group was asked to stay in a classroom and was given a cup of coffee (to stimulate the need to smoke).  This group really had a high need to smoke.  The other group went outside the classroom. They were also given a cup of coffee, but the experimenter lit up a cigarette, and all of the participants did too.  For this group, the coffee was a way to help us make sure enough time went by that the nicotine would help dull the need to smoke.

Before getting on to what they thought the study was really about, the participants were asked if they were interested in buying raffle tickets.  For half of the people, the prize was three cartons of cigarettes that would be given out in a drawing a week later.  For the other half of the people, the prize was an amount of money equivalent to what you’d pay for three cartons of cigarettes.  So what happened?

People who were offered the raffle to win cigarettes were somewhat more likely to buy tickets if they had a high need to smoke than if they had a low need to smoke.  That is, there was some evidence for this idea of valuation. 

People who were offered the raffle to win cash bought tickets if they had a low need to smoke.  The group in the classroom (who had a high need to smoke) bought almost no tickets at all.  That is, when people had a high need to smoke, they were really uninterested in cash. 

There are two important things to take away from this.  First, even though people know abstractly that cash can be used to purchase cigarettes, their goals have a very concrete effect on their preferences.  So, things that are not obviously related to smoking are devaluated.  Second, the people who are inside the classroom and need a cigarette probably walked out of the classroom after the study and smoked a cigarette.  Presumably, if we had offered this raffle to them after having their cigarette instead, they would have been more interested in buying tickets to win cash.  So, cigarettes can change people’s preferences pretty rapidly.

This means that a smoker trying to quit smoking can talk quite a bit about how they are going to resist the urge to smoke.  But when that need gets strong, the goal system is going to make cigarettes more and more attractive and is going to make everything else less and less attractive to help make the smoker satisfy their need to smoke.  This operation of the motivational system is usually a good thing.  It operates for all sorts of beneficial goals that people have.  But for habitual smokers, it creates a lot of problems.