jim.shamlin.com

Predictably Irrational

Author: Dan Ariely
HarperCollins: 2008

This originally was a book-length manuscript, but is extremely chatty and pack with "filler" material, so it has been condensed into a single page of notes.

Introduction

The introduction to the book provides a few examples of plainly irrational choices, then degrades into a cocktail party argument in which the author made someone mode educated than himself seem foolish.

Aside of the smarminess and narcissism, he raises a valid point: traditional approaches to economics presume that all of mankind's decisions are guided entirely by the angels of our better nature - when we are completely rational and self aware, free from vulgar desires and psychological dysfunction.

His criticism of traditional economics is this: it is based on rational and informed decisions. But in real life, people seldom have complete information about a decision and make highly irrational choices. And that's what this book is about - the real decisions that people actually make, as opposed to the ideal decisions that people do not.

(EN: And the danger of this book, and particularly of those who have read it without much knowledge of economics or psychology, is in swinging the pendulum to the opposite extreme and suggesting that all decisions are entirely irrational and flawed. In reality, people attempt to be rational, and the choices they make usually reflect this - to suggest that people are either entirely or not at all rational sets up a binary, all-or-nothing fallacy that is completely unrepresentative of reality. And while psychology does provide some insight into consumer behavior that seems irrational, a great deal of caution should be taken when considering the kind of dime-store "pop" psychology that books like this contain. It champions an under-represented perspective that should likely be given more serious consideration, though likely by a different author.)

1: The Truth about Relativity

Economics is made in a very narrow scope of information available to an individual. In the first place, we do not know what we want until we learn what we want. And in the second place, when we learn what we want, we choose it within a limited context.

He uses the example of a magazine subscription, which offers one price for an online subscription, a second price for print, and a third price for both - only the third price is the same as the second. In an informal experiment, he shows that most people (84%) choose the print-and-web option - but if the second option is removed (print only), fewer people (32%) make that choice. Ergo, some people who chose print-and-web didn't really want print at all: they chose it because it looked best by comparison.

(EN: The experiment was among students who were not actually purchasing the magazine, so this is a case of an invalid sample and an unrealistic situation, but it seems plausible that a valid experiment would also show the same difference, likely to a lesser degree.)

This is the reason many retailers and manufacturers stock more than one product of a given kind - so that customers do not see it as a choice between buy and don't-buy, but which item to buy (and purchasing is a foregone conclusion).

Another interesting example is a bread-machine manufacturer who could not get many people to take interest in the machine. But then, they created a "deluxe" version of the machine and their standard version sold like hotcakes.

Through a few other examples, he illustrates a basic sales tactic: give people three options - a good and a bad choice (A or B), and then introduce a slightly inferior version of the good choice (A-minus). This allows them to see the price of the A choice as relative to that of the A-minus choice (and slightly more expensive) rather than seeing the A choice as relative to the B choices (significantly more expensive).

There's also a bit about relativity in terms of scope. People will go to a different retailer to save $10 on a $25 pen, but they will not go to a different retailer to save $10 on a $500 suit. The net value is still the same (seven dollars) - but since it's nearly a half of the price of a pen but only two percent of the price of the suit, the former seems more important. Rationally, seven dollars is seven dollars.

A final bit considers how people feel "wealthy" in regard to their own cultural reference group. The rich do not compare themselves to the poor, but only to other rich people, and feel dissatisfaction if they are not better off than their peers. The poor likewise do not compare themselves to the rich, but to other people in their neighborhood, and feel affluent if they're better off than their neighbors.

(EN: This is actually a common criticism of the media - the unrealistic way that characters in television and in books are portrayed gives people a glimpse at a skewed lifestyle. E.g., a sitcom about a working-class character in her twenties lives is set in an apartment that no-one that age could afford, filled with things they likewise could not afford, which causes people to feel dissatisfied with their own lot.)

2: The Fallacy of Supply and Demand

Ariley next attacks the economic principle that prices are determined by supply and demand of goods. (EN: Whether he is ignorant or merely setting up a straw man to facilitate an easy victory is unclear, but this is a common misinterpretation. A buyer's willingness to pay is a reflection of his desire for an item independent of its availability. The more he wants it, the more he is willing to pay. Supply and demand meet in the market when there are insufficient goods for all buyers, and sellers seek the best price they can get by selling to those who want it most, hence offer the highest price. So price has always been set by desire, not supply, and the effects of supply and demand are secondary.)

He mentions how a market was created for black pearls. At first people rejected them, but a promoter ran an advertisement showing them among diamonds, emeralds, and other precious stones. He also showcased a string of black pearls at a jewelry shop, with an outrageous price tag. And soon enough, people started buying them at premium prices.

He suggests that our willingness to pay five dollars for Starbuck's coffee rather than a dollar at Dunkin' Donuts is all a matter of stage dressing: an elegant shop, products with Italian names, and the like give us the sense that Starbuck's is more precious and worth five times as much.

He then speaks a bit more about anchors: how we are tricked into buying a large coffee because it is only fifty cents more than the medium size. This goes back to his earlier point about relativity in pricing - if there was only one size we would question whether the product was worth it, but when we have different sizes to compare we see the prices as relative to one another and the "not worth it at all" option is completely ignored.

He mentions the ability to deceive others by depicting something as pleasant. Consider how Tom Sawyer tricked Huckleberry Finn into painting a fence because he pretended it was enjoyable rather than laborious. This is the same as when people follow crowds: if there are a lot of people waiting in line, they assume it's for something good.

This is also self-reinforcing. Because we expected something to be good, we often convince ourselves it actually was good - we would rather prove ourselves to have been right than to make an objective assessment. And when we buy something habitually, each purchase confirms the right-ness of the last one - and we avoid the discomfort of having to admit we were wrong.

The more we spend on something, the more likely we are to feel it was a good choice. This can be seen in the medical field quite often, where an expensive medicine or even a surgical procedure led to patients reporting greater satisfaction - which did not occur in a blind comparison.

Even non-commercial choices like voting are tied to the assumption that we have only two choices and must take the lesser of two evils. Many people report supporting one candidate not because of his merits, but because they dislike the other option.

He then speaks of anchoring: the famous experiment in which students were told to write down the last two digits of their SSN in a mock auction, and found that those who had higher digits also bid higher for the items. Anchoring also allows us to compare a price to an initial expectation - the "sale" event at a store creates mania not because products are worth the sale price, but because there is a discount from the regular price, which was an anchor.

People also become creatures of habit very easily, making the same choices without pausing to think, because thinking is hard work. But if you reflect on a decision, you may realize the error of your ways. Ask yourself:

If you run through these questions you can break bad economic habits and form better ones.

3: The Cost of Zero Cost

The concept of "free" is a false notion that drives people insane. It's a false notion because nothing is ever really free: "buy one get one free" requires you to buy something, and if something is ever given away for free it still requires the consumer to make an effort to get it, and is usually a lure that is expected to create a financial benefit (you feel the need to reciprocate by purchasing something when a retailer gives you a gift or a free sample.)

Another informal experiment: the author set up a stand that sold truffles for fifteen cents and chocolate drops for a penny. A significant number of people paid more to have the truffle because they preferred it. He lowered the truffle to fourteen cents and the drops to free - and very few people bought truffles. Even though the price difference was still the same (fourteen cents more for a truffle), the fact that the drops were free made people choose them. (EN: The flaw in this experiment is that he did not count the number of non-buyers in the first group who chose nothing at all, who chose to take nothing when a price was attached, but took a drop because it had no price in the second experiment.)

To get around the flaw, he mentions another experiment in which he offered children who came to his door at Halloween with a choice: he gave them three chocolate drops and the choice to trade two for a small candy bar or three for a large candy bar. Later, he gave them the choice to trade one drop for a large candy bar or take a small candy bar for free - and the majority of children took the small bar because they didn't have to give up anything to get it.

The offer of a "free" product short-circuits the human mind in the same way that a computer will throw an error when it is asked to divide something by zero. We simply do not have the ability to assign a realistic value to something when its price is nothing.

He also mentions the increase in individual order value at Amazon when they mentioned free shipping for orders over $20 - a screw-up in France had them offer one-Franc shipping (20 cents), and there was virtually no increase in order size. (EN: I looked into it, and this was not a mistake but a reaction to a government ban on calling it free shipping if there was a purchase limit.)

A final observation: people wait a long time in line to get something for free, even if it is something they have no use for, failing to consider the value of their time. The magic of "free" blinds people to the non-monetary costs of obtaining something.

The psychological basis, he speculates, is the fear of loss. When we have to pay for something, even a nominal cost, we are poignantly aware that we are giving something up in order to get it, and worry we might have made better use of the resource we are losing. "Free" eliminates this decision-making.

4: The Value of Social Norms

The author draws a distinction between economic and social exchanges. An economic exchange is one of an explicit trade: you are giving money for good, labor for money, good for good, labor for labor, etc. A social exchange is one in which someone does a kindness for someone else without an explicit or specific demand for something in return.

The examples he gives are offering to pay someone for a mean when they have invited you for dinner, or talking to a date about how much you are paying for her food and entertainment in exchange for sex. Both are considered improper to the point of being offensive.

(EN: Another example, all too common in the design profession, is where an accountant asks a designer to help out with their web site as a favor, and then when the designer asks for the accountant to help with her taxes, he sends a bill. This has happened to a handful of people I know - but then, accountants aren't exactly known for their social graces.)

Social norms still require the return of favor for favor, but it's much more vague - and there is often the negative impact of a friend who seems to always be asking for favors and does little to nothing in return, or the person who reminds you of a small favor they did when asking you to do something that is entirely inappropriate given the "level" of your relationship to them.

Another informal experiment: people were offered various amounts of money for participating in a computer test. Those offered five dollars performed a task 159 times in five minutes; those offered fifty cents performed the task 101 times; and those offered nothing performed the task 168 times - more than either group that was paid. The implication is that people will undertake more effort to do someone a favor than they will if they are being paid.

An example: AARP asked attorneys to offer legal services to senior citizens at a discounted price of $30 and got few takers. When they instead asked them to offer their services for free, a lot of attorneys volunteered. (EN: The flaw in this observation is that pro-bono work is required or at least recognized by some professional associations whereas discounted work does not count. So it's the value of the esteem that is missing here.)

It's suggested that employees report less satisfaction with their benefits when employers inform them of the dollar value. If they do not know the cost, they regard it as an act of kindness and generosity - but if the employer discloses the cost, it is seen as part of their compensation.

In a non-commercial situation, he mentions a daycare that imposed a fine on parents for picking up children late. This actually caused the number of parents who were late to increase, because they saw the fine as a late charge - it costs an extra amount to leave the kid there longer - rather than considering timeliness to be a social norm.

It's also mentioned that people whose jobs are very risky don't do it for the money. Police officers, firefighters, and soldiers are not well paid, and they do it because they take great self-esteem from risking themselves for the sake of others. It may be speculated that if the salaries were increased, you would draw a different pool of candidates who were more mercenary than altruistic.

It's also suggested that companies that attempt to market on social norms, claiming to care about customers, often fare worse than those who are forthright about the nature of the relationship. "You can't treat your customers like family one moment and then treat them impersonally a moment later when this becomes more convenient and profitable" because this creates a certain expectation that you immediately violate, and are seen as untrustworthy and deceitful.

And so, it can be valuable to leverage social norms - people will give more than they would if they were paid to do the same task - but you have to be genuine.

5: The Influence of Arousal

In this chapter, Ariley focuses entirely on an experiment that was done regarding sexual arousal. Separate groups of undergraduates were asked a number of questions regarding their sexual practices: whether they found elderly people, children, or animals to be sexually attractive, whether they might drug a woman to have sex with her, or whether they would refuse to have unprotected sex. Naturally, there was a higher positive response rate when the subjects were sexually aroused at the time the questions were asked.

The same principle applies to other states of arousal: decisions made when we are influenced by anger, irritation, or sexual arousal are often very bad or inappropriate decisions.

Meanwhile, economics as well as much of market research is based on decisions that are made in a calm, cool, and collected state - the unrealistic situation of a laboratory when the subject is comfortable and nothing real is at stake.

For some industries, particularly restaurants, such research can lead them astray. When a person is not hungry, he might be more circumspect in his decisions - but people seldom choose a restaurant until they are already feeling the pangs of hunger.

(EN: Negative emotions are also sometimes exploited - sales calls are actually scheduled when they expect people are having dinner because they want the marks to agree to something when they are in an irritated state, and might agree to something they would have refused in a more collected state, because they want to get back to their meal.)

In a later chapter (EN:" moved here because it is more germane) he mentions the "ice method" of controlling personal spending: freezing credit cards in a block of ice causes people to cool down before they make a purchase. If they wait until the ice has melted, they will have time for their rational faculties to engage, rather than purchasing in the heat of the moment.

6: Procrastination and Self Control

The author speaks of an informal experiment he did in a class in which three papers were due.

The implication is that those who have fixed deadlines are most focused on their work.

Another example is the manner in which people save for retirement - or fail to do so. When people are young, the deposits will earn tremendous interest, but because retirement is so far away they see little point in saving. Meanwhile as people get older and retirement draws near, they tend to make large deposits, even though there is little time for their investments to mature.

He also mentioned the way in which Honda improved the maintenance of its vehicles by setting service needs at specific intervals. Wheras other manufacturers had different maintenance schedules for each task, Honda simply offered service packages that were not optimal from an engineering standpoint, but were more convenient for customers to remember and attend - and as such the vehicles were better maintained and preserved their value better.

The problem is that people have little discipline or self-control and will make financially unwise choices because it is simply more convenient for them. They want rewards fast and payments slow, which explains the reason that credit is constantly in crisis and savings are low in modern American culture.

In non-financial situations: consider that people who wish to diet or exercise to improve their health have better success if they give themselves little rewards along the way to achieve progress. They are more motivated by the little rewards than the major one because they are faster to earn.

There's a brief mention of social reinforcement: telling others about your goals so that they will express disappointment when you fail to keep your commitments can be effective. Weight Watchers and Alcoholics Anonymous leverage the fear of shame in a social group to keep people committed to a long-range goal.

7: The High Price of Ownership

The "endowment effect" maintains that we value the things that we own more than the things that we don't - and as a result we value the things we own more than other people value them.

One reason for this is that we remember the effort it required to obtain them. Consider an experiment a Duke university, whose basketball stadium is insufficient to provide tickets for the entire student body. Those who wish to get tickets must camp out and perform various stunts just to be entered into a lottery - and an informal telephone survey showed that the students would demand $2,400 for the tickets they had "earned" in this manner. Meanwhile, others who wanted the tickets were surveyed to find that they would pay about $170 to have them.

(EN: The same is true of product satisfaction. An example that comes to mind is that instant cake mixes failed terribly when all that had to be added was water. But when reformulated to require them to add an egg, the product became successful because of the "effort" required enabled them to feel like they were doing something to "make" the cake.)

Another reason for the effect is that we recognize the value of what we have in terms of the benefits it provides. Having owned and used it, and even customized it for our specific needs, we perceive it as having higher value that someone who is considering it but has no intimate knowledge. This is the reason that homeowners, particularly those who have customized their houses, demand much higher prices than buyers are willing to offer.

And again, there is the fear of loss that is inherent to any exchange. When we sell a possession, we are not considering its market value against other choices a buyer might have, but are instead focused on the need to be compensated for losing the benefits of ownership we currently have, which we sense on an emotional rather than rational level.

One way to leverage this proclivity is to offer customers with a sample or a free trial, or a period during which they can return an unwanted item with no questions asked. The belief is that once a person has possession of something, they will experience it, and will then fear the loss of it.

Possessiveness is such a strong instinct that it extends before the sale: people bidding in auctions, such as eBay, often feel that the item is "theirs" before they have won the auction, and this causes them to bid up the price even beyond the retail price - it's about protecting something they feel they already own.

(EN: In sales, the take-away has the same effect. Allow a customer to handle a product and then snatch it out of their hands. Their natural inclination is to want to grab it back or ask them to return it, because in that brief moment they have formed the idea that it is "theirs" and are upset at the thought of losing it, even if they don't sense its benefits.)

8: Options: Keeping Doors Open

The chapter opens with the legend of Xiang Yu, a Chinese warlord who is rumored to have burned the ships and smashed the cooking pots when his armies arrived on enemy shores to eliminate the options of turning back of staying in place so that they would be more motivated to seek out and conquer the enemy stronghold.

This reflects that a great deal of stagnation and procrastination comes from the desire to keep options available, even at the cost of not doing things that would cause an option to close, even if that option is less desirable than another.

He mentions some colleagues of his who did an experiment involving a game in which there were three doors, each of which would pay a reward. Players tried out all three and then stuck to the one that gave the highest payouts. But when they modified a game to eliminate a door if it was not chosen in the last twelve turns, players would click those doors to prevent them from expiring, even though they offered lower payouts. Even when they discovered the door that paid the best, they still wanted to keep their options open.

We see the same thing in the way that students balk at choosing a major - it is a highly stressful decision because it's not merely a matter of choosing something appealing, but giving up on all other options. So many students spend time and money for an extra year or two of college because they find it difficult to commit to one option and leave others behind.

The same can be said of marriage in the present day, when there is less of a stigma on the "single life" - to commit to one person means breaking ties with all others, and people are less likely to do so and, by the divorce statistics, less confident of the choice they have made and more attracted to being "free" to have multiple shallow relationships rather than one deep and fulfilling one.

Another interesting aspect of the bewildering array of choices in modern life is to retreat to a system where there is less choice allowed. Becoming devoted to a strict religion or adhering to a bizarre diet is a person's way of eliminating the complexity of choice and submitting to a system that tells them what to do, to save them the effort of deciding and the anxiety of making a wrong choice.

(EN: All of this is nothing new to economics, which considers the "opportunity cost" of choosing not to pursue certain options. The problem arises when the fear of losing options distorts our evaluations - we tend to undervalue opportunities we don't want and overvalue the ones we do, and end up skewing the analysis to make the most emotionally comfortable choice seem like the most financially sound decision as well.)

9: The Effect of Expectations

In psychology, expectations work in the manner of filters: we see, hear, feel, etc. the things we expect to experience simply because we expect to experience them - the mind ignores contrary details and sometimes fabricates other details to support the recognition of what we expect.

Ariely mentions another informal experiment in which he gave students a "taste test" between two beers - when in reality it was the very same commercial beer, though the second was tainted with a few drops of balsamic vinegar. When students were unaware of the difference, most chose the tainted beer; but when they were made aware, they chose the untainted one. This is presumably because the word "vinegar" primes them for an unpleasant taste.

A second experiment claimed to be a taste-test of coffee, but really paid attention to whether students used some unusual ingredients (cloves, nutmeg, cardamom, etc.) to season the coffee. When the condiments were placed in Styrofoam cups, none of the students used them. When placed in "fancy" containers, they were used. Again, it was the expectation created by the presentation that made the unusual condiments appealing.

He observes that in restaurants, dishes are described as being "exotic" or "delicious" in order to make patrons more inclined to like them, even without tasting them. And in the same way, more upscale restaurants also use fancier plates and silverware to enhance the experience of consuming food.

He mentions the cola taste-tests specifically, suggesting that Pepsi usually wins out when the tests are blind, but Coke wins when they know what the brands are - allegedly because of the greater association of Coca-Cola to the memories and associations created from many years of advertising. This was even tested in an MRI, and it was found that when people are told the brand, it activates the memory center of the brain in addition to the taste center.

Another experiment suggested this also works on self-perception. A group of Asian women were given a math test. Before taking the test, one group ansewed some questions about their race, another about their gender - and the group that was caused to think of themselves as "Asian" (stereotypically skilled at math) scored higher than those cause to think of themselves as women (stereotypically poor at math).

10: The Power of Price

A high price is associated to product quality, and the association is so strong that people believe a product to have higher quality simply because it has a higher price.

Ariely presents a few examples of medical studies in which patients who were told the price of medicine reported differing effects according to the price they were told: when told the drug costs $2.50 per dose, "nearly all" subjects reported pain relief. When told it was ten cents, only half reported pain relief.

A separate study of college athletes found that students reported less fatigue when told that the drink was more expensive. The same deception was attempted on students who took a 15-question quiz, and found that the group who was told it was expensive answered 9/15 questions correctly (equal to a control group) while a group who was told it was cheap got 6/5/15 correct - suggesting that the belief in high price did nothing to improve performance, but a belief in low price may have cause performance to decline.

This leads to some speculation about the use of placebos in medical treatment. There is strong evidence that a patient who is lied to about the effectiveness of a drug often shows improvement, even over others who are given actual medicine.

11: Dishonesty, Part 1

To test the degree to which students cheat, the author set up an informal experiment asked students to take a 50-question quiz for which they would be paid 10 cents per correct answer. The set-up was that they would write their answers, then transfer them to a bubble-sheet for scanning.

A control group got 32.6 questions correct, and this was considered a benchmark. Another group got a tainted bubble-sheet that indicated the correct answers, and scored an average of 36.2 questions correctly (it was presumed they cheated to get an additional 3.6 questions right). A third group who were not required to turn in the handwritten answers got 35.9 answers correct, and a fourth who were allowed to destroy both sheets got 36.1 answers correctly. He takes these difference to mean that many honest people will cheat, and that the risk of getting caught is of little influence.

He also observes that people cheat or steal by only a little bit. Students had the opportunity to fill in all the correct answers, but instead only cheated a little bit. He speculates that a person in an office might take a pen for personal use, but would likely be reluctant to steal an entire box of pens. Since the value stolen is negligible, he reckons most people don't engage their ethical system for minor cheating or minor theft.

He presents some statistics on theft, particularly the juxtaposition of insider theft (by employees) at $600 billion per year, and external theft (robbery/burglary) as $525 billion in the same year. The irony of it is that a thief who steals from another form is well aware that what he is doing is criminal, whereas most employees consider themselves to be honest. It's also speculated that there are probably more incidents of employee theft, albeit for smaller amounts.

Another test was done that gave opportunities to cheat - but the first item on the test for one group was to recall ten books they read in high school, and the test for another was to recall the ten commandments. While the first group cheated by about 33%, there was no cheating in the second group - regardless of how many commandments they could recall, just being made to think about morality made them behave more morally.

To eliminate the religious element, he removed the question byt asked some participant to sign a statement indicating that they are aware that their participation in the study falls under the college's honor code - and this also eliminated cheating (even though the college in question did not have an honor code). This suggests that it is not necessarily religious belief, but the mere reminder of a "moral benchmark" of any kind that causes people to consider the ethics of their behavior.

12: Dishonesty, Part 2

There's a return to the example of office theft: a person might make off with a pen from the office for personal use, but would be reluctant to steal a box of pens. But would they be willing to steal a dollar from petty cash in order to purchase a pen for personal use? Highly unlikely. People regard things that are worth money as being different to money itself - and this makes it easier for them to cheat or steal.

Consider an informal experiment done in the dorms of a college (EN: an ivy-league school, so it's presumed the students come from affluent families and have no serious financial problems). When he left a six-pack of soda in the refrigerator, all the drinks vanished within 72 hours. When he left six dollar bills in the refrigerator, none were taken.

He also repeated his earlier experiment, in which students were paid for correctly answering items on a test - most people cheated a little bit, but none cheated to a perfect score. He repeated this, substituting tokens as a reward in the lab, which could be exchanged for cash in a separate location. The amount of cheating was triple, and 16% of participants cheated their way to a perfect score.

He had previously mentioned that people tend to cheat a little bit on business expense reports, perhaps taking a friend to dinner and reporting it as a business expense - but here he suggests that executives who have a secretary prepare the report often cheat quite a lot, adding entirely fictitious expenses for significant amounts (such as expenses from a cross-country trip they never actually took). Being one step removed from cash makes people far more likely to cheat.

He speculates about recent accounting scandals, and suggests that the perpetrators considered themselves to be moral persons. They would not consider robbing thousands of people at gunpoint, but what they did with the ledgers were tantamount to exactly the same thing.

The "white collar" crimes that are committed on paper are considered, even by society, to be less heinous than armed robbery or burgling someone's home, though the economic impact is no different.

13: Social Decisions

Ariely's experiment in conformity takes place in a tavern, where the house is offering samples of four different kinds of beer to patrons in exchange for reviews.

When there were four or fewer people sitting at a table and the order was taken verbally, each of them would order a different beer. When they switched to using order forms, such that the choice was made silently, people very often made the same choices.

He reckons because America is an individualistic society, that people wanted to express their individuality in the beer that they ordered, and further speculates that had this happened in a collectivistic society such as japan or China, that people would be more likely to order what other people are ordering. (EN: This seems like a rather huge leap of logic, especially given that the experiment was rigged to elicit atypical behavior - studying the patterns in normal restaurant orders would be more reasonable ... in which case a person ordering the same thing as another person is not as unusual as he seems to suggest.)

A step further: when people ordered silently, he believes that they were more satisfied with their choices - given the ratings they gave to the beers they sampled in both situations. He believes this to indicate that consumer choices, particularly ones of which other people are aware, are often more influenced by the impression a person wishes to make on others than any consideration of the personal enjoyment they might experience.

In the end, he suggests that the way to be happy with a choice is to make it for yourself - in a restaurant, decide what you want and order it rather than changing your mind because of what others might think of you.

(EN: I'm dropping the last few bits of this chapter - the first of which suggests that it may be possible to use unethical tactics to trick people into doing things that benefit them. Intentional deception is never ethical, and the notion of presuming to know what is better for another person and manipulating them to that end is not only immoral, but revolting.)