A few words on subliminal advertising

Another summer of movies has come and gone.  If you are like me, you spent some good quality time in the theater watching a few blockbusters.  Along the way, you may even have stopped by the snack counter for some popcorn, candy, or a drink.   

Thinking about the movies and snacks often gets people thinking about subliminal advertising as well.  In the 1950’s, an advertiser suggested they had spliced images of brand names into movies at speeds too fast to be noticed and had influenced people’s purchases.  While this story turned out to be a hoax, many people believe that these kinds of brief presentations can really affect what you buy.

So, what is really going on?

Let’s take this in pieces.  First, it really is possible to present items to people that affect their behavior without awareness.  The word subliminal means “below the threshold,” and refers to items that are presented too fast to be noticed consciously.

If you flash something for one frame of a movie, it is presented for about 1/60^th of a second.  Because there is an image shown before and after it, you will not notice the image consciously, but it will be processed by your visual system, and some information will get through.

The information that gets into the cognitive system makes it easier to think about the concept that was flashed.  Lots of work has shown that subliminal presentations of words will make you faster to respond to other related words.

So, how will this affect choices?

A nice study in the April, 2011 issue of the Journal of Consumer Psychology by Thijs Verwijmeren, Johan Karremans, Wolfgang Stroebe, and Daniel Wigboldus lays out the main factors.  They were interested in choices for drinks.  They measured participants’ level of thirst.  They also determined how much people typically buy two different brands of drinks (one was an ice tea and the other was a bottled water). 

After making these judgments, people were asked to do a simple task in which they saw a row of capital Bs (BBBBBBBB), but on some trials there was also a lower-case b in the row (BBBbBBBB).  They had to count the number of trials where there was a lower-case b.  Before each of the rows of Bs were presented, half the participants saw the brand name of the ice tea flashed on the screen subliminally.

At the end of the study, participants were allowed to select either the ice tea or the water to drink.

The pattern of data is a bit complicated, though it ultimately makes a lot of sense.

When you are not thirsty, the subliminal message has very little effect on your choices.  You tend to pick the drink you generally like.

If you are thirsty, and you have a strong preference for the brand that was shown subliminally, it has no real effect on your choices.

If you are thirsty and you have no real preference for either drink, then you tend to pick the brand that was shown subliminally.

If you are thirsty and your less-preferred drink was shown subliminally, you tend to pick the brand that was shown subliminally (which goes against your habit).

Putting all this together, then, subliminal advertising can have some effects on your choices, though it will not turn you into a robot.  First, subliminal ads only have an effect if you are already motivated to pursue a goal.  So, the subliminal ad will not make you do something you don’t want to do.  Second, subliminal ads have their strongest effect when they make it easier for you to think about something that is not normally your habit.  That is, the subliminal ads tend to favor the underdog.

So go to the movies <follow me on Twitter>, and don’t worry <follow me on Twitter> about ads affecting your trips to the snack bar.

How long will it take to get home?

At the University of Texas, I have two offices: one in my department and one that I use in my role as the director of the masters program in the Human Dimensions of Organizations.  As a result, I often have to walk from one side of campus to the other for meetings.  It is important to do a good job of estimating the amount of time it will take to walk from one place to the other to ensure that I am not late for an appointment, but also that I do not waste time sitting around waiting for a meeting to start.

The walk is familiar enough now, that I know how long it takes, but when it first started, I had to guess.  What factors affect these guesses about how long it takes to go from one location to another?

An interesting paper by Priya Raghubir, Vicki Morwitz, and Amitav Chakravarti in the April, 2011 issue of the Journal of Consumer Psychology looked at how familiarity with the destination affects your judgments of how long a trip will take. 

They started by asking whether people thought it would take longer to go from their home to a location than to go from that location to their home.  They asked students in a classroom either to estimate how long it would take to walk from their home to a classroom or from that classroom back home.  The estimates were about 4 minutes longer to go from home to the classroom than to go home from the classroom.

Why would judgments about home be shorter?

The authors reason that people consider their home to be a bigger location than the destination.  That is, you are familiar with your home and surrounding area, and so when going from home to another place, it will take you longer to feel like you have really left your home area.  Similarly, when you are going from somewhere else to home, you will feel like you are almost there earlier in the trip.

To test this possibility, they performed another study in which people made judgments about a journey between two cities that were several hours’ drive away.  One city was their home city.  The participants were given elaborate driving directions like those you would get from Mapquest or Google Maps.  Of interest, people had to judge when they felt that the journey was under way and when they thought the journey was halfway complete. 

When people were thinking about driving from their home city to another location, they felt like they had to drive further before the journey was really under way or when they were halfway there than when they were driving from another city to their home.  That is, the familiarity of the locations near home made it feel like the early stages of the journey were long.

This work suggests one reason why you often feel like the travel required to go somewhere on a vacation seems to take much longer than the travel required to get back home. 

This phenomenon may also affect your beliefs about the amount of time that it will take to get a project done.  If you are doing a project where the first steps are familiar, you may choose to start early, because you know how much work is required to get started.  If you are doing a project where the ending parts of the project are familiar, though, you may feel confident starting later, because the project will feel like it is nearly completed faster.

Why is the belief in global warming affected by temperature?

Last week, I wrote about a strange finding.  People are more likely to believe in global warming when the weather is warmer than usual for that time of year.  In that blog entry, I summarized research by Ye Li, Eric Johnson, and Lisa Zaval that people’s ratings of their agreement that humans are causing global climate change are higher on unseasonably warm days than on unseasonably cool days.  This belief influences their willingness to perform actions like donating money to charity.

Why does this happen?

As luck would have it, a fascinating paper on this topic by Jane Risen and Clayton Critcher was published in the May, 2011 issue of the Journal of Personality and Social Psychology. 

These researchers explore two possible explanations for the relationship between belief in global warming and temperature.  One possibility is that when it is warm, that makes it easier to think about the concept of heat.  When it is easier to think about heat, then it is also easier to think about global warming, and so people’s belief in global warming goes up.

A second possibility is that when it is warm, it is easier for people to simulate what it would be like when the planet is warmer.  When it is easier to imagine a hot planet, it is easier to believe in global warming.

Now, a lot of times psychology research is criticized for demonstrating things that people already know are true.  So, before I go on and describe some of the studies, which of these possibilities do you think is correct?  (It is possible that they are both true, so don’t feel like you need to pick only one.)

Ok.  The first few studies in this paper just demonstrate the link between temperature and belief.  For example, in one study participants judged their beliefs about many political issues.  On a scale from 0 (low belief) to 10 (high belief).  The judgments were made while sitting in a room heated with a space heater to 81 degrees or in a room at a normal room temperature (73 degrees).  Participants also expressed whether their political ideology was more liberal or more conservative.  Overall, the participants who rated themselves as liberal expressed a greater belief in global warming than those who were conservative.  For both groups, though, their ratings were about a scale point higher when sitting in the hot room than when sitting in the cold room.

A few other studies in this paper explored whether making it easier to think about heat influenced judgments about global warming.  In one study, people did a version of the sentence unscrambling task developed by Bob Wyer and Thom Srull.  In this task, people are given five words like “sizzling bacon is the for” and they have to take four of the words and assemble them into sentences (“The bacon is sizzling”).  One group created sentences that were related to heat, while the control group had no sentences relating to heat.  This technique is well-known to increase how easy it is to think about the concept primed by the sentences.

To demonstrate that people in this study found it easier to think about heat, participants filled in word fragments like F _ A M E, which could be completed to form words relating to heat (like FLAME) or words unrelated to heat (like FRAME).  As expected, when primed with heat, people tended to fill in the word fragments to form heat-related words.  These participants also rated their belief in global warming.  Priming participants with the concept of heat did not affect their belief in global warming.

A third group of studies explored the prospect that heat makes it easier for people to imagine a warmer planet, which in turn influences beliefs in global warming.  In one study, participants sitting in a hot room or a room of normal temperature looked at a series of pictures.  A few of those pictures showed hot dry landscapes.  A few showed images of snow and ice.  The rest of the pictures were not related to climate.  The quality of the pictures was degraded using the transparency feature of a photo editor so that they were between 40 and 60 percent transparent.

Later in the experimental session, participants were shown clear versions of the images they had seen earlier and were allowed to adjust the transparency of the images themselves and were told to match the images to what they had seen before.  The people in the hot room adjusted the images of hot scenes to be clearer than those sitting in a room of normal temperature.  The images of cold scenes were not affected by whether the room temperature was hot or normal.  This result suggests that being hot makes it easier to imagine hot scenes. 

Finally, another group of participants rated their belief in a variety of political issues while looking at pictures relating to those issues.  For most of the issues (like charter schools), degraded quality images were shown using the transparency feature used earlier.  For global warming, one group saw degraded images, while another group saw clear images.  Thus, for the group that saw the clear pictures of hot scenes, the clarity was very salient.  The group that saw clear images rated their belief in global warming as much stronger than the group that saw degraded images.

Putting all of these studies together, it seems that when you are hot, it gets easier to imagine a world that is suffering the effects of global warming, and that increases your belief in global warming.  When it is hot, it may also become easier to think about heat conceptually, but that conceptual ease does not seem to translate into changes in beliefs about global warming.

This work also suggests that when you want to try to persuade someone about the importance of an issue whose effects won’t be felt for years to come, it is important to make the long-term consequences feel as vivid as possible.  Thus, documentaries like Al Gore’s An Inconvenient Truth may be most effective when they provide clear images of what will happen to the Earth as temperatures rise.

On a diet? Leave the credit card at home.

Unpacking groceries after a trip to the store is sometimes a process of discovery.  You have managed to get most of what you went to the store to purchase, but invariably a few items leapt off the shelves and into your cart as a result of impulse purchases.  Most of these impulse purchases are high-calorie high-fat foods like cookies, candies, or cakes.

There are lots of factors that can affect the number of impulse purchases you make in a shopping trip.  For example, there is evidence to support the feeling most of us have that we buy more food on trips to the supermarket when we’re hungry than we do on trips when we have just eaten.

An interesting set of studies by Manoj Thomas, Kalpesh Kaushik Desai, and Satheeshkumar Seenivasan in the June, 2011 issue of the Journal of Consumer Research suggests that your method of payment at the store may also affect your impulse purchases.

They suggest that many people find paying with cash as mildly painful.  That is, actually handing bills and coins to someone else really brings home the nature of spending.  Paying with a credit or debit card is less painful, because the numbers are more abstract.  Because you make impulse purchases in order to feel good, the authors suggest that paying with cash should decrease the number of impulse purchases that people make.

The researchers start by analyzing the purchase patterns of 1000 households at a grocery store.  The data for studies like this comes from people who use loyalty cards at supermarkets.  The loyalty cards allow stores to track your purchases over time.

The researchers separated purchases of foods that are not normally bought impulsively (like vegetables, baby food, and tea) from purchases of unhealthy foods that are often bought on impulse (like cookies and candy).  They found that people who paid with credit or debit cards tended to buy more of these unhealthy foods than those who paid with cash.  Spending on healthy foods did not differ across the methods of payment.

The researchers then did a series of studies in the lab.  In these studies, were shown pictures of 20 foods.  Half were healthy and half were unhealthy.  People were asked to pretend they were at the supermarket and they had to decide whether to purchase each of these products.  Before starting the study, the participants were either shown the logos of four major credit card companies and told that the store accepts credit cards, or they were told that the store only takes cash.

In these studies, participants chose more of the unhealthy items when they could pay with a credit card than when they could only pay with cash.  As with the analysis of the real purchases, there was no difference between groups in the amount of healthy food people chose.

Additional studies used surveys to measure how much pain people usually experience when making purchases with cash.  Those people who are most reluctant to spend cash are the ones who tended to avoid the unhealthy items when paying with cash.  Those people who do not find it at all painful to part with cash bought unhealthy foods at the same rate regardless of the method of payment.

What does this mean for you?  If you are trying to cut down on the amount of unhealthy food you buy, think a bit about your usual spending habits.  If you are stingy with your cash, then you will probably make fewer impulse purchases at the store if you pay with cash than with credit cards.  If so, then leave your credit cards at home when you go to the supermarket.

Finally, if you are worried about the number of impulse purchases you make, always bring a list to the supermarket.  You make fewer impulse purchases when you have a list than when you don’t. 

Global warming, like politics, is local

Texas is unlike most of the world in many ways, but there is one way in which it seems similar.  On a warm spring day in March, when the temperature climbs to near 90 degrees, global warming seems to be the talk of the town.  Of course, 6 weeks earlier when the temperatures in Texas were an unseasonably cold 26 degrees, the typical reaction was to scoff at global warming.

The problem, of course, is that global warming reflects a rise in the average temperature across the globe over the span of years.  The rate of change overall is very slow.  But very few actual days sit on the average temperature.  Instead, there is a lot of variability from day-to-day.  During the winter in Texas, it is not uncommon to see temperatures rise or fall 40 degrees in a day.  

These kinds of anecdotes are fun, but is there really any truth to the idea that people’s belief in global warming is affected by the temperature that day?  This question was explored in a paper by Ye Li, Eric Johnson, and Lisa Zaval in the April, 2011 issue of Psychological Science.

In one study, participants were people in the US and Australia (where it was summer in the northern hemisphere and winter in the southern hemisphere).  People rated the strength of their belief in global warming.  They also rated whether they thought the temperature that day was warmer, colder, or about normal for that time of year.  When people felt the day was warmer than usual, they also expressed a higher belief in global warming than when they felt the day was cooler than usual. They also expressed more concern for global warming when the day was warmer than usual rather than cooler than usual.  The order in which people were asked about global warming and the day’s temperature did not affect the ratings.

Does this really matter for people’s behavior?

In a second study, another group of people performed an internet study for which they were going to be paid for their participation.  Along the way, they answered the same questions about global warming and the day’s temperature.  At the end of the study, participants were asked to donate as much of their payment for the study as they desired to a charity that focuses on cleaner air and prevention of global warming.  The participants in this study donated over four times as much money when the day was much warmer than usual than when the day was much cooler than usual. 

These data really highlight why it is so hard to create policies to deal with long-term problems like global warming.  At a conceptual level, it is easy to make arguments about the importance of many problems like global warming, hunger, and human rights.  However, our behavior is influenced most strongly by our local conditions.  Unless these abstract world problems can be turned into specific issues that affect our world right now, we are unlikely to do much about them.

Unfortunately, for many problems like global warming, by the time we really feel their effects, it will be too late to do anything about them.

After all these years, I know less about you than I thought

The stereotype of an old married couple is one where the partners are constantly finishing each other’s sentences.  They have been together so long, they can predict each other’s orders at restaurants and guess what movies they will want to see.

At the same time, there is a nagging sense that this can’t really be the way long-term relationships affect people.  People who have been married for years will complain about the presents that their spouses buy for them wondering if their long-term partner really knows them at all.

So, how well do couples really understand each other’s likes and dislikes?

This question was explored in a paper by Benjamin Schiebehenne, Jutta Mata, and Peter Todd in the April, 2011 issue of the Journal of Consumer Psychology. 

They compared young couples (generally people in their 20s) who had been together for around 2 years to older couples (people in their late 60s and 70s) who had been together for around 40 years.  Each member of a couple judged preferences for about 40 examples of each of three types of items.  One type (foods) was one that had high daily relevance.  A second type (movies) a common but not daily part of people’s lives.  A third type (kitchen furnishings) was low in daily relevance.  Each person rated how much they liked the items and also how much they thought their partner liked the items.  Finally, each person estimated the number of their predictions for their partner’s preferences that they thought they got correct.

As you might expect, the more frequently something is encountered, the more accurate people’s beliefs about their partner’s preferences.  Overall, people were best able to judge their partner’s preferences for foods.  They were least accurate at judging their partner’s preferences for kitchen designs.  Their judgments about movies came out in between.

The more surprising result is that the younger couples were far more accurate at making these judgments overall than the older couples.  So, couples who had been together for only 2 years were 5-10% more accurate at judging their partner’s preferences than the couples who had been together for 40 years.

The older couples were far more confident in the accuracy of their predictions than the younger couples.  Thus, the older couples believed they knew more about their partner’s preferences than the younger couples even though their actual predictions were less accurate.

Why does this happen?

When studying a topic like this, it is hard to do a true experiment.  You can’t randomly assign people to be in long-term relationships.  So, it is hard to know exactly why couples get less accurate over time.

One thing that may be happening is that younger couples spend a lot of effort trying to learn about each other.  They are paying more careful attention to preferences, because the relationship is new. 

Older couples have already learned a lot about each other.  There are a few factors that may affect the accuracy of their predictions in the long-run.  First, the older couples may not notice changes in their partner’s preferences over time, because they are focused on what their partner liked early in the relationship. 

Second, partners often compromise over things they do, and that can affect beliefs about preferences.  For example, a man might like action movies, while his wife is not so wild about them.  However, she might go to see action movies with him just so that they can watch movies together.  Over time, then, he might think his wife likes these movies more than she actually does.  In this case, her compromise has masked her true preferences.  This compromise often helps couples to share each other’s lives, but it does influence the accuracy of perceptions partners have of each other.

One issue that has not been explored is whether being accurate is valuable.  It is important that the older couples in this study had all been together for about 40 years.  That means that these couples had successfully navigated the difficulties of long-term relationships in a world where 50% of marriages end in divorce.  So, even though we think that knowing our partners accurately is a good thing, it is possible that something about what makes relationships successful in the long-term also decreases our knowledge about what our partners like.

You can talk yourself into anything with time.

Whenever the news reports on a tragedy, there are often amazing reports of heroism that go along with them.  Following the tragic shooting of Representative Gabrielle Giffords and others in a parking lot in Arizona in January, 2011, details emerged about ordinary people who helped subdue the shooter.  After the massive earthquake in Japan in March, 2011, there were many stories and pictures of groups helping to pull victims from buildings that had collapsed.

One reason why it is so incredible to read these stories is that from the comfort of your own home, it is hard to envision whether you would have the courage to help out in the same circumstance.  Yet, when you read interviews with people in these situations, they just report that they did what they felt was the right thing to do at the time.

Of course, one big difference between being at the scene of an event and reading about it later is the amount of time that you have to think about it.  When you witness a crime or experience and earthquake, the situation unfolds quickly.  You have to make a snap judgment about whether to take action.  Reading about it later, you can think carefully about what you think is the right thing to do. 

Joshua Greene and his colleagues suggest that when people are faced with moral dilemmas, there are two different reasoning systems that influence your decisions about what to do.  One is a fast-acting emotion-based system that provides a gut-reaction about how to act.  In many cases that have a moral dimension, this system favors actions that fit with your responsibilities in the moment.  The other is a slower reasoning system that allows you to take broader societal good into account. 

Over the past several years, research on moral reasoning has developed a number of problems that place these two kinds of decisions in opposition.  For example, imagine you go on a cruise in the Caribbean.  An engine on the boat explodes, and the ship starts to sink.  Only a few of the ship’s lifeboats are operational, and people start to climb aboard them.  The boat you are on is so full of people, it is likely to sink, but if you push a few people off the lifeboat, then many more will be saved.   What would you do?

In this case, choosing to push some people off the boat would save many at the expense of a few.  However, pushing some people off the boat means that you have deprived those people of their right to try to survive as well.

A paper in the June, 2011 issue of Cognition by Renata Suter and Ralph Hertwig explored whether people would give different responses to dilemmas like this depending on the amount of time they had to respond.  In one study, people were told that they were going to read about a number of dilemmas.  Some were told that they should go with their first reaction and respond as quickly as possible.  Others were told to think about the dilemma for as long as they needed to and then make a choice.

They found that people who could take as much time as they needed tended to make the choice that would save the most people at the expense of the few.  That is, they would choose to push someone off the boat in order to save everyone else.  In contrast, those who chose quickly, elected not to push people off the boat, despite the risk to everyone else.

These results support the idea that we reason about dilemmas in very different ways depending on the time we have to make the choice.  In the moment, we are driven by a variety of emotions.  At times, our choices in the moment seem humane (trying to save everyone or risking life and limb during a tragedy).  Of course, those same emotions can cause us to lash out in anger at someone or to give into temptation in other circumstances.

When we have the time to justify our choices, we can reduce the influence of our emotions on the eventual choice.  Whether this reason-based system leads to a good choice, though, depends a lot on the system of rules we adopt for making choices.  Given enough time, we can probably talk ourselves into anything.