Subliminal: how your unconscious mind rules your behavior - Leonard Mlodinow 2013

Remembering and Forgetting
The Two-Tiered Brain

A man sets himself the task of portraying the world. Through the years he peoples a space with images of provinces, kingdoms, mountains, bays, ships, islands, fishes, rooms, instruments, stars, horses, and people. Shortly before his death, he discovers that the patient labyrinth of lines traces the image of his face. —JORGE LUIS BORGES

JUST SOUTH OF the Haw River in central North Carolina lies the old mill town of Burlington. It’s a part of the country that is home to blue herons, tobacco, and hot, humid summer nights. The Brookwood Garden Apartments is a typical Burlington complex. A pleasant single-story building made of gray brick, it is situated a few miles east of Elon College, now Elon University, a private school that, with the decline of the mills, came to dominate the town. On one of those hot nights in July 1984, a twenty-two-year-old Elon student named Jennifer Thompson was asleep in bed when a man snuck up to her back door.¹ It was three o’clock in the morning. As her air conditioner hummed and rattled, the man cut Jennifer Thompson’s phone line, busted the lightbulb outside her door, and broke in. The noise was not enough to rouse her from her sleep, but the man’s footsteps inside her apartment eventually did. She opened her eyes and made out the form of someone crouching in the darkness at her side. A moment later the man jumped on her, put a knife to her throat, and threatened to kill her if she resisted. Then, as the intruder raped her, she studied his face, focusing on being able to identify him if she survived.

Thompson eventually tricked the rapist into allowing her to turn on a light and fix him a drink, at which point she escaped, naked, out the back door. She frantically pounded on the door of the next unit. The sleeping occupants didn’t hear her, but the rapist did, and he came after her. Thompson raced across the lawn toward a brick house that had a light on. The rapist gave up and moved on to a nearby building, where he again broke in, and raped another woman. Thompson, meanwhile, was taken to Memorial Hospital, where the police obtained samples of her hair and vaginal fluid. Afterward, they took her to the station, where Thompson recounted her study of the rapist’s face for the police sketch artist.

The next day the tips started pouring in. One pointed to a man named Ronald Cotton, twenty-two, who worked at a restaurant near Thompson’s apartment. Cotton had a record. He had previously pleaded guilty to a charge of breaking and entering and, while a teenager, to sexual assault. Three days after the incident, Detective Mike Gauldin summoned Thompson to headquarters to look at six photos, which he lined up on a table. According to the police report, Thompson studied the photos for five minutes. “I can almost remember feeling like I was at an SAT test,” she said. One of the photos was a shot of Cotton. She picked him out. A few days later, Gauldin presented Thompson with a physical lineup of five men. Each man was asked to step forward, utter a line, then turn and step back. At first unsure whether the rapist was the fourth man or the fifth, Thompson eventually settled on the fifth. Cotton again. According to Thompson, when informed that this was the same man she had identified from the photo lineup, she thought to herself, “Bingo, I did it right.” In court Thompson pointed her finger at Cotton and once more identified him as her rapist. The jury reached a verdict in forty minutes, and the judge sentenced Cotton to life plus fifty years. Thompson said it was the happiest day of her life. She celebrated with champagne.

The first sign that something was amiss, other than the defendant’s denials, came after Cotton, working in the prison kitchen, encountered a man named Bobby Poole. Poole bore a resemblance to Cotton and, therefore, also to the face in the police sketch based on Thompson’s description. What’s more, Poole was in prison for the same crime, rape. Cotton confronted Poole about the Thompson case, but Poole denied any involvement. Luckily for Cotton, Bobby Poole blabbed to another inmate that he had indeed raped Thompson and the other woman. Ronald Cotton had by pure chance run into the actual rapist. As a result of the prison confession, Cotton won a new trial.

At the second trial Jennifer Thompson was asked again if she could identify her rapist. She stood fifteen feet from both Poole and Cotton and looked them over. Then she pointed at Cotton and reaffirmed that he was her rapist. Poole looked something like Cotton, but thanks to the experiences that she had had during the time after the rape—her identifying Cotton in a photo, then in a lineup, then in the courtroom—Cotton’s was the face forever burned into her memory of that night. Instead of becoming a free man, Cotton emerged from his second trial with an even harsher punishment: he got two life sentences.

Seven more years passed. What was left of the evidence from the ten-year-old crime, including a fragment of a single sperm from the perpetrator, languished on a shelf in the Burlington Police Department. Meanwhile, the new technology of DNA testing was making the news, thanks to the double-murder trial of O. J. Simpson. Cotton prodded his attorney to request that the sperm fragment be tested. Eventually, his attorney was able to get the test done. The result proved that Bobby Poole, not Ronald Cotton, had raped Jennifer Thompson.

In the Thompson case, all we know is that the victim misremembered her attacker. We’ll never know how accurately or inaccurately Thompson remembered the other details of her attack because no objective record of the crime exists. But it is difficult to imagine a witness more reliable than Jennifer Thompson. She was bright. She stayed relatively calm during the assault. She studied her attacker’s face. She focused on remembering it. She had no prior knowledge of or bias against Cotton. Yet she fingered the wrong man. That has to be disturbing, for if Jennifer Thompson was mistaken in her identification, perhaps no eyewitness can be trusted to reliably identify an unknown assailant. There’s plenty of evidence to suggest that this is the case—some of it from the very people who organize lineups like the one that resulted in Cotton’s arrest.

About seventy-five thousand police lineups take place each year, and statistics on those show that 20 to 25 percent of the time witnesses make a choice that the police know is incorrect. They can be sure of this because the witnesses have chosen one of the “known innocents” or “fillers” that the police inserted into the lineup simply to fill it out.² These are often police detectives themselves, or inmates plucked from the local jail. Such false identifications don’t get anyone in trouble, but think about the implications: the police know that a fifth to a quarter of the time a witness will identify an individual who they are certain did not commit the crime, yet when a witness fingers the person who is their suspect, the police—and the courts—assume that that identification is reliable. As the above statistics reveal, it’s not. In fact, experimental studies in which people are exposed to mock crimes suggest that when the true culprit is not in the lineup, more than half the time eyewitnesses will do exactly what Jennifer Thompson did: they will choose someone anyway, selecting the person who best matches their memory of the criminal.³ As a result, false eyewitness identification seems to be the leading cause of wrongful conviction. An organization called the Innocence Project, for example, found that of the hundreds of people exonerated on the basis of postconviction DNA testing, 75 percent had been imprisoned because of inaccurate eyewitness identification.⁴

You would think that such findings would result in a massive overhaul of the process and the use of eyewitness identification. Unfortunately, the legal system is resistant to change, especially when the changes are fundamental—and inconvenient. As a result, to this day the magnitude and probability of memory error has gone virtually unnoticed. Certainly the law occasionally pays lip service to the fact that eyewitnesses can be mistaken, but most police departments still rely heavily on lineups, and you can still convict someone in court solely on the eyewitness testimony of a stranger. In fact, judges often prohibit the defense from introducing testimony about the scientific research on the flaws of eyewitness identification. “Judges say it’s either too complicated, abstract, and unconnected for jurors to understand, and other times they say it’s too simplistic,” says Brandon Garrett, the author of a book called Convicting the Innocent.⁵ The courts even discourage jurors who are deliberating from using the trial transcript to aid their memory of the testimony they heard in court. The state of California, for example, recommends that judges inform juries that “their memories should prevail over the written transcript.”⁶ Lawyers will tell you there are practical reasons for that policy—for instance, that deliberations would take too long if jurors pored over the trial transcripts. But to me, that seems outrageous, like saying we should believe someone’s testimony about an incident rather than a film of the incident itself. We’d never settle for such thinking in other areas of life. Imagine the American Medical Association telling doctors not to rely on patients’ charts. “Heart murmur? I don’t remember any heart murmur. Let’s take you off that medication.”

IT’S RARE TO have proof of what actually happened, so in most cases we’ll never know how accurate our memories really are. But there are exceptions. In fact, there is one example in which those who study memory distortion were provided with a record that couldn’t have been surpassed had they orchestrated the incident themselves. I’m referring to the Watergate scandal of the 1970s. That scandal concerned a break-in by Republican operatives at the headquarters of the Democratic National Committee and the subsequent cover-up by the administration of President Richard Nixon. A fellow named John Dean, the White House counsel to Nixon, was deeply involved in orchestrating the cover-up, which eventually led to Nixon’s resignation. Dean was said to have an extraordinary memory, and as millions around the world watched on live television, he testified at hearings held by the United States Senate. In his testimony, Dean recalled incriminating conversations with Nixon and other principals in such great detail that he became known as the “human tape recorder.” What endows Dean’s testimony with scientific importance is the fact that the Senate committee later discovered that there was also a real tape recorder listening in on the president: Nixon was secretly recording his conversations for his own later use. The human tape recorder could be checked against reality.

The psychologist Ulric Neisser did the checking. He painstakingly compared Dean’s testimony to the actual transcripts and cataloged his findings.⁷ John Dean, it turns out, was more like a historical novelist than a tape recorder. He was almost never right in his recollections of the content of the conversations, and he was usually not even close.

For example, on September 15, 1972—before the scandal engulfed the White House—a grand jury concluded its investigation by handing down indictments against seven men. They included the five Watergate burglars but only two of the people involved in planning the crime, and they were the “small fish”—Howard Hunt and Gordon Liddy. The Justice Department said it had no evidence on which to indict anyone higher up. That seemed to be a victory for Nixon. In his testimony, Dean had this to say about the president’s reaction:

Late that afternoon I received a call requesting me to come to the President’s Oval Office. When I arrived at the Oval Office I found Haldeman [Nixon’s chief of staff] and the President. The President asked me to sit down. Both men appeared to be in very good spirits and my reception was very warm and cordial. The President then told me that Bob—referring to Haldeman—had kept him posted on my handling of the Watergate case. The President told me I had done a good job and he appreciated how difficult a task it had been and the President was pleased that the case had stopped with Liddy. I responded that I could not take credit because others had done much more difficult things than I had done. As the President discussed the present status of the situation I told him that all I had been able to do was to contain the case and assist in keeping it out of the White House. I also told him there was a long way to go before this matter would end and that I certainly could make no assurances that the day would not come when this matter would start to unravel.

On comparing this meticulous account of the meeting to the transcript, Neisser found that hardly a word of it was true. Nixon didn’t make any of the statements Dean attributed to him. He didn’t ask Dean to sit down; he didn’t say that Haldeman had kept him posted; he didn’t say that Dean had done a good job; and he didn’t say anything about Liddy or the indictments. Nor did Dean say any of the things he attributed to himself. In fact, not only did Dean not say that he “could make no assurances” that the matter wouldn’t start to unravel, he actually said pretty much the opposite, reassuring Nixon that “nothing is going to come crashing down.” Of course, Dean’s testimony sounds self-serving, and he might have been intentionally lying about his role. But if he was lying, he did a poor job of it, because, on the whole, his Senate testimony is just as self-incriminating as the actual, though very different, conversations revealed by the transcripts. And in any case, what is most interesting are the little details, neither incriminating nor exonerating, about which Dean seemed so certain, and was so wrong.

Perhaps you are thinking that the distortions so frequent in the memories of those who were the victims of serious crimes (or those who, like Dean, were trying to cover up such crimes) don’t have much to do with your everyday life, with how well you remember the details of your personal interactions. But memory distortions occur in everyone’s life. Think, for example, about a business negotiation. The various parties to the negotiation go back and forth, over the course of some days, and you are sure that you remember both what you and what the others said. In constructing your memory, however, there is what you said, but there is also what you communicated, what the other participants in the process interpreted as your message, and, finally, what they recalled about those interpretations. It’s quite a chain, and so people often strongly disagree in their recollections of events. That’s why when they are having important conversations, lawyers take notes. Though this doesn’t eliminate the potential for memory lapses, it does minimize it. Unfortunately, if you go through life taking notes on all your interpersonal interactions, chances are you won’t have many.

Cases like those of John Dean and Jennifer Thompson raise the same questions that have been raised, over the years, in thousands of other court cases: What is it about the way human memory works that produces such distortions? And how much can we trust our own memories of day-to-day life?

THE TRADITIONAL VIEW of memory, and the one that persists among most of us, is that it is something like a storehouse of movies on a computer’s hard drive. This is a concept of memory analogous to the simple video camera model of vision I described in the last chapter, and it is just as misguided. In the traditional view, your brain records an accurate and complete record of events, and if you have trouble remembering, it is because you can’t find the right movie file (or don’t really want to) or because the hard drive has been corrupted in some way. As late as 1991, in a survey conducted by the psychologist Elizabeth Loftus, most people, including the great majority of psychologists, still held this traditional view of memory: that whether accessible or repressed, clear or faded, our memory is a literal recorder of events.⁸ Yet if memories were indeed like what a camera records, they could be forgotten or they could fade so that they were no longer clear and vivid, but it would be difficult to explain how people—like Thompson and Dean—could have memories that are both clear and vivid while also being wrong.

One of the first scientists to realize that the traditional view does not accurately describe the way human memory operates had his epiphany after a case of false testimony—his own. Hugo Münsterberg was a German psychologist.⁹ He hadn’t started out intending to study the human mind, but when he was a student at the University of Leipzig he attended a series of lectures by Wilhelm Wundt. That was in 1883, just a few years after Wundt had started his famous psychology lab. Wundt’s lectures not only moved Münsterberg, they changed his life. Two years later Münsterberg completed a PhD under Wundt in physiological psychology, and in 1891 he was appointed assistant professor at the University of Freiburg. That same year, while attending the First International Congress in Paris, Münsterberg met William James, who had been impressed by his work. James was then officially the director of the new Harvard Psychological Laboratory, but he wanted to resign from the post to focus on his interests in philosophy. He lured Münsterberg across the Atlantic as his replacement, despite the fact that although Münsterberg could read English he could not speak it.

The incident that inspired Münsterberg’s particular interest in memory occurred a decade and a half later, in 1907.¹⁰ While he was vacationing with his family at the seashore, his home in the city was burglarized. Informed of this by the police, Münsterberg rushed back and took stock of the condition of his house. Later, he was called to testify under oath about what he had found. He gave the court a detailed account of his survey, which included the trail of candle wax he had seen on the second floor, a large mantel clock the burglar had wrapped in paper for transport but then left on the dining room table, and evidence that the burglar had entered through a cellar window. Münsterberg testified with great certainty, for as a scientist and a psychologist, he was trained in careful observation, and he was known to have a good memory, at least for dry intellectual facts. “During the last eighteen years,” Münsterberg once wrote, “I have delivered about three thousand university lectures. For those three thousand coherent addresses I had not once a single written or printed line or any notes whatever on the platform.… My memory serves me therefore rather generously.” But this was no university lecture. In this case, each of the above statements proved to be false. His confident testimony, like Dean’s, was riddled with errors.

Those errors alarmed Münsterberg. If his memory could mislead him, others must be having the same problem. Maybe his errors were not unusual but the norm. He began to delve into reams of eyewitness reports, as well as some early pioneering studies of memory, in order to investigate more generally how human memory functions. In one case Münsterberg studied, after a talk on criminology in Berlin, a student stood up and shouted a challenge to the distinguished speaker, one Professor Franz von Liszt, a cousin of the composer Franz Liszt. Another student jumped to his feet to defend von Liszt. An argument ensued. The first student pulled a gun. The other student rushed him. Then von Liszt joined the fray. Amid the chaos, the gun went off. The entire room erupted into bedlam. Finally von Liszt shouted for order, saying it was all a ruse. The two enraged students weren’t really students at all but actors following a script. The altercation had been part of a grand experiment. The purpose of the exercise? To test everyone’s powers of observation and memory. Nothing like a fake shootout in psych class to liven things up.

After the event, von Liszt divided the audience into groups. One group was asked to immediately write an account of what they had seen, another was cross-examined in person, and others were asked to write reports a little later. In order to quantify the accuracy of the reports, von Liszt divided the performance into fourteen bite-sized components, some referring to people’s actions, others to what they said. He counted as errors omissions, alterations, and additions. The students’ error rates varied from 26 to 80 percent. Actions that never occurred were attributed to the actors. Other important actions were missed. Words were put into the arguing students’ mouths, and even into the mouths of students who had said nothing.

As you might imagine, the incident received a fair amount of publicity. Soon staged conflicts became the vogue among psychologists all over Germany. They often involved, as the original had, a revolver. In one copycat experiment, a clown rushed into a crowded scientific meeting, followed by a man wielding a gun. The man and the clown argued, then fought and, after the gun went off, ran out of the room—all in less than twenty seconds. Clowns are not unheard of in scientific meetings, but they rarely wear clown costumes, so it is probably safe to assume that the audience knew the incident was staged, and why. But although the observers were aware that a quiz would follow, their reports were grossly inaccurate. Among the inventions that appeared in the reports were a wide variety of different costumes attributed to the clown and many details describing the fine hat on the head of the man with the gun. Hats were common in those days, but the gunman had not worn one.

From the nature of these memory errors, and those documented in many other incidents he studied, Münsterberg fashioned a theory of memory. He believed that none of us can retain in memory the vast quantity of details we are confronted with at any moment in our lives and that our memory mistakes have a common origin: they are all artifacts of the techniques our minds employ to fill in the inevitable gaps. Those techniques include relying on our expectations and, more generally, on our belief systems and our prior knowledge. As a result, when our expectations, beliefs, and prior knowledge are at odds with the actual events, our brains can be fooled.

For example, in his own case, Münsterberg had overheard police conversations about the burglar entering through the cellar window and, without realizing it, incorporated that information into his memory of the crime scene. But there was no such evidence, for, as the police later discovered, their initial speculation had been wrong. The burglar had actually entered by removing the lock on the front door. The clock Münsterberg remembered packed in paper for transport had actually been packed in a tablecloth, but, as Münsterberg wrote, his “imagination gradually substituted the usual method of packing with wrapping paper.” As for the candle wax he so clearly remembered having seen on the second floor, it was actually in the attic. When he spotted it, he wasn’t aware of its importance, and by the time the issue came up, he was focused on the strewn papers and other disorder on the second floor, apparently causing him to recall having seen the candle wax there.

Münsterberg published his ideas about memory in a book that became a best seller, On the Witness Stand: Essays on Psychology and Crime.¹¹ In it, he elaborated on a number of key concepts that many researchers now believe correspond to the way memory really does work: first, people have a good memory for the general gist of events but a bad one for the details; second, when pressed for the unremembered details, even well-intentioned people making a sincere effort to be accurate will inadvertently fill in the gaps by making things up; and third, people will believe the memories they make up.

Hugo Münsterberg died on December 17, 1917, at age fifty-three, after suffering a cerebral hemorrhage and collapsing while delivering a lecture to a class at Radcliffe.¹² His ideas on memory, and his pioneering work in applying psychology to law, education, and business, had made him famous, and he’d counted as friends notables like President Theodore Roosevelt and the philosopher Bertrand Russell. But one person Münsterberg did not consider to be a friend in his later years was his onetime sponsor and mentor, William James.¹³ For one, James had become fascinated with psychics, communication with the dead, and other mystical activities, which Münsterberg and many others considered to be pure quackery. For another, James, if not a convert to psychoanalysis, had at least followed Freud’s work with interest and saw value in it. Münsterberg, on the other hand, was blunt about his view of the unconscious, writing, “The story of the subconscious mind can be told in three words: there is none.”¹⁴ In fact, when Freud visited Boston in 1909 to speak—in German—at Harvard, Münsterberg showed his disapproval by remaining conspicuously absent.

Between them, Freud and Münsterberg had come up with theories of mind and memory that were of great importance, but unfortunately the men had little impact on each other: Freud understood much better than Münsterberg did the immense power of the unconscious, but he thought that repression, rather than a dynamic act of creation on the part of the unconscious, was the reason for the gaps and inaccuracies in our memory; while Münsterberg understood much better than Freud did the mechanics and the reasons for memory distortion and loss—but had no sense at all of the unconscious processes that created them.

HOW COULD A memory system that discards so much of our experience have survived the rigors of evolution? Though human memory is subject to the distortion of memory reconstruction, if those subliminal distortions had proved seriously detrimental to our ancestors’ survival, our memory system, or perhaps our species, would not have survived. Though our memory system is far from perfect, it is, in most situations, exactly what evolution requires: it is good enough. In fact, in the big picture, human memory is wonderfully efficient and accurate—sufficient to have enabled our ancestors to generally recognize the creatures they should avoid and those they should hunt down, where the best trout streams are, and the safest way back to camp. In modern terms, the starting point in understanding how memory works is Münsterberg’s realization that the mind is continuously bombarded by a quantity of data so vast that it cannot possibly handle all of it—the roughly eleven million bits per second I mentioned in the last chapter. And so we have traded perfect recall for the ability to handle and process that staggering amount of information.

When we hold a baby’s birthday party in the park, we experience two intense hours of sights and sounds. If we crammed all of them into memory, we’d soon have a huge warehouse of smiles, frosting mustaches, and poopy diapers. Important aspects of the experience would be stored amid irrelevant clutter, such as the patterns of color on each mother’s blouse, the small talk made by each dad, the cries and screams of every child present, and the steadily growing number of ants on the picnic table. The truth is, you don’t care about the ants or the small talk, and you don’t want to remember everything. The challenge that the mind faces, and that the unconscious meets, is to be able to sift through this inventory of data in order to retain the parts that actually do matter to you. If the sifting doesn’t occur, you just get lost in the data dump. You see the trees but not the forest.

There is, in fact, a famous study that illustrates the downside of an unfiltered memory, a case study of an individual who had such a memory. The study was performed over the course of thirty years, starting in the 1920s, by the Russian psychologist A. R. Luria.¹⁵ The man who couldn’t forget was a famed mnemonist named Solomon Shereshevsky. Shereshevsky apparently remembered in great detail everything that happened to him. Once Luria asked Shereshevsky to recount their initial meeting. Shereshevsky recalled that they were in Luria’s apartment and described exactly what the furniture looked like and what Luria was wearing. Then he recited without error the list of seventy words that—fifteen years earlier—Luria had read aloud and asked him to repeat.

The downside of Shereshevsky’s flawless memory was that the details often got in the way of understanding. For instance, Shereshevsky had great trouble recognizing faces. Most of us store in memory the general features of the faces we remember, and when we see someone we know, we identify the person by matching the face we’re looking at to a face in that limited catalog. But Shereshevsky’s memory housed a great many versions of every face he had ever seen. To Shereshevsky, each time a face changed its expression or was seen in different lighting, it was a new face, and he remembered them all. So any given person had not one face but dozens, and when Shereshevsky encountered someone he knew, matching that person’s face to the faces stored in his memory meant performing a search of a vast inventory of images to try to find an exact equivalent to what he was seeing.

Shereshevsky had similar problems with language. If you spoke to him, though he could always play back your exact words, he had trouble understanding your point. The comparison with language is apt, because this is another trees-and-forest problem. Linguists recognize two types of language structure: surface structure and deep structure. Surface structure refers to the specific way an idea is expressed, such as the words used and their order. Deep structure refers to the gist of the idea.¹⁶ Most of us avoid the problems of clutter by retaining the gist but freely discarding details. As a result, although we can retain deep structure—the meaning of what was said—for long periods of time, we can accurately remember surface structure—the words in which it was said—for just eight to ten seconds.¹⁷ Shereshevsky apparently had an exact and long-lasting memory of all the details of the surface structure, but those details interfered with his ability to extract the gist of what was being said. His inability to forget the irrelevant became so frustrating that at times he would write things down on paper and then burn the page, hoping his memory of them would also go up in flames. It didn’t work.

Read the following list of words, and please pay careful attention: candy, sour, sugar, bitter, good, taste, tooth, nice, honey, soda, chocolate, heart, cake, eat, and pie. If you read only the first few words carefully and then skimmed the rest because you lack patience and feel silly allowing yourself to be ordered around by a book, please reconsider—it is important. Please read through the list. Study it for half a minute. Now cover the list so you can’t see the words, and keep it covered while you read the next paragraph.

If you are a Shereshevsky you’ll have no trouble recalling all the words on the list, but chances are, your memory works a bit differently. In fact, I have given the little exercise I am about to give you to a dozen groups over the years, and the result is always the same. I’ll tell you the punch line after I explain the exercise. It is simple: just identify which of the following three words appeared on the above list: taste, point, sweet. Your answer doesn’t have to be just one word. Perhaps all of them were listed? Or none of them? Please give this some thought. Assess each word carefully. Can you picture seeing it on the list? Are you confident? Don’t choose a word as being on the list unless you are sure of it and can picture it there. Please settle on your answer. Now please uncover the list in the previous paragraph and see how you did.

The vast majority of people recall with great confidence that “point” was not on the list. The majority also recall that “taste” was. The punch line of the exercise has to do with the other word: “sweet.” If you recalled seeing that word, it is an illustration of the fact that your memory is based on your recollection of the gist of the list you saw and not the actual list: the word “sweet” was not on the list, but most of the words on the list were related thematically to the concept of sweetness. The memory researcher Daniel Schacter wrote that he gave tests like this to many audiences and the great majority of people claimed that “sweet” was on the list, even though it was not.¹⁸ I have also given this test to many large groups, and while I did not find a great majority remembering that “sweet” was on the list, I did consistently get about half of my audience claiming it was—about the same number who correctly recalled that “taste” was on it. That result was consistent across many cities and countries. The difference between my results and Schacter’s may stem from the way I phrase the question—for I always stress that people should not designate a word unless they are sure, unless they can picture the list and vividly see that the word is on it.

Our process of remembering can be said to be analogous to the way computers store images, except that our memories have the added complexity that the memory data we store changes over time—we’ll get to that later. In computers, to save storage space, images are often highly “compressed,” meaning that only certain key attributes of the original image are kept; this technique can reduce the file size from megabytes to kilobytes. When the image is viewed, the computer predicts, from the limited information in the compressed file, what the original image looked like. If we view a small “thumbnail”-sized image made from a highly compressed data file, it usually looks very much like the original. But if we blow the image up, if we look closely at the details, we see many errors—blocks and bands of solid color where the software guessed wrong and the missing details were incorrectly filled in.

That’s how both Jennifer Thompson and John Dean got fooled, and it’s essentially the process Münsterberg envisioned: remember the gist, fill in the details, believe the result. Thompson recalled the “gist” of her rapist’s face, and when she saw a man in the lineup of photographs who fit the general parameters of what she remembered, she filled in the details of her memory with the face of the man in front of her, working off the expectation that the police wouldn’t show her a set of pictures unless they had reason to believe the rapist’s photo was among them (though as it turned out, it wasn’t). Similarly, Dean remembered few of the details of his individual conversations, but when he was pressed, his mind filled them in, using his expectations about what Nixon would have said. Neither Thompson nor Dean was aware of those fabrications. And both had them reinforced by repeatedly being asked to relive the events they were remembering, for when we are repeatedly asked to re-create a memory, we reinforce it each time, so that in a way we are remembering the memory, not the event.

You can easily see how this happens in your own life. Your brain, for example, might have recorded in its neurons the feeling of being embarrassed when you were teased by a fourth-grade boy because you brought your favorite teddy bear to school. You probably wouldn’t have retained a picture of the teddy bear, or the boy’s face, or the look on that face when you threw your peanut butter sandwich at him (or was it ham and cheese?). But suppose that years later you had reason to relive the moment. Those details might then have come to mind, filled in by your unconscious. If, for some reason, you returned to the incident again and again—perhaps because in retrospect it had become a funny story about your childhood that people always enjoyed hearing—you most likely created a picture of the incident so indelibly vivid and clear to yourself that you would believe totally in the accuracy of all the details.

If this is so, you may be wondering, then why have you never noticed your memory mistakes? The problem is that we rarely find ourselves in the position that John Dean was in—the position of having an accurate recording of the events we claim to remember. And so we have no reason to doubt our memories. Those who have made it their business to investigate memory in a serious fashion, however, can provide you with plenty of reasons for doubting. For example, the psychologist Dan Simons, ever the scientist, became so curious about his own memory errors that he picked an episode from his own life—his experiences on September 11, 2001—and did something few of us would ever make the effort to do.¹⁹ He investigated, ten years later, what had actually happened. His memory of that day seemed very clear. He was in his lab at Harvard with his three graduate students, all named Steve, when they heard the news, and they spent the rest of the day together, watching the coverage. But Simons’s investigation revealed that only one of the Steves was actually present—another was out of town with friends, and the third was giving a talk elsewhere on campus. As Münsterberg might have predicted, the scene Simons remembered was the scene he’d have expected, based on prior experience, since those three students were usually in the lab—but it wasn’t an accurate picture of what happened.

THROUGH HIS LOVE of case studies and real-life interactions, Hugo Münsterberg advanced the frontiers of our understanding of how we store and retrieve memories. But Münsterberg’s work left open a major issue: How does memory change over time? As it turned out, at about the same period when Münsterberg was writing his book, another pioneer, a laboratory scientist who, like Münsterberg, swam against the Freudian tide, was studying the evolution of memory. The son of a shoemaker from the tiny country town of Stow-on-the-Wold in England, the young Frederic Bartlett had to take over his own education when the town’s equivalent of a high school closed.²⁰ That was in 1900. He did the job well enough that he ended up an undergraduate at Cambridge University, where he remained for graduate school; he eventually became the institution’s first professor in the new field of experimental psychology. Like Münsterberg, Bartlett did not go into academia planning to study memory. He came to it through an interest in anthropology.

Bartlett was curious about the way culture changes as it is passed from person to person, and through the generations. The process, he thought, must be similar to the evolution of an individual’s personal memories. For example, you might remember a crucial high school basketball game in which you scored four points, but years later, you might remember that number as being fourteen. Meanwhile, your sister might swear you spent the game in a beaver costume, dressed as the team’s mascot. Bartlett studied how time and social interactions among people with differing recollections of events change the memory of those events. He hoped, through that work, to gain an understanding of how “group memory,” or culture, develops.

Bartlett imagined that the evolution of both cultural and personal memories resembles the whisper game (also called the telephone game). You probably recall the process: the first person in a chain whispers a sentence or two to the next person in the chain, who whispers it to the next person, and so on. By the end, the words bear little resemblance to what was said at the beginning. Bartlett used the whisper game paradigm to study how stories evolve as they pass from one person’s memory to the next. But his real breakthrough was to adapt the procedure to study how the story can evolve over time within an individual’s memory. Essentially, he had his subjects play the whisper game with themselves. In his most famous work, Bartlett read his subjects the Native American folktale “The War of the Ghosts.” The story is about two boys who leave their village to hunt seals at the river. Five men in a canoe come along and ask the boys to accompany them in attacking some people in a town upriver. One of the boys goes along, and, during the attack, he hears one of the warriors remark that he—the boy—had been shot. But the boy doesn’t feel anything, and he concludes that the warriors are ghosts. The boy returns to his village and tells his people about his adventure. The next day, when the sun rises, he falls over, dead.

After reading the story to his subjects, Bartlett asked them to remind themselves of the tale after fifteen minutes, and then at irregular intervals after that, sometimes over a period of weeks or months. Bartlett studied the way that his subjects recounted the stories over time, and he noted an important trend in the evolution of memory: there wasn’t just memory loss; there were also memory additions. That is, as the original reading of the story faded into the past, new memory data was fabricated, and that fabrication proceeded according to certain general principles. The subjects maintained the story’s general form but dropped some details and changed others. The story became shorter and simpler. With time, supernatural elements were eliminated. Other elements were added or reinterpreted so that “whenever anything appeared incomprehensible, it was either omitted or explained” by adding content.²¹ Without realizing it, people seemed to be trying to alter the strange story into a more understandable and familiar form. They provided the story with their own organization, making it seem to them more coherent. Inaccuracy was the rule, and not the exception. The story, Bartlett wrote, “was robbed of all its surprising, jerky and inconsequential form.”

This figurative “smoothing out” of memories is strikingly similar to a literal smoothing out that Gestalt psychologists in the 1920s had noted in studies of people’s memory for geometric shapes: if you show someone a shape that is irregular and jagged, and quiz them about it later, they’ll recall the shape as being far more regular and symmetrical than it actually was.²² In 1932, after nineteen years of research, Bartlett published his results. The process of fitting memories into a comfortable form “is an active process,” he wrote, and depends on the subject’s own prior knowledge and beliefs about the world, the “preformed tendencies and bias which the subject brings to the task” of remembering.²³

For many years Bartlett’s work on memory was forgotten, though he went on to an illustrious career in which he helped train a generation of British researchers to work in experimental psychology. Today Bartlett’s memory research has been rediscovered, and replicated in a modern setting. For example, the morning after the explosion of the space shuttle Challenger, Ulric Neisser, the man who did the John Dean study, asked a group of Emory University students how they’d first heard the news. The students all wrote clear accounts of their experiences. Then, about three years later, he asked the forty-four students who were still on campus to again recall that experience.²⁴ Not one of the accounts was entirely correct, and about one-quarter of them were entirely wrong. The act of hearing the news became less random and more like the dramatic stories or clichés you might expect someone to tell, just as Bartlett might have predicted. For example, one subject, who’d heard the news while chatting with friends at the cafeteria, later reported how “some girl came running down the hall screaming ’the space shuttle just blew up.’ ” Another, who’d heard it from various classmates in her religion class, later remembered, “I was sitting in my freshman dorm room with my roommate and we were watching TV. It came on a news flash and we were both totally shocked.” Even more striking than the distortions were the students’ reactions to their original accounts. Many insisted that their later memories were more accurate. They were reluctant to accept their earlier description of the scene, even though it was in their own handwriting. Said one, “Yes, that’s my handwriting—but I still remember it the other way!” Unless all these examples and studies are just strange statistical flukes, they ought to give us pause regarding our own memories, especially when they conflict with someone else’s. Are we “often wrong but never in doubt”? We might all benefit from being less certain, even when a memory seems clear and vivid.

HOW GOOD AN eyewitness are you? The psychologists Raymond Nickerson and Marilyn Adams invented a neat challenge. Just think of—but don’t look at—an American penny. It’s an object you might have viewed thousands of times, but how well do you really know it? Can you draw one? Take a moment to try, or at least try to imagine one. What are the main features on each side? When you are done, have a look at the graphic on the following page and try this easier task: pick out the correct penny from among the beautiful sketches Nickerson and Adams kindly provided.²⁵

If you picked A, you would have been in the minority of subjects who chose the correct coin in Nickerson and Adams’s experiment. And if your drawings or imaginings have all eight features of the penny—features such as the profile of Abraham Lincoln on one side, and phrases like IN GOD WE TRUST and E PLURIBUS UNUM—then you are in the top 5 percent in memory for detail. If you did poorly on this test, it doesn’t mean you have a bad memory. Your memory for general features might be excellent. In fact, most people can remember previously viewed photographs surprisingly well, even after a long interval. But they are remembering only general content, not precise form.²⁶ To not store in memory the details of a penny is for most of us an advantage; unless we have to answer a question on a game show with a lot of money at stake, we have no need to remember what’s on a penny, and to do so could get in the way of our remembering more important things.

Reprinted from R. S. Nickerson and M. J. Adams, “Long-Term Memory for a Common Object,” Cognitive Psychology 11, 287—307, copyright 1979, with permission from Elsevier

One reason we don’t retain details of the images that our eyes pick up is that in order for us to remember them, the details first have to have captured our conscious attention. But while the eye delivers a multitude of details, our conscious mind doesn’t register most of them. The disparity between what we see and what we register and, therefore, remember, can be dramatic.

The key to one experiment investigating that disparity was the fact that when you study an image with many objects in it, your eye will shift among the different objects displayed. For example, if an image shows two people seated at a table with a vase on it, you might look at one person’s face, then the vase, then the other person’s face, then perhaps the vase again, then the tabletop, and so on, all in rapid succession. But remember the experiment in the last chapter, in which you stood facing a mirror and noted that there were blanks in your perception during the time your eyes were moving? The researchers who performed this study cleverly realized that if, during the split second their subjects’ eyes were in motion, the image the subjects were looking at changed subtly, the subjects might not notice. Here is how it worked: Each subject started by looking at some initial image on a computer screen. The subject’s eyes would move from object to object, bringing different aspects of the scene into focus. After a while, during one of the subject’s numerous eye shifts, the experimenters would replace the image with one that was slightly altered. As a result, once the subject’s eyes settled on the new target object, certain details of the image were different—for example, the hats the two men in the scene had been wearing were exchanged. The great majority of subjects didn’t notice. In fact, only half the subjects noticed when the two people exchanged heads!²⁷

It’s interesting to speculate how important a detail has to be to register with us. To test if memory gaps like this also happen when the objects that change from shot to shot are the focus of attention, Dan Simons and his fellow psychologist Daniel Levin created videos depicting simple events in which the actor playing a particular character changed from scene to scene.²⁸ Then they recruited sixty Cornell University students, who agreed to watch the videos in exchange for candy. In a typical video, as depicted by the sample frames below, a person sitting at a desk hears a phone ring, gets up, and walks toward the door. The video then cuts to a view of the hallway, where a different actor walks to the telephone and answers it. The change is not as drastic as, say, replacing Brad Pitt with Meryl Streep. But neither were the two actors hard to tell apart. Would the students notice the switch?

Figure provided by Daniel Simons

After viewing the film, the students were asked to write a brief description of it. If they didn’t mention the actors’ change, they were asked directly, “Did you notice that the person who was sitting at the desk was a different person than the one who answered the phone?” About two-thirds of them admitted that they hadn’t noticed. Surely during each shot they were aware of the actor and her actions. But they didn’t retain in their memory the details of her identity. Emboldened by that startling find, the researchers decided to go a step further. They examined whether this phenomenon, called change blindness, also occurred in real-world interactions. This time they took their experiment outdoors, onto the Cornell University campus.²⁹ There, a researcher carrying a campus map approached unsuspecting pedestrians to ask for directions to a nearby building. After the researcher and pedestrian had spoken for ten or fifteen seconds, two other men, each holding one end of a large door, rudely passed between them. As the door passed, it blocked the pedestrian’s view of the experimenter for about one second. During that time, a new researcher with an identical map stepped in to continue the direction-asking interaction while the original researcher walked off behind the door. The substitute researcher was two inches shorter, wore different clothing, and had a noticeably different voice than the original. The pedestrian’s conversational partner had suddenly morphed into someone else. Still, most of the pedestrians didn’t notice, and were quite surprised when told of the switch.

Figure provided by Daniel Simons

IF WE’RE NOT very good at noticing or remembering the details of scenes that occurred, an even more serious issue is recalling something that never happened at all. Remember the people in my audiences who reported seeing in their mind’s eye a vivid picture of the word “sweet” on the list I had presented to them? Those people were having a “false memory,” a memory that seemed real but wasn’t. False memories feel no different than memories that are based in reality. For example, in the many variations of the word list experiment researchers have performed over the years, people who “remembered” phantom words rarely felt they were taking a shot in the dark. They reported recalling them vividly, and with great confidence. In one of the more revealing experiments, two word lists were read to volunteers by two different readers, a man and a woman.³⁰ After the readings, the volunteers were presented with another list, this one containing words they both had and had not heard. They were asked to identify which were which. For each word they remembered hearing, they were also asked whether it had been uttered by the male or the female speaker. The subjects were pretty accurate in recalling whether the man or woman had said the words they’d actually heard. But to the researchers’ surprise, the subjects almost always also expressed confidence in identifying whether it was the man or the woman who had spoken the words they were wrong about having heard. That is, even when the subjects were remembering a word that had not actually been uttered, their memory of its utterance was vivid and specific. In fact, when told in a postexperiment debriefing that they hadn’t really heard a word they thought they had heard, the subjects frequently refused to believe it. In many cases the experimenters had to replay the videotape of the session to convince them, and even then, some of the subjects, like Jennifer Thompson in Ronald Cotton’s second trial, refused to accept the evidence that they were mistaken—they accused the researchers of switching the tape.

The idea that we can remember events that never happened was a key plot element of the famous Philip K. Dick story “We Can Remember It for You Wholesale,” which begins with a man approaching a company to have the memory of an exciting visit to Mars implanted in his brain. As it turns out, planting simple false memories is not that hard, and requires no high-tech solution like the one Dick envisioned. Memories of events that supposedly happened long ago are particularly easy to implant. You might not be able to convince anyone that they have been to Mars, but if your child’s fantasy is a ride in a hot air balloon, research has shown that it is possible to supply that memory with none of the expense or bother of arranging the actual experience.³¹

In one study scientists recruited twenty subjects who had never been in a hot air balloon, as well as one accompanying family member. Each family member secretly provided the researchers with three photos depicting the subject in the midst of some moderately significant event that occurred when the subject was between four and eight years old. They also provided other shots, which the researchers used to create a bogus photo of the subject in a hot air balloon. The photos, both real and faked, were then presented to the subjects, who were not aware of the ruse. The subjects were asked to recall everything they could about the scene depicted by each photo and were given a few minutes to think about it, if needed. If nothing came to them, they were asked to close their eyes and try to picture themselves as they appeared in the photo. The process was repeated two more times, at intervals of three to seven days. When it was over, half the subjects recalled memories of the balloon trip. Some recounted sensory details of the ride. Said one subject after being told the photo was a phony, “I still feel in my head that I actually was there; I can sort of see images of it.… ”

False memories and misinformation are so easy to plant that they have been induced in three-month-old infants, gorillas, and even pigeons and rats.³² As humans, we are so prone to false memories that you can sometimes induce one simply by casually telling a person about an incident that didn’t really happen. Over time, that person may “remember” the incident but forget the source of that memory. As a result, he or she will confuse the imagined event with his or her actual past. When psychologists employ this procedure, they are typically successful with between 15 and 50 percent of their subjects. For example, in a recent study, subjects who had actually been to Disneyland were asked to repeatedly read and think about a fake print advertisement for the amusement park.³³ The copy in the fake ad invited the reader to “imagine how you felt when you first saw Bugs Bunny with your own eyes up close.… Your mother pushing you in his direction so you would shake his hand, waiting to capture the moment with a picture. You needed no urging, but somehow the closer you got, the bigger he got.… He doesn’t look that big on TV, you thought.… And it hits you hard. Bugs, the character you idolized on TV, is only several feet away.… Your heart stops but that doesn’t stop your hands from sweating. You wipe them off just before reaching up to grab his hand.… ” Later, when asked in a questionnaire about their personal memories of Disneyland, more than a quarter of the subjects reported having met Bugs Bunny there. Of those, 62 percent remembered shaking his hand, 46 percent recalled hugging him, and one recalled that he was holding a carrot. It was not possible that such encounters really occurred, because Bugs Bunny is a Warner Brothers property, and Disney inviting Bugs to roam Disneyland is something like the king of Saudi Arabia hosting a Passover Seder.

In other studies people have been led to believe that they had once gotten lost in a shopping mall, been rescued by a lifeguard, survived a vicious animal attack, and been uncomfortably licked on the ear by Pluto.³⁴ They have been made to believe that they once had a finger caught in a mousetrap,³⁵ spilled a punch bowl at a wedding reception,³⁶ and were hospitalized overnight for a high fever.³⁷ But even when memories are entirely fabricated, they are usually based on something true. Kids might be induced into believing they took a ride on a hot air balloon—but the details the child fills in to explain the bogus balloon ride photo percolate from the child’s unconscious, from a body of stored sensory and psychological experiences and the expectations and beliefs that stem from those experiences.

THINK BACK ON your life. What do you remember? When I do that, I find that it is not enough. Of my father, for example, who died more than twenty years ago, my memory holds but meager scraps. Walking with him after his stroke, as he leans for the first time on a cane. Or his glittering eyes and warm smile at one of my then-infrequent visits home. Of my earlier years I recall even less. I remember his younger self beaming with joy at a new Chevrolet and erupting with anger when I threw away his cigarettes. And if I go back still further, trying to remember the earliest days of childhood, I have yet fewer, ever more out-of-focus snapshots: of my father hugging me sometimes, or my mother singing to me while she held me and stroked my hair.

I know, when I shower my children with my usual excess of hugs and kisses, that most of those scenes will not stay with them. They will forget, and for good reason. I would not wish upon them the unforgetting life of a Shereshevsky. But my hugs and kisses do not vanish without a trace. They remain, at least in aggregate, as fond feelings and emotional bonds. I know that my memory of my parents would overflow any tiny vessel formed from merely the concrete episodes that my consciousness recalls, and I hope that the same will be true of my children. Moments in time may be forever forgotten, or viewed through a hazy or distorting lens, yet something of them nonetheless survives within us, permeating our unconscious. From there, they impart to us a rich array of feelings that bubble up when we think about those who were dearest to our hearts—or when we think of others whom we’ve only met, or the exotic and ordinary places we’ve lived in and visited, or the events that shaped us. Though imperfectly, our brains still manage to communicate a coherent picture of our life experience.

In the last chapter we saw how our unconscious takes the incomplete data provided by our senses, fills in what’s missing, and passes the perception to our conscious minds. When we look at a scene we think we are seeing a sharp, well-defined picture, like a photograph, but we really see only a small part of the picture clearly, and our subliminal brains paint in the rest. Our brains use the same trick in memory. If you were designing the system for human memory, you probably would not choose a process that tosses out data wholesale and then, when asked to retrieve it, makes things up. But for the vast majority of us, the method works well, most of the time. Our species would not have survived if that weren’t so. Through evolution, perfection may be abandoned, but sufficiency must be achieved. The lesson that teaches me is to be both humble and grateful. Humble, because any great confidence I feel in any particular memory could well be misplaced; but grateful, both for the memories I retain and the ability to not retain all of them. Conscious memory and perception accomplish their miracles with a heavy reliance on the unconscious. In the chapter that follows, we’ll see that this same two-tier system affects what is most important to us: the way we function in our complex human societies.