When analyzing something, scientists ask two types of questions: “How does it work?” and “How well does it work?” So far in this chapter on memory, you’ve read about how it works: Information enters memory through a series of stages (sensory, short-term, and long-term memory) and is retained in long-term memory in ways that can be understood by semantic network, PDP, or embodied cognition models.

Let’s now turn to questions about how well memory works. How accurate are our memories? What can you do to improve your memory?

Memory is imperfect. People forget things: names, song lyrics, answers to exam questions. The imperfection in these cases is our failure to remember something seen or heard earlier.

This type of memory failure isn’t surprising. Many of the psychological processes you learned about in this chapter—a lack of attentional effort, a confusion of similar sounds, a failure to process information deeply—can cause you to forget material that you have seen. Another type of memory imperfection, however, is surprising: false memory, or memory for things you have not seen.

FALSE MEMORY. False memory occurs when people “remember” events that never happened in the first place. They recall a personal experience—seeing something occur, being in a particular place at a particular time—but history shows that the experience never actually happened.

In a simple demonstration of false memory (Roediger & McDermott, 1995; see Chapter 1’s Try This!), participants first heard a list of words that were semantically related, for example:

Next, they saw a new set of words. Some of the new words were on the original list and others were not. Participants tried to remember which new words had been seen previously, on the original list.

You can envision how it felt to participate in this experiment through a simple exercise. Cover up the list of words above. Now try to answer the following questions:

240

If you’re like most of the participants in this study, you remembered that the list did not contain the words “radio” or “shoe” and that it did contain the word “pie.” And, like most participants, you probably remembered that the list contained the word “sweet.” But look back at the list now and try to find “sweet.” It’s not there! The researchers found that words like “sweet”—words similar in meaning to the original words but not actually presented—were remembered “at about the same level as items actually presented in the middle of the list” (Roediger & McDermott, 1995, p. 806; recall from earlier in the chapter that there is a serial position effect in memory for lists). When trying to remember words from these lists, participants experienced false memory, just as you probably did right now. They remembered something that hadn’t actually occurred.

This research finding shows that human memory processes differ substantially from memory storage in an electronic device, such as the hard drive of your computer. In electronic systems, memory is passive. The electronic device merely records information that is entered into it, places the information into a permanent storage system, and retrieves it when instructed to do so. The process is so simple that the computer couldn’t experience false memory. Human memory processes, however, involve active thinking. When you “input” information (in this case, read a list of words to be remembered), you do not merely store the information that is presented. You think about words and images that go beyond the information given. Being presented with words like “sugar,” “chocolate,” and “tooth” activates additional concepts in your semantic network, such as “sweet” (and perhaps also “dentist”). This capacity for active thought makes us more creative than computers. However, it also makes us prone to memory errors because, when we try to recall the information, a synonym or another word related to those on the list may come to mind.

People exhibit false memory not only for simple information, such as words on a list, but also for personal experiences. Elizabeth Loftus asked adult research participants whether they could remember details from four childhood events (Loftus, 1997). Three of the episodes she presented were real events that actually had happened in the participants’ lives. The fourth, however, was fictional; it was a story that involved the participant getting lost in a shopping mall at age 5. After presenting each episode, Loftus asked participants if they could recall experiencing the event.

If human memory were error-free, the participants would remember the real events and report no memory for the false one. As it turns out, the participants did remember real events most of the time; 68% of participants reported remembering some details about the events that had actually occurred. Surprisingly, more than one-fourth of the participants also reported that they “remembered” experiencing the fictional event (Loftus, 1997).

241

How could human memory be prone to this error? Loftus explains that if someone suggests to you that an event happened (or merely that it might have happened), your imagination goes to work. You form a mental image of the event and imagine your thoughts and feelings in this situation. Even if the event never occurred, this “act of imagination … makes the event seem more familiar and that familiarity is mistakenly related to childhood memories rather than to the act of imagination” (Loftus, 1997, p. 74).

Loftus’s findings have substantial implications not only for psychology, but also for legal proceedings. As an example, in 1994 a 30-year-old man, Steven Cook, charged that 17 years earlier he had been sexually molested by Cardinal Joseph Bernardin of Chicago. The evidence Cook presented in court was his memories. He reported that, while in therapy, he had suddenly recalled the abuse experience after repressing it for more than a decade and a half. Members of a jury might think that this memory proved the cardinal was guilty. Loftus’s research, however, raises the possibility that the memory was false, even if Cook honestly believed that his report was true. In this case, the plaintiff himself came to doubt the accuracy of his own memory. Cook eventually withdrew his case against the cardinal, concluding that his memory was not reliable (Nolan, 1994).

BIASES IN EYEWITNESS MEMORY. A second form of memory error also is relevant to legal proceedings. It is eyewitness memory, which is memory for events that we personally observe (as opposed to events we merely hear about). If you observe a car accident and report the details to police, or witness a crime and pick the perpetrator out of a line-up of suspects, you’re relying on eyewitness memory.

We often believe that eyewitness memories can’t be wrong. “I saw it with my own eyes!” people say to express this belief. But, in fact, these memories can be wrong. After witnessing an event, people may hear information that distorts their eyewitness memory (Figure 6.11). Research by Loftus shows how this can work.

242

In one study, Loftus showed participants a film with eight political demonstrators disrupting instruction in a classroom (Loftus, 1975). Afterward, participants were asked one of two questions: “Was the leader of the four demonstrators who entered the classroom a male?” or “Was the leader of the 12 demonstrators who entered the classroom a male?” Later, the participants were asked to recall how many demonstrators took part in the protest. The questions affected people’s eyewitness memory. Although everyone had seen 8 demonstrators, people who had been asked the question mentioning 12 demonstrators remembered that the number of demonstrators was approximately 9. Those who had been asked the question mentioning 4 demonstrators remembered that there were about 6 of them.

In related research, participants saw a film of a traffic accident and later were asked how fast the cars in the accident had been going (Loftus, 2003). In different experimental conditions, Loftus varied the phrasing of the question about the cars’ speed at impact. She asked some participants how fast the cars were going when they “hit” each other, and others how fast they were going when they “smashed into” each other. Again, the language used in the question altered people’s eyewitness memory (Loftus, 2003). Participants who heard the words “smashed into” as opposed to “hit” remembered the cars as going faster.

THE ACCURACY OF FLASHBULB MEMORY. A third type of memory error involves flashbulb memory, which is vivid memory of unexpected, highly emotional, and significant events. Every once in a while, as a society, we experience an unexpected and emotionally stimulating event: a political assassination, a terrorist attack, or a sudden natural disaster. When people first hear of the event, their emotions are aroused and senses heightened. When people look back on the event, they often feel they can see perfectly, in their mind, not only the event but also where they were when they first heard about it, almost as if a flashbulb had gone off originally and imprinted an indelible image onto their memory.

243

Because they are so vivid, flashbulb memories seem like a form of memory that is not prone to error. You might expect that they are more accurate than ordinary memories of everyday events. But are they?

To find out, researchers began a study of memory immediately after a major “flashbulb” event: the 9/11 terrorist attacks. On September 12, 2001, they asked participants to report their experiences in detail—where they were, whom they were with, what they were doing—when they first heard of the attacks. They also asked participants how confident they were that their memories of the events were accurate. To compare flashbulb memory with ordinary memory, participants also answered these questions with respect to a second event, namely, an everyday activity that had occurred recently.

The researchers then asked participants these same questions 7, 18, and 224 days later (Talarico & Rubin, 2003). This enabled them to study changes in both the accuracy of flashbulb and regular memories, and people’s confidence in that accuracy. Accuracy was measured in terms of the consistency between the original memory and the memory at a later time; if memory at a later time is not consistent with memory only one day after the event, then the later memory is inaccurate.

As you can see in Figure 6.12, over time people’s confidence in their everyday memories declined, but their confidence in their flashbulb memories remained high. Was their relatively high level of confidence in flashbulb memories matched by a high level of accuracy? No, people remembered “flashbulb” events with no greater accuracy than everyday events. Participants’ perception that their flashbulb memories were more accurate than their memory for everyday events was false.

244

CULTURAL OPPORTUNITIES

Much of this chapter has concerned memory’s limitations: Storage in sensory memory is brief; the capacity of short-term memory is limited; some long-term “memories” are not accurate—the memories are false. After learning of the limitations, you might be asking, “Is there anything I can do to improve my memory?”

People have asked this question for thousands of years. In fact, the need to improve memory was greater in the past than today (Danziger, 2008). Before the invention of the printing press, copies of documents, even important ones such as sacred texts and legal proclamations, were scarce. People who needed to know the information in these documents had to memorize it. The demand for memory skills was so high that in ancient Rome, for example, training in memory was part of a formal education.

246

Today, technology does much of our memory work for us. We can access facts and figures on the Internet, so we don’t need to memorize them. Yet we often still do need that “old-fashioned” memory device, the human mind; whether you’re trying to remember a formula during a math exam or someone’s name at a party, you need memory skills. So let’s look at two strategies for enhancing them. The first, chunking, is a technique for packing more information into short-term memory. The second, mnemonics, is a set of techniques for organizing information in long-term memory in a way that makes it easier to retrieve.

CHUNKING. As you’ll recall from earlier in this chapter, short-term memory can hold only a few pieces of information. If somebody reads you a list of 15 names, numbers, or words, you’re likely to remember only about a half-dozen of them.

Chunking is a strategy for increasing the amount of information you can retain in short-term memory. The strategy is to group different pieces of information into one memorable piece, or “chunk” of information. As an example, quickly read the line of letters below, then cover up the line, and see how many you can remember:

A L J F K G W B B H O

If you’re like most people, you remembered a couple letters at the beginning (A, L) due to primacy effects and some at the end (H, O) due to recency effects, but you couldn’t remember half or more of the letters.

Now read the letters again, while trying to group together some letters. Here’s a hint: presidents of the United States. Let’s group them so that the virtue of the hint becomes obvious:

A L    J F K    G W B    B H O

There are four chunks of information. Each one contains the initials of an American president: Abraham Lincoln, John F. Kennedy, George W. Bush, and Barack H. Obama. Each chunk packs two or three pieces of information into one memorable group. By remembering the four chunks, which is easy, you’re able to remember 11 individual letters, which was hard.

An even more effective example of the benefits of chunking involves the following list of numbers. Read it quickly and try to recall the numbers:

1 4 9 1 6 2 5 3 6 4 9 6 4 8 1

Again, you probably recalled just a few, from the beginning and end of the list—unless you noticed the sequence: 1², 2², 3², 4², 5², 6², 7², 8², 9². Once you notice that, the original 15 pieces of information (the 15 numbers) are reduced to one chunk; all you have to do is remember the single fact that the string of numbers is the result of listing the square of integers starting with 1.

Chunking doesn’t expand the storage capacity of short-term memory; your memory holds the same number of “packages” of information. Chunking simply stuffs more information into each package.

The second memory enhancement tool works on a different principle. It enhances memory by providing memorable methods of organizing the information you need to recall.

MNEMONICS. Memory can be enhanced by mnemonics, which are strategies for organizing information in memory. Mnemonic strategies usually add a small amount of information to material that you need to remember; the additional information increases the organization of information, making it easier to recall the material you need (Higbee, 1996). Just as a library cataloging system is useful for finding books in a library, mnemonics are useful for locating information in memory.

One simple type of mnemonic uses acronyms (abbreviations formed from the initial parts of other words). In a trigonometry class, you may have needed to remember that the sine of an angle equals the opposite side of a triangle divided by the hypotenuse, the cosine is the adjacent side divided by the hypotenuse, and the tangent is the opposite side divided by the adjacent side. How are you going to remember all that? You can use the acronym SOHCAHTOA (Sine Opposite Hypotenuse, etc.). The acronym is an extra piece of information to remember, but it’s useful: It organizes information that you otherwise might forget.

247

SOHCAHTOA only helps in trig classes, however. Other mnemonic strategies boost recall of diverse types of information (Higbee, 1996). For example:

248