46 7.3. Problems with Memory

As we have seen, our cultural values and encoding strategies can affect the memories that we store. Similarly, a memory pulled from long-term storage into short-term memory is flexible and subject to change. New events can be added and we can unintentionally alter what we think we remember about past events, resulting in inaccuracies and distortions (Roediger & DeSoto, 2015). Therefore, memories are constructed, rather than accurate action replays of what actually happened.

Memory Errors

Daniel Schacter (2001, 2003; 2020), a well-known memory researcher, offers seven different ways that our memories can (and often do) fail us. He calls them the seven sins of memory and categorizes them into three groups: forgetting, distortion, and intrusion (Table 7.1). Let us focus first on why he suggests that we sometimes forget. One caveat about these “sins” is that they are based on studies with WEIRD, mostly undergraduate samples, we have provided information from other studies when available.

Table 7.1. Schacter’s Seven Sins of Memory







Memory fades over time

Forget events that occurred long ago



Forgetting caused by lapses in attention- encoding failure

Forget where your phone is



Accessibility of information is temporarily blocked

Tip of the tongue



Source of memory is confused

Recalling a dream memory as a waking memory



False memories

Result from leading questions



Memories distorted by current belief system

Align memories to current beliefs



Inability to forget undesirable memories

Traumatic events

Why do we forget?


Most memories are somewhat transient, meaning that they tend to fade over time, especially if they are not activated regularly. In 1885, German psychologist, Hermann Ebbinghaus conducted a study to measure the transience of memory. He was the sole participant in this famous study! First, he memorized long lists of meaningless trigrams (TQM, LMX, etc.) Then he measured how many of them he remembered at various intervals of time for a month (Figure 7.12). The curve in Figure 7.12 shows his rate of forgetting and is called the Ebbinghaus forgetting curve (Ebbinghaus, 1885/1964). Its general shape has been replicated many times in studies of LTM. We can see that initially, there is a rapid, steep loss of information—he forgot more than half the trigrams in the first hour! Then the curve begins to level off, so that after a month, a little bit of information is still retained. You may have had similar experiences if you cram for an exam. You remember very little information a month after the exam.

A line graph has an x-axis labeled “elapsed time since learning” with a scale listing these intervals: 0, 20, and 60 minutes; 9, 24, and 48 hours; and 6 and 31 days. The y-axis is labeled “retention (%)” with a scale of zero to 100. The line reflects these approximate data points: 0 minutes is 100%, 20 minutes is 55%, 60 minutes is 40%, 9 hours is 37%, 24 hours is 30%, 48 hours is 25%, 6 days is 20%, and 31 days is 10%.
Figure 7.12. The Ebbinghaus forgetting curve shows how quickly memory for new information decays.

Absentmindedness and encoding failure

Are you constantly losing your cell phone or your ID? Have you ever gone back home to make sure you turned off the stove or locked the door? Have you ever walked into a room for something, but forgotten what it was? You probably answered yes to at least one, if not all, of these examples—but don’t worry, you are not alone. We are all prone to the memory error that Schacter refers to as absentmindedness, these are lapses in memory when we fail to pay adequate attention to what we were doing. This then makes it difficult to remember what we actually did. We can consider absentmindedness to be a failure of encoding. We know that information only can move into LTM if we pay attention to it and process it deeply. For example, people from individualistic societies are less likely to encode background information in a scene than people from collectivistic cultures, and therefore are less likely to remember that information in future (Wang, 2021).

When we are trying to multi-task, we cannot pay attention to everything and so something that we are doing – like putting our phone down—does not get encoded properly. Other times, we think we have paid attention to something, but we have not processed it deeply enough to remember it. There are many instances of this in our daily lives. For example, you might spend many hours re-reading your notes to study for a test, but when you get to the test, you realize that you only skimmed your notes because you were tired or distracted by other tasks and people. The final section of this chapter gives suggestions for better ways to study. Failing to recognize coins that we use in our daily lives is another example where frequent shallow processing makes us believe that we know something better than we actually do. For instance, think of how many times in your life you’ve seen a U.S. penny. Do you think you can accurately remember what the front of it actually looks like? Nickerson and Adams (1979) found that most Americans are not able to pick out a picture of a real penny from similar pictures of “fake” pennies. Most of us never look closely enough to encode the details of the penny. We only encode enough information to be able to distinguish it from other coins. Try it for yourself with another US coin. See if you can figure out which of the nickels in Figure 7.13 is the real one. Look for the answer at the end of the chapter (at the top of the Reference section).

Four illustrations of nickels have minor differences in the placement and orientation of text.
Figure 7.13. Can you tell which coin, (a), (b), (c), or (d) is the accurate depiction of a US nickel? You will found the answer above the Reference section at the end of this chapter.


Blocking refers to a failure in the retrieval of information that has been successfully encoded and stored in LTM. For example, if someone asks you a question such as – What’s the name of the Black actor who starred in The Shawshank Redemption? You might feel very confident that you know the answer, but you cannot remember it in the moment. This is a very frustrating feeling for most of us, and is often referred to as the “tip of the tongue” (ToT) phenomenon. You feel that the information is right there – just waiting to come out of your mouth. Often you can recall other details about the information that you are searching for. For example, you might be able to clearly picture the actor’s face, or name other movies he starred in, such as Oblivion and the Dark Knight trilogy. You might even remember that his first name begins with an M and that his last name has two syllables. As you activate all of this other information in your memory, more and more parts of the semantic network that includes the actor’s name become active, and then suddenly it pops into your head (Morgan Freeman!) Research suggests that when we experience a strong ToT sensation, it is likely that we are correct in thinking that we know the information, we are just slow in retrieving it. Older adults experience ToT more than younger ones, and it is common across many languages and cultures, including people who use American Sign Language (Schwartz & Metcalfe, 2011). What do you do when you experience ToT? D’Angelo and Humphreys (2015) found that encouraging people to say what they are thinking out loud (e.g., He was in Oblivion and the Dark Knight trilogy – I think his name begins with M) often helped them to remember the ToT information. Or do you (as I often do) Google it, rather than waiting for your brain to provide the answer? It certainly resolves feelings of frustration and allows me to move on with other things, but how does it affect my memory? Psychologists have begun to explore how the Internet affects our memory. Sparrow and colleagues found that people who rely on the Internet are less likely to put effort into remembering information (Sparrow & Chatman, 2013). Also, D’Angelo and Humphreys (2015) showed that when people experienced ToT for newly learned words, they were more likely to experience ToT in the future if they were given the answer, than if they remembered the word by themselves. Therefore, putting effort into trying to retrieve ToT information yourself, probably makes it more memorable in the future.


Another reason why we fail to successfully retrieve familiar information is because other information that we know is interfering with our remembering. There are two types of interference: proactive and retroactive interference (Figure 7.14). In proactive interference, older information that you learned previously prevents you remembering newer information. For example, you might remember your friend’s old phone number, where they used to live, or their previous partner’s name, but you cannot remember their current information. Retroactive interference happens in the opposite direction, newly learned information interferes with the recall of older information. For example, you learn your college email address, but then have difficulty remembering your email address from high school.

A diagram shows two types of interference. A box with the text “learn combination to high school locker, 17–04–32” is followed by an arrow pointing right toward a box labeled “memory of old locker combination interferes with recall of new gym locker combination, ??–??–??”; the arrow connecting the two boxes contains the text “proactive interference (old information hinders recall of new information.” Beneath that is a second part of the diagram. A box with the text “knowledge of new email address interferes with recall of old email address, nvayala@???” is followed by an arrow pointing left toward the “early event” box and away from another box labeled “learn sibling’s new college email address, npatel@siblingcollege.edu”; the arrow connecting the two boxes contains the text “retroactive interference (new information hinders recall of old information.”
Figure 7.14. Sometimes forgetting is caused by a failure to retrieve information. This can be due to interference, either retroactive or proactive.

Memory distortions

It is often hard for us to believe that our memory is not as accurate as we think it is, especially in the case of vivid and detailed flashbulb memories. However, it is important to remember that memory is constructed and so can be influenced by multiple different factors, including our beliefs, expectations, emotions, prior experiences, and interference from external sources. Memory distortions are common in our daily lives, but can have devastating effects when people testify about the details of a crime that they witnessed. People’s freedom and lives often rest on the accuracy of these memories. Unfortunately, eyewitness misidentification is the leading cause for wrongful conviction in the United States (Innocence Project, 2023). Among the 375 people exonerated on the basis of DNA evidence between 1989 and 2020, 69% were convicted on the basis of eye witness testimony (Innocence Project, 2023). Let’s take a look at three common errors of memory distortion: misattribution, suggestibility, and bias, and consider how they affect eyewitness memory and other kinds of memories in our daily lives.

Misattribution/Source confusion

Misattribution happens when you make a mistake about where you got your information—this is sometimes called source confusion or a source-monitoring error. You might remember something—but where did that knowledge come from? Did you read it online? Did someone tell you? Or did you experience it yourself? Sometimes people believe that they witnessed a detail personally, but they actually learned about it by reading, or watching something on TV or on the Internet. Source confusion led to the wrongful arrest of Australian eyewitness expert, Donald Thomson, in 1975.

Australian eyewitness expert, Donald Thomson, appeared on a live TV discussion about the unreliability of eyewitness memory. He was later arrested, placed in a lineup and identified by a victim as the man who had raped her. The police charged Thomson although the rape had occurred at the time he was on TV. Initially, they dismissed his alibi that he was in plain view of a TV audience and in the company of the other discussants, including an assistant commissioner of police. Eventually, the investigators discovered that the rapist had attacked the woman as she was watching TV—the very program on which Thomson had appeared. Authorities eventually cleared Thomson. The woman had confused the rapist’s face with the face that she had seen on TV (Baddeley, 2004, p. 133)

Source confusion can also influence jurors’ decision making abilities. Ruva and Guenther (2015) found that mock jurors often confused negative publicity about a case with actual evidence they had heard during a trial, which affected their verdicts. It is often difficult to remember where exactly we learned something, because we tend to remember general information (the gist), rather than specific details (Schacter, 2003; 2021).

Psychologists have studied the phenomena of misattribution in the lab, using a technique referred to as the Deese-Roediger-McDermott (DRM) paradigm (Roediger & McDermott, 2000). Participants are shown (or read) a list of words—many of which are semantically related to a common theme, e.g., FURNITURE, and so could include words like chair, stool, desk, sofa, cabinet, etc., in addition to unrelated words (e.g., elephant, banana, ocean). Participants are more likely to remember the related words than the unrelated ones, because they are inter-connected within a semantic network. However, they will also often falsely remember a common item of furniture that was not among the words presented. The DRM effect also helps us to understand how people organize information in semantic memory. We frequently build schemas (cognitive structures) for situations that we experience frequently, which help us to quickly recognize places and situations. For example, you likely have a schema for a classroom (students, teacher, desks, board etc.), which is very different from that for a restaurant or a dentist’s office. False memories can be elicited with the DRM paradigm when we activate existing schema— showing it is easy to misremember something if we would normally expect it to be present in a commonly encountered setting.

Suggestibility/Misinformation Effect

Our memories are also vulnerable to suggestion. People around us can influence our memories by asking leading questions or giving us incorrect information. Elizabeth Loftus, a much cited memory researcher, refers to this as the misinformation effect and has conducted multiple studies with WEIRD participants showing how misinformation can generate false memories. Loftus has testified in court many times about the unreliability of eyewitness testimony.

In one well-known study on the misinformation effect, Loftus and Palmer (1974) asked 45 U.S. college students to watch a video of two cars colliding, and then asked them to estimate the speed of the cars involved (Figure 7.15).

Photograph A shows two cars that have crashed into each other. Part B is a bar graph titled “perceived speed based on questioner’s verb (source: Loftus and Palmer, 1974).” The x-axis is labeled “questioner’s verb, and the y-axis is labeled “perceived speed (mph).” Five bars share data: “smashed” was perceived at about 41 mph, “collided” at about 39 mph, “bumped” at about 37 mph, “hit” at about 34 mph, and “contacted” at about 32 mph.
Figure 7.15. Misinformation Effect. When people are asked leading questions about an event, their memory of the event may be altered.

The participants’ answers depended on the format of the question. Participants who were asked: How fast were the cars going when they “smashed” into each other? estimated that the cars were traveling at a much higher speed than participants who heard the verb “contacted” instead of “smashed”. In a follow-up study one week later, they asked the participants whether they saw any broken glass at the scene of the accident (none was actually present in the video). Participants who had been in the “smashed” group were more than twice as likely to indicate that they did remember seeing glass. Loftus and Palmer demonstrated that a leading question can generate false memories. Misinformation can corrupt memory quite easily when it is encountered in social situations (Gabbert et al., 2004). This is a problem particularly in cases where more than one person witnesses a crime, because witnesses often talk to one another as they wait for the police to arrive. However, different witnesses are likely to see or notice different things, so when they talk about the crime, they not only reinforce common memories for the event, they also contaminate each other’s memories (Gabbert et al., 2003; Paterson & Kemp, 2006; Takarangi et al., 2006).

Most studies of the misinformation effect have been conducted with WEIRD participants and point to the dangers of sharing information with others. However, Vredeveldt and de Bruïne (2022) have emphasized the benefits and cultural importance of memory sharing within some non-WEIRD communities. Some communities, particularly those in Rwanda, East Timor, and Sierra Leone particularly value the importance of collective memory, they embrace the fact that multiple versions of an event are pooled together because they believe that they enrich the memory. In such cultures, memory is considered social rather than individual, and contributes to feelings of community.

Link to Learning

In 1984, Jennifer Thompson, a 22-year-old college student in North Carolina, was brutally raped at knife-point. She tried to memorize every detail of her rapist’s face and physical characteristics, vowing that if she survived, she would help get him convicted. However, in two line-ups (photo and in-person) she misidentified an innocent man, Ronald Cotton, as her assailant. Ronald Cotton spent 11 years in jail for a crime he did not commit, until he was finally exonerated on the basis of DNA evidence. There were many factors that contributed to Jennifer Thompson’s misidentification of her rapist.

Watch this first video about Ronald Cotton who was falsely convicted and then watch this second video about the task of his accuser to learn more about why Jennifer Thompson identified the wrong person as her attacker.

Can you think of any other factors that might have influenced Jennifer Thompson’s memory about her attacker?

Ever since Loftus published her first studies on the suggestibility of eyewitness testimony in the 1970s, social scientists, police officers, therapists, and legal practitioners have been aware of the flaws in interview practices. Consequently, steps have been taken to decrease suggestibility of witnesses. One way is to modify how witnesses are questioned. When interviewers use neutral and less leading language, children more accurately recall what happened and who was involved (Goodman, 2006; Pipe, 1996; Pipe et al., 2004). Another change is in how police lineups are conducted. It’s recommended that a blind photo lineup be used. This way the person administering the lineup doesn’t know which photo belongs to the suspect, minimizing the possibility of giving leading cues. Additionally, judges in some states now inform jurors about the possibility of mis-identification and can suppress eyewitness testimony if they deem it unreliable. Imagine that you were a juror listening to an eyewitness identify someone in court as the perpetrator of a crime. How persuasive do you think you would find the evidence?

Controversies over Repressed and Recovered Memories

Loftus and colleagues have conducted several studies where they have used the power of suggestion to implant false autobiographical memories. The researchers used fake materials, which they told the participants had come from their family members. These included fake stories and digitally altered photographs showing their participants in situations that had never occurred, such as meeting Bugs Bunny at Disney Land (Bugs Bunny isn’t a Disney character) or riding in a hot air balloon (Braun et al., 2002; Wade et al., 2002). Loftus has used this research when testifying in court cases to demonstrate the fallibility of memory and its susceptibility to suggestion. Some of these court cases have involved individuals with recovered repressed memories. Repressed memories are a controversial area in psychology. Some psychologists believe that it is possible to bury (repress) deeply traumatic memories, which can be recovered many years later during therapy sessions.  They argue that some children’s experiences were so traumatizing and distressing that they had to lock those memories away in order to lead some semblance of a normal life. They believe that repressed memories can be buried for decades, but can be accurately recalled through hypnosis and guided imagery techniques (Devilly, 2007). Cheit (2007) has suggested that repressing these memories creates psychological distress in adulthood. The Recovered Memory Project was created so that victims of childhood sexual abuse can recall these memories and allow the healing process to begin (Cheit, 2007; Devilly, 2007).

Repressed memories have received a lot of publicity, particularly as they frequently relate to memories of events (such as sexual abuse) that do not have independent witnesses. Loftus and other memory researchers have challenged the idea that individuals can repress memories of traumatic events from childhood, including sexual abuse, and then recover those memories years later during therapy. Loftus is not denying the occurrence of childhood sexual abuse, but she questions the accuracy of “recovered memories”. She is particularly skeptical of the questioning process used to access these memories, given her research on misinformation effects. Of relevance to her arguments are studies that have demonstrated that young children do not always have reliable memories about where people have touched them. For example, Ceci and Brucks (1993, 1995) asked three-year-old children to use an anatomically correct doll to show where their pediatricians had touched them during an exam. Fifty-five percent of the children pointed to the genital/anal area on the dolls, even though they had not received any form of genital examination.


Schacter (2001) points out that our feelings and view of the world can distort our memory of past events. Here are some examples of different types of bias that are known to affect memory.

Stereotype bias

Our stereotypes about people can affect what we remember about them. For example, racial stereotypes were shown to affect the memory of Asian American and European American research participants when they were asked to remember a name presented with a job (basketball player or politician). Half of the names in the study were commonly associated with African Americans, such as Jamal and Tyrone, and half were typically White names such as Greg and Howard. They found that participants made more mistakes when “African American” names were paired with “politician” and when “White names were paired with “basketball player” (Payne et al., 2004). The study shows that people have better memory for information when it is consistent with their stereotypes about particular groups of people.

Egocentric bias

Egocentric bias is a phenomenon found with WEIRD participants, where they tend remember past events that boost self-esteem, by focusing on more positive events than negative ones. This bias is likely exclusive to individualistic societies, as it is not found among East Asian participants (Wang, 2021).

Hindsight bias

Hindsight bias describes the “I knew it all along” phenomenon. In other words, we often misremember what we knew or believed before an event. For example, you might be watching a murder-mystery and at some point in the movie, you announce that you know who committed the crime. At the end of the movie, however, it turns out that someone else was the murderer. However, you are pretty convinced that despite your earlier incorrect announcement, you really knew who it was – you just didn’t say it out loud. In an Internet study, Pohl and colleagues found that hindsight bias was similar among participants from Asia, Australia and North America (Pohl et al., 2002).


Persistence of memory occurs when highly unpleasant memories of traumatic events intrude into our thoughts in such a way that we are unable to function effectively. An example of this includes frequent involuntary flashbacks that are a common symptom among people with PTSD. Arousal theory explains why emotional memories are more easily encoded than non-emotional ones. In the case of traumatic memories, stress hormones may speed the consolidation of memories, and may contribute to strong re-experiencing of emotions when memories of the trauma are recalled.


People’s memory can also be affected by medical conditions (e.g., stroke, viral infections, Alzheimer’s disease and other forms of dementia, substance abuse, and traumatic head injury)—resulting in a condition known as amnesia. There are two common types of amnesia: anterograde amnesia and retrograde amnesia (Figure 7.16).

A single-line flow diagram compares two types of amnesia. In the center is a box labeled “event” with arrows extending from both sides. Extending to the left is an arrow pointing left to the word “past”; the arrow is labeled “retrograde amnesia.” Extending to the right is an arrow pointing right to the word “present”; the arrow is labeled “anterograde amnesia.”
Figure 7.16. Timeline of retrograde and anterograde amnesia. Memory problems that extend back in time before the injury and prevent retrieval of information previously stored in long-term memory is known as retrograde amnesia. Conversely, memory problems that extend forward in time from the point of injury and prevent the formation of new memories is called anterograde amnesia.


Anterograde Amnesia

People with anterograde amnesia cannot make new explicit (semantic or autobiographical) memories. The hippocampal structures are usually affected, which prevents the successful transfer of information from short-term to long-term memory. Early Alzheimer’s disease is the most common cause of anterograde amnesia (Markowitsch, 2011).

Patient H.M. who had his hippocampal structures removed to treat his epilepsy, had severe anterograde amnesia. He could read the same magazine over and over again but would have no memory of ever reading it before—it was always new to him. H.M. was unable to remember meeting new people after his surgery. If they were introduced to H. M. and then left the room, he would not know them when they returned a few minutes later. However, he, like other people with hippocampal damage, was still able to form new procedural memories, even though he had no explicit memory of doing so (Bayley & Squire, 2002; Corkin, 1968; Squire, 2009).

Retrograde Amnesia

Retrograde amnesia is the loss of explicit memories that occurred prior to damage to the brain. People with retrograde amnesia cannot remember some, or even all of their past episodic memories. People with damage to their hippocampus often also have some degree of retrograde amnesia for memories that have not yet been consolidated. HM had retrograde amnesia that affected memories from the 2-3 years before his surgery.

Link to learning

Watch this video about EP who had a virus that destroyed the hippocampal structures in his brain.

Watch this video about the neuroscience of memory for a brief summary about different aspects of memory and how memories are biologically stored.


Icon for the Creative Commons Attribution 4.0 International License

Introduction to Psychology (A critical approach) Copyright © 2021 by Rose M. Spielman; Kathryn Dumper; William Jenkins; Arlene Lacombe; Marilyn Lovett; and Marion Perlmutter is licensed under a Creative Commons Attribution 4.0 International License, except where otherwise noted.