CONSTRUCTIVE MEMORY

Memory retrieval is not just a straightforward process but is fundamentally reconstructive,

relying on various strategies like searching for cues and drawing inferences to retrieve original
memory traces and reconstruct experiences. Kolodner (1983) developed an artificial-intelligence
model that illustrates this reconstructive memory process. In practical terms, however, memory is
also constructive. This means that our prior experiences influence both what we recall and how we
recall it (Davis & Loftus, 2007; Grant & Ceci, 2000; Sutton, 2003).

A classic study by Bransford and Johnson (1972) highlights this constructive nature. Participants who
realized a passage was about washing clothes remembered the details far better. In a subsequent
study, Bransford and Johnson (1973) further demonstrated this idea: participants interpreted an
ambiguous passage in distinct ways, depending on the context they believed it described. For
example, participants remembered different details about a passage mentioning an atmosphere that
did not require special clothing, depending on whether they thought it described outer space or a
peace march.

In another domain, Bower, Karlin, and Dueck (1975) studied the constructive aspect of memory with
“droodles”—abstract images that can have multiple interpretations. Participants given an
interpretive label for each droodle were able to reproduce nearly 20% more of them than those in
the control group. This supports the idea that constructive memory helps in organizing and recalling
ambiguous or complex information by linking it to prior knowledge or context.

Autobiographical Memory
Autobiographical memory, which encompasses an individual’s memory of personal history, is
inherently constructive. Instead of recalling events exactly as they occurred, individuals reconstruct
these experiences, often influenced by personal interpretations. Autobiographical memory is
generally reliable, yet subject to distortions. Interestingly, people tend to remember events from
their youth and early adulthood more vividly than more recent experiences (Read & Connolly, 2007;
Rubin, 1982, 1996). Diary studies have been used to explore this further. For instance, Linton (1982)
recorded daily experiences over six years, finding a linear rate of forgetting, an unexpected outcome
as most memory curves show rapid initial forgetting, then slowing. Linton’s study also revealed no
strong link between the emotionality or salience of events and their memorability, highlighting how
certain memories fade regardless of emotional significance.

Another study by Sehulster (1989) examined a researcher’s recollection of opera performances over
25 years, finding better recall for performances at the beginning and end of the period, a pattern
known as the serial-position effect. Additionally, self-esteem has been shown to impact memory
recall: individuals with positive self-esteem tend to remember positive experiences, while those with
negative self-esteem or depression recall more negative events (Christensen, Wood, & Barrett, 2003;
Wisco & Nolen-Hoeksema, 2009). Typically, memory distortions affect minor details rather than the
core aspects of events, which may help prioritize significant information (Bjork et al., 2005;
Goldsmith et al., 2005).

A distinctive type of autobiographical memory is flashbulb memory, which captures intense,

emotionally charged events with vivid detail, as if “preserved on film” (Brown & Kulik, 1977). Events
like the assassination of President Kennedy or the September 11 attacks often trigger flashbulb
memories due to their emotional weight and surprise factor. Conway (1995) suggests that a memory
becomes a flashbulb memory when it is significant, surprising, and emotionally impactful. While
these memories are often confidently recalled, their vividness may result from repeated rehearsal,
either through recounting or silent reflection, which can enhance perceived clarity but not
necessarily accuracy (Bohannon, 1988; Weaver, 1993). Studies of the 9/11 attacks, for instance, show
that forgetting is faster in the first year but stabilizes thereafter. Additionally, flashbulb memories are
prone to distortion; a notable example is that over 70% of people recalled seeing the first plane hit
the tower on 9/11, despite footage only being released the following day (Pezdek, 2003, 2006). This
indicates that while flashbulb memories feel vivid, they are just as reconstructive and subject to
similar encoding and retrieval processes as regular memories (Neisser, 2003; Read & Connolly, 2007).

Biologically, the medial temporal lobe plays a crucial role in recalling autobiographical memories.
Damage to this area, as shown in studies by Kirwan et al. (2008), can impair the recall of recent
memories but often leaves older memories intact, underscoring the importance of specific brain
regions in memory retrieval.

Memory Distortions
People often experience distortions in their memories (Aminoff et al., 2008; Roediger & McDermott,
2000; Schacter & Curran, 2000; Schnider, 2008). Remarkably, simply suggesting that something
occurred can make a person more likely to believe it actually happened, even if it did not (Ackil &
Zaragoza, 1998). Schacter (2001) categorized these common memory distortions into the “seven sins
of memory”:

1. Transience: Memory naturally fades over time. For instance, many people remember that
O.J. Simpson was acquitted of criminal charges, yet few recall exactly how they first learned
about his acquittal.

2. Absent-mindedness: People can be forgetful in daily tasks, like brushing their teeth twice or
entering a room and forgetting what they were looking for.

3. Blocking: Sometimes known as the "tip-of-the-tongue" phenomenon, blocking occurs when

a person cannot retrieve something they know well, like a name or synonym.

4. Misattribution: This happens when people recall information but mix up its source. For
example, in eyewitness testimony, individuals might think they saw something that was
suggested to them but wasn’t actually witnessed.

5. Suggestibility: People are highly susceptible to suggestion. For example, if asked about
seeing a fictitious film of a plane crash, many may falsely remember seeing it.

6. Bias: Personal biases can color memory recall. Someone experiencing chronic pain, for
example, may be more inclined to remember past pain than someone who is pain-free,
regardless of their actual past experiences.

7. Persistence: Some memories linger excessively, such as a single notable failure that
overshadows numerous successes, despite its relative insignificance.

These seven tendencies reveal how memory is not a perfect recording device but a dynamic and
often subjective reconstruction, vulnerable to a range of influences and distortions.
The Eyewitness Testimony Paradigm
Eyewitness testimony has been found to play a significant role in wrongful convictions. A survey of
U.S. prosecutors estimated that around 77,000 suspects are arrested each year based on eyewitness
identifications (Dolan, 1995). However, the accuracy of these identifications is often questionable. In
fact, in the first 180 U.S. cases where DNA evidence exonerated individuals who had been wrongfully
convicted, more than three-quarters of those cases involved errors in eyewitness testimony (Wells et
al., 2006). This suggests that eyewitness testimony may be one of the most common causes of
wrongful convictions in the United States (Modafferi et al., 2009).

The proportion of mistaken eyewitness identifications varies widely. Estimates range from as low as a
few percent to greater than 90% (Wells, 1993, p. 554). Even the most conservative estimates suggest
that a significant portion of eyewitness identifications could be mistaken, raising serious concerns
about the reliability of such testimony in the justice system.

An example of the consequences of such errors can be seen in the case of Timothy, who was
convicted in 1986 for the brutal murder of a mother and her two young daughters (Dolan, 1995).
Timothy was sentenced to death and spent over two years on death row, despite the lack of physical
evidence linking him to the crime. The conviction was based on eyewitness testimony that placed
him near the scene of the murder. However, it was later discovered that another man, who
resembled Timothy, was frequently seen in the neighborhood where the victims lived. As a result,
Timothy was granted a second trial and was acquitted. This case highlights the potentially
devastating impact of faulty eyewitness testimony.

What Influences the Accuracy of Eyewitness Testimonies?

Eyewitness testimony is often relied upon in criminal cases, but it poses significant risks of wrongful
conviction when it serves as the sole or primary basis for a conviction. Research has shown that
eyewitness testimony can be deeply flawed, with mistaken identifications leading to serious
miscarriages of justice (Loftus & Ketcham, 1991; Loftus, Miller, & Burns, 1987; Wells & Loftus, 1984).
One of the major issues with eyewitness testimony is its persuasive power in court, particularly when
the eyewitness appears highly confident, even if their recollections lack detail or contain conflicting
responses. People may also mistakenly believe they remember events simply because they have
thought about or imagined them (Garry & Loftus, 1994). As a result, it is estimated that around
10,000 people each year may be wrongfully convicted based on faulty eyewitness identification
(Cutler & Penrod, 1995; Loftus & Ketcham, 1991).

The susceptibility to error in eyewitness testimony is a significant concern. People are particularly
prone to imagining things they didn’t actually see, leading to inaccuracies in their recollections
(Loftus, 1998). This tendency to create or distort memories has been supported by studies on the
nature of memory itself. In one experiment, participants watched a car stop at a stop sign and then
turn right, allegedly hitting a pedestrian. Afterward, they were asked questions, some of which
contained accurate details (e.g., stop sign) and others with incorrect information (e.g., yield sign).
The participants who were exposed to the inconsistent information (yield sign) showed a 34% lower
accuracy in recalling the event correctly compared to those who received consistent questions (stop
sign) (Loftus, Miller, & Burns, 1978). This demonstrates how external influences, such as suggestive
questioning, can distort eyewitness memories.

Furthermore, line-up procedures also contribute to errors in eyewitness testimony. In line-ups,

witnesses may assume the perpetrator is included, leading them to wrongly identify an individual
who was not the criminal. Research has shown that when the perpetrator is not present in the line-
up, witnesses often select someone else, convinced they have identified the right person (Wells,
1993). Additionally, the identities of the other people in the line-up—known as distractors—can
affect the witness’s judgment. If the distractors resemble the perpetrator or are similar in other
ways, it can increase the likelihood of a false identification (Wells, Luus, & Windschitl, 1994). This
highlights how the composition of a line-up and the assumptions witnesses bring to the process can
impact the accuracy of identifications.

These factors illustrate the complexity of eyewitness testimony and underscore the potential for
errors. From suggestive questioning to the flawed nature of memory itself, multiple factors can lead
to wrongful convictions based on mistaken identifications.

In one study, for example, participants viewed a series of slides depicting a (simulated) automobile
accident. The automobile, a red Datsun, came to either a stop sign (for half the participants) or a
yield sign (for the other half) before becoming involved in an accident with a pedestrian. The
experimental manipulation came in the questioning that followed the slideshow. About half of the
participants (half of whom had seen a stop sign; the other half, a yield sign) were asked, "Did another
car pass the red Datsun while it was stopped at the stop sign?" The other half of the participants
were asked, "Did another car pass the red Datsun while it was stopped at the yield sign?" After
answering these and other apparently routine questions, participants worked on an unrelated
activity for 20 minutes. Then they were given a recognition test of several slides. Included in the test
was a critical test pair depicting a red Datsun stopped either at a stop sign or at a yield sign.
Participants were to decide which of the two slides they had originally seen. Those who received a
question consistent with the slide they had originally seen (for example, a question about the stop
sign when the slide they had previously seen contained a stop sign, not a yield sign) correctly
recognized the slide 75% of the time. Participants who received an inconsistent question, however,
had an overall accuracy rate of 41%, a dramatic decrease given that guessing alone would have
produced an overall accuracy rate of 50%. Other studies by Loftus (1975) have demonstrated that
people's memories can be altered by presenting misleading questions. For example, some
participants viewed a film and were then asked, "How fast was the white sports car going when it
passed the barn while traveling along the country road?" Other participants were merely asked,
"How fast was the white sports car going while traveling along the country road?" Actually, no barn
was presented in the film. One week later, all participants were asked whether they had seen a barn.
Fewer than 3% of the participants in the second condition reported having seen a barn, whereas 17%
of the participants who had been asked the misleading question reported having seen a barn. Lane,
Mather, Villa, and Morita (2001) found that experimental "witnesses" who were asked to focus on
specific details of a videotaped crime were more likely to confuse what they'd witnessed with the
information given to them in post-event questions than were "witnesses" asked only to summarize
the major aspects of the crime.

"Memory malleability" fits well with some laboratory studies of sentence recall, both supporting
Bartlett's conception of memory as a constructive process. A classic study by Bransford, Barclay, and
Franks (1971) illustrates this idea. They gave participants a list of sentences, all derived from four
basic sentences, such as "The ants were in the kitchen," "The jelly was on the table," "The jelly was
sweet," and "The ants ate the jelly." The sentences the participants saw included two of the
preceding sentences, combinations of two of the simple sentences (for example, "The sweet jelly was
on the table"), and combinations of three of the simple sentences (for example, "The ants ate the
sweet jelly on the table"). On a later recognition test, the participants were asked to decide, for each
sentence presented, if they had seen that exact sentence before and to rate their confidence in their
judgment. They were most confident in "recognizing" the sentence that combined all four of the
simple sentences, "The ants in the kitchen ate the sweet jelly that was on the table," even though it
had never been presented.

Confessions can significantly impact the accuracy of eyewitness testimony. A study by Hasel and
Kassin (2009) demonstrated how confessions influence the identification process. Participants
watched a staged robbery and were later presented with a line-up of suspects, although the actual
perpetrator was not included. After some time, participants were informed that one of the suspects
had made a confession. Remarkably, 61% of those who had previously made an identification
changed their choice, and 50% of those who had not initially identified anyone went on to select the
confessor. This shows the profound effect that hearing a confession can have on the eyewitness's
confidence in their identification of a suspect.

Feedback given to eyewitnesses can also alter their testimony, as shown by the post-identification
feedback effect. In an experiment, eyewitnesses who were told they had identified the correct
perpetrator felt more confident in their choice, while those who were informed they had identified
an incorrect person (a "filler") quickly backed away from their judgment (Wells, 2008; Wright &
Skagerberg, 2007). This effect highlights how feedback can distort a witness's confidence and
influence the accuracy of their testimony.

Another factor that weakens eyewitness identification is the racial or ethnic group of the suspect
relative to that of the witness. Research has shown that eyewitnesses are less accurate when
identifying people from a racial or ethnic group other than their own (Bothwell, Brigham, & Malpass,
1989; Brigham & Malpass, 1985; Pezdek, Blandon-Gitlin, & Moore, 2003; Shapiro & Penrod, 1986).
The problem does not lie in remembering faces from other racial or ethnic groups, but rather in the
initial encoding of those faces, suggesting that cross-racial or cross-ethnic identifications are more
prone to error (Walker & Tanaka, 2003).

The accuracy of eyewitness recall and identification is also impacted by the level of stress
experienced by the witness. Research has shown that as stress increases, the accuracy of both recall
and identification declines (Deffenbacher et al., 2004; Payne et al., 2002). Since most crimes occur
under highly stressful conditions, this finding further calls into question the reliability of eyewitness
testimony in real-world settings. These various factors—confessions, feedback, racial or ethnic
biases, and stress—highlight the complexities that can undermine the accuracy of eyewitness
accounts.
Not everyone is skeptical of the reliability of eyewitness testimony. Some researchers, such as
Zaragoza, McCloskey, and Jamis (1987), argue that the issues surrounding eyewitness testimony are
not as clear-cut as they might appear. Specifically, it is still uncertain whether misleading information
presented after an event actually displaces the original memory or if it merely competes with it.
Some psychologists believe that more research is needed to fully understand the circumstances
under which eyewitness testimony can be impaired before making broad claims about its reliability
in court (McKenna, Treadway, & McCloskey, 1992). As a result, the ultimate verdict on the reliability
of eyewitness testimony is still unresolved.

However, it is crucial for everyone involved in legal proceedings to understand the limitations of
eyewitness testimony. While defense attorneys are generally aware of the limitations of eyewitness
accounts, research has shown that prosecutors are less informed about these limitations.
Prosecutors often overestimate the reliability of eyewitness statements and tend to underestimate
their role in contributing to wrongful convictions (Wise et al., 2009). These findings highlight the
importance of educating not only the public but also legal professionals about the potential fallibility
of eyewitness testimony to ensure fairer and more informed legal proceedings.

Children as Eyewitnesses
The reliability of eyewitness testimony is particularly questionable when it comes to children.
Research by Ceci and Bruck (1993, 1995) reveals that children's recollections are highly susceptible to
distortion, especially when they are asked leading or suggestive questions, which are common in
courtroom settings. The younger the child, the less reliable their testimony tends to be. Preschool-
aged children are especially vulnerable to suggestive questioning compared to older children or
adults, which makes their recollections more likely to be influenced by the way questions are asked
or by external pressures. This vulnerability to suggestion underscores the need for caution when
interpreting the testimony of young witnesses.

Additionally, children are more likely to conform to what they believe the questioner wants to hear. If
an adult appears coercive or indicates a desired response, children may feel compelled to provide
the answer they think will please the adult. This tendency can lead children to offer responses that
align with the expectations of the questioner, rather than what they truly remember. Such pressures
can be particularly evident in the stressful environment of a courtroom, where children may feel the
weight of authority figures.

Children are also prone to recalling events based on what others have said, rather than their own
actual observations. When a child hears a story about an incident, they may come to believe they
witnessed it themselves. This effect is more pronounced when there is a delay between the event
and the recall, particularly for children with intellectual disabilities. Such factors can significantly
distort their memories and affect the accuracy of their testimony (Henry & Gudjonsson, 2003).

Furthermore, research shows that the presence of authority figures, such as uniformed officers, can
influence children's eyewitness accuracy. A study in the United Kingdom found that children made
more mistakes in identifying suspects during a line-up when a uniformed officer was present,
suggesting that children may be unduly influenced by the presence of authority figures (Lowenstein
et al., 2010). Given these various factors, children's eyewitness testimony should be treated with
even greater caution than that of adults, as it is highly susceptible to distortion and external
influence.

Can Eyewitness Testimonies Be Improved?

Several steps can be taken to improve the accuracy of eyewitness identification and reduce biases in
the process. Wells (1993) outlines methods that aim to minimize potential biases, decrease the
pressure on witnesses to choose a suspect from a limited set of options, and ensure diversity within
line-ups while still ensuring that each suspect fits the description given by the eyewitness. One of the
main goals is to create a fairer and more accurate process for identifying perpetrators.

Suggestive interviews, which can lead to biases in memory, also pose a problem for eyewitness
testimony. Research by Melnyk and Bruck (2004) highlights that biases are particularly likely when
interviews occur shortly after the event. Since witnesses are typically interviewed as soon as possible
after a crime, it is crucial to ensure that the questions asked are not leading, especially when the
witness is a child. Leading questions can distort memories, so taking care to avoid them can help
preserve the accuracy of the eyewitness's recollection.

Gary Wells (2006) made several recommendations to enhance identification accuracy in line-ups.
One suggestion is to present only one suspect at a time in the line-up, so witnesses do not feel
pressured to choose between multiple potential perpetrators. It is also essential to ensure that all
individuals in the line-up resemble each other in significant ways, reducing the chance that someone
is mistakenly identified based on shared characteristics with the actual perpetrator. Another crucial
recommendation is to caution witnesses that the suspect may not even be in the line-up, which
reduces the pressure to identify someone, even when they are uncertain.

In addition to these procedural changes, some psychologists, such as Loftus (1993a, 1993b), and
defense attorneys believe that jurors should be informed that a witness's confidence in their
identification does not necessarily correlate with the accuracy of that identification. Confidence in an
identification is not a reliable indicator of its accuracy, so jurors should be made aware of this
potential discrepancy. However, some psychologists, like Egeth (1993) and Yuille (1993), as well as
many prosecutors, argue that the evidence against eyewitness testimony, particularly based on
simulated studies rather than real-life cases, is not strong enough to question its credibility in a way
that could risk sending a true criminal free. They express concerns that undermining eyewitness
testimony could lead to guilty individuals not being convicted, thus allowing them to continue
committing crimes. This ongoing debate highlights the complexity of balancing the reliability of
eyewitness testimony with the risks of wrongful convictions.

IN THE LAB OF ELIZABETH LOFTUS

Research on False Memories

Research on the malleability of human memory has shown that people can be influenced by external
suggestions to believe in experiences they never had. One study by Braun, Ellis, and Loftus (2002)
examined how autobiographical advertising, like Disney ads that evoke nostalgia, could create false
memories. Participants who saw an ad suggesting they had shaken hands with Mickey Mouse at
Disneyland later became more confident in recalling this event, even if it had never occurred. This
raised the question of whether such ads triggered a true memory or planted a false one.

To investigate this further, a second study involved a fake ad for Disney that suggested participants
had shaken hands with Bugs Bunny, a character from Warner Brothers, which was impossible
because Bugs Bunny would never be at Disney. Despite this, about 16% of participants later believed
they had met Bugs Bunny at Disneyland as children, providing detailed (though fictitious)
recollections of the encounter, such as touching his ears or hearing his catchphrase, "What’s up
Doc?"

Another study by Berkowitz et al. (2008) aimed to plant a false memory of an unpleasant experience
with the Pluto character. About 30% of participants believed they had had a negative encounter with
Pluto during a childhood trip to Disney. This false memory even influenced their behavior, as these
individuals were less willing to pay for a Pluto souvenir. This demonstrates how false memories can
affect not only thoughts but also behaviors.

In addition, studies have shown that false memories can influence preferences and consumption. For
instance, Bernstein and Loftus (2009) successfully implanted false memories related to food, such as
becoming ill after eating egg salad, which affected how much participants liked or consumed the
food. These studies contribute to a larger body of research that shows how easily human memory
can be manipulated by suggestive influences, such as advertisements. This raises important
questions about the accuracy of our memories and how much of what we recall is shaped by
suggestion, imagination, or other mental processes.

Repressed Memories
The concept of repressed memories, particularly in the context of traumatic childhood events, has
been a topic of debate within psychology. Some therapists believe that individuals may repress
memories of traumatic experiences, such as childhood abuse, pushing them into unconsciousness
due to the distress these memories cause. Techniques like hypnosis are used to help "recover" these
memories, which are thought to be inaccessible but potentially retrievable (Briere & Conte, 1993).
However, there is limited support for the existence of repressed memories, and many psychologists
remain skeptical (Ceci & Loftus, 1994; Pennebaker & Memon, 1996).

The skepticism around repressed memories stems from several factors. First, therapists might
unintentionally influence their clients, planting false memories during the therapeutic process. False
memories can be easily created, even in individuals without any psychological issues, through
suggestive questioning or the use of non-emotional stimuli (Roediger & McDermott, 1995). Second,
proving that memories are false can be incredibly difficult, especially in cases like childhood abuse,
where the issue may become a matter of one person’s word against another’s (Schooler, 1994).
Currently, there is no conclusive evidence proving the existence of repressed memories, though it
has not been definitively ruled out either.

One way that memory distortion is studied in the lab is through the Roediger-McDermott (1995)
paradigm, where participants are presented with a list of 15 words related to a critical word that is
not actually presented. For example, words like "sleep," "bed," and "tired" might be presented, but
the word "sleep" itself is not. Despite this, participants often falsely recognize the non-presented
word "sleep" as if it were included in the list. This phenomenon, known as false recognition, has
been replicated in several studies (McDermott, 1996; Schacter, Verfaellie, & Pradere, 1996), and even
shorter lists can produce this effect, although to a lesser degree (Coane et al., 2007). This false
recognition can be enhanced when the list is embedded in a narrative, especially for young children
(Dewhurst, Pursglove, & Lewis, 2007).

Several cognitive theories help explain these memory errors. One possibility is a source-monitoring
error, which occurs when people misattribute the origin of a memory. For instance, a person might
mistakenly believe they read about an event in a reputable source when they actually saw it in a less
credible outlet. Similarly, in memory recall tasks, people might confuse the source of a word they
remember, thinking it was part of a list when it was actually activated by related words (Foley et al.,
2006; Johnson, 1996, 2002). Another explanation is spreading activation, where exposure to related
words activates other associated words in the memory network. For example, the word "nap" might
activate words like "sleep," "bed," and "cat." This process can lead to false recognition when
participants mistakenly believe that a word like "sleep" was part of the list (Dodd & MacLeod, 2004;
Hancock et al., 2003). Though spreading activation provides a plausible explanation for memory
errors, it is not universally accepted (Meade et al., 2007).

Context Effects on Encoding and Retrieval

Memory retrieval is influenced by various factors, both internal and external. Internal context
includes emotions, moods, and states of consciousness, which all play a significant role in how we
retrieve memories. For example, when information is encoded during a particular mood (e.g., happy
or sad), it is easier to recall when in the same mood again, a phenomenon known as mood-
congruent memory (Baddeley, 1989; Bower, 1983). Similarly, state-dependent memory suggests
that memories encoded while under the influence of substances like alcohol or drugs are more easily
retrieved in the same state (Eich, 1980, 1995). However, moods can also have negative effects on
memory retrieval, especially in people suffering from depression, who may find it easier to recall sad
memories, which can further perpetuate their depression (Baddeley, 1989; Wisco & Nolen-
Hoeksema, 2009). On the other hand, positive moods tend to facilitate the recall of pleasant
memories, while unpleasant or neutral memories may be more difficult to retrieve (Matlin &
Underhill, 1979; Monnier & Syssau, 2008).

External context also affects memory retrieval, as we tend to recall information more effectively
when we are in the same environment in which we encoded it. This was demonstrated in the study
by Godden & Baddeley (1975), where divers who learned words underwater recalled them more
effectively when tested underwater than on land. Similarly, context-dependent memory was
observed in infants, who showed stronger recall of learned behaviors when in the same context in
which they had initially learned them (Butler & Rovee-Collier, 1989; Borovsky & Rovee-Collier, 1990).

The encoding specificity principle emphasizes that memory retrieval is more successful when the
context at retrieval matches the context during encoding (Tulving & Thomson, 1973). This principle
was illustrated in an experiment by Watkins & Tulving (1975), where participants performed better
in cued recall (using stimulus words) than in recognition tests. The stimulus words were better cues
for retrieval because they matched the context in which the words were learned.

Schemas also influence memory retrieval, particularly when it comes to expertise. Experts in a
specific field tend to have more elaborated schemas, which are mental frameworks for organizing
and representing knowledge (Chase & Simon, 1973; Frensch & Sternberg, 1989). These schemas help
experts fill in gaps in their memory, organize new information, and retrieve memories more easily.
Therefore, experts are more confident in their recollections, as their schemas provide a stronger
cognitive context for encoding, storing, and retrieving information.

Another important factor in memory retrieval is the self-reference effect, where people recall
information more effectively when they generate their own cues for retrieval. This was demonstrated
in a study by Mantyla (1986), where participants who created their own cues for word lists showed
almost perfect recall. The cues were most effective when they were compatible and distinctive to the
target word, highlighting the importance of personally meaningful cues for memory retrieval.
In practical terms, understanding these factors can improve memory retrieval, especially in
educational or therapeutic settings. When studying for a test, for example, organizing the
information in a way that matches how it will be tested can enhance recall. Additionally, matching
the level of processing during encoding and retrieval can improve memory performance, as shown
by Moscovitch & Craik (1976). By taking these contextual factors into account, individuals can
optimize their memory retrieval and recall.