by David M. Reutter

“Remembrance of things past is not necessarily the remembrance of things as they were.”

– Marcel Proust

When jurors weigh evidence in a criminal trial, few types of evidence are more persuasive than a witness pointing across a courtroom and declaring with absolute certainty: “That’s the person who did it.” Jurors lean forward. Prosecutors rest their case. Defense attorneys brace for the worst. The confidence, the emotion, the vivid details – all signal that truth has spoken.

Yet decades of memory research reveal a disturbing reality that our criminal justice system has been reluctant to accept: human memory is not a faithful recorder of events but a fragile, malleable process prone to distortion and outright fabrication. Witnesses can recall events that never occurred with the same conviction and detail as genuine memories. They can identify innocent strangers as perpetrators with unwavering certainty. And they can do so without any intent to deceive – their minds simply filling gaps, incorporating suggestions, and reconstructing the past to feel like truth.

While courts scrutinize physical evidence with chain-of-custody protocols and expert analysis, testimonial evidence faces only a basic credibility test, despite research showing that confidence, detail, and emotion are unreliable indicators of accuracy. The consequences ripple through the justice system: innocent people imprisoned, guilty offenders free to harm again, public trust eroded, and taxpayers bearing the cost of wrongful convictions. Understanding how false memories form, why they feel real, and what safeguards can protect against them is not merely an academic exercise – it is an urgent necessity for a justice system that claims to seek truth and justice.

Steve Titus, who was 31 in 1980, after spending an evening at a Seattle restaurant with his girlfriend was subject to a traffic stop. Actually, Titus was pulled over because his vehicle was similar to the description of one driven by a rapist who raped a woman earlier that evening. Titus also was a fit for the general description of the rapist.

Police later showed the victim a photo lineup that included Titus. She pointed to Titus and said, “That one’s the closest.” Titus was charged with rape based upon that inconclusive identification. At the ensuing trial, the victim pointed to Titus and testified, “I’m absolutely positive that’s the man.” Without any corroborating or physical evidence tying him to the crime, the jury found Titus guilty of rape. That is the power of eyewitness testimony.

Situations where a victim goes from making an inconclusive identification to absolute certainty are not rare. In fact, it is not unusual for a victim to identify during a second or third lineup a person who also appeared, but was not identified, in the initial lineup(s). According to the Innocence Project, about 69% of DNA exonerations involve mistaken identity. Information from the National Registry of Exonerations shows that across all exonerations, about 28% involve misidentification. The difference between these statistics shows the full magnitude of wrongful convictions caused by misidentification because absent DNA or other evidence to prove the wrong person was identified, a conviction stands. Even in the face of DNA or other exonerating evidence, prosecutors regularly fight to prevent reversal of the conviction.

There are vast differences between what experts and lay people believe about memory. Results of a 2011 telephone survey of 1,838 lay people published in the Public Library of Science by psychology professors Daniel J. Simmons and Christoher F. Chabris found that 37.1% of respondents mostly agreed or strongly agreed that “the testimony of one confident eyewitness should be enough evidence to convict a defendant”; 63% mostly agreed or strongly agreed that “[h]uman memory works like a video camera, accurately recording the events we see and hear so that we can review and inspect later”; and almost 48% mostly agreed or strongly agreed that “[o]nce you have experienced an event and formed a memory of it, that memory does not change.” Importantly, none of the 73 memory experts surveyed agreed on any level with those statements.

This chasm between public misinformation and scientific fact has profound implications for the courtroom. Jurors arrive believing memory functions like a recording device, making them naturally inclined to trust eyewitnesses who speak with certainty and rich detail – the very characteristics that research shows can accompany false memories just as readily as genuine ones. When a witness takes the stand and confidently identifies a defendant, jurors interpret that confidence as proof of accuracy, unaware that the human brain routinely generates false memories with the same neural signatures as true ones.

Meanwhile, judges admit eyewitness testimony under standards that assume memory’s basic reliability, rarely requiring expert testimony to educate juries about memory’s actual limitations. The result is a justice system built on a foundation of outdated folk psychology rather than established science, where the most persuasive eyewitness may simply be the one whose memory has been most thoroughly – and unwittingly – reconstructed. Until courts and juries understand that confidence does not equal accuracy, that detail does not guarantee truth, and that emotion reflects conviction rather than correctness, testimonial evidence will continue to send innocent people to prison while the guilty remain free.

When Imagination Becomes Evidence: The Psychology of False Memory

Experts define a false memory as a fabricated or distorted recollection of an event, which can range from an entirely new memory of an event that never occurred (a memory of commission) to a significantly altered memory of an event that did happen (a memory of reconstruction or distortion). False memories are distinct from simple, trivial errors in recollection, such as misremembering a date or name. Instead, they represent a significant blurring of the real and the imagined, a phenomenon sometimes referred to as “confabulation” when describing the spontaneous, pathological production of distorted memories.

Several common characteristics of false memories include a high degree of subjective confidence in the memory’s accuracy, a lack of awareness that the memory is false or distorted (meaning the person genuinely believes the memory is true and is making no attempt to lie or deceive), and the memory’s content is usually drawn from elements of the person’s existing knowledge and experiences. False memories can be provoked by external suggestion, as happened when the rape victim was shown a photo lineup that unintentionally reinforced a false identification, leading to the wrongful conviction of Steve Titus. They may also develop spontaneously (e.g., due to source monitoring failures) or be associated with certain psychological or neurological conditions, such as dementia.

The various ways in which false memories are created and adopted have been thoroughly investigated in cognitive psychology. The following discussion explores the mechanisms and external factors – ranging from individual cognitive biases to social pressure – that drive these memory errors, starting with the influence of the imagination.

Imagination inflation is a well-established memory phenomenon in cognitive psychology. Researchers have concluded that the repeated imagining of a fictitious or plausible event increases a person’s subjective confidence that the event actually happened, thus creating a false memory.

This phenomenon occurs because imagination activates similar neural pathways to those used in processing real experiences. The imagined event gains details and feels more familiar or “fluent” over time, a process which can lead to a source monitoring error, where the individual mistakenly attributes the vivid, internally generated memory to an external, real-life source.

Closely related to imagination inflation is the more general concept of source monitoring failure. Source misattribution is a common memory phenomenon that occurs when people accurately recall information or a memory fragment but attribute it to the wrong source or context. It is a fundamental type of memory error within the Source Monitoring Framework, which describes the cognitive processes used to determine the origin of a memory.

A classic example of source misattribution is cryptomnesia, or “unconscious plagiarism,” which occurs when an individual remembers an idea, piece of music, or story but completely forgets its original source, leading them to genuinely believe they came up with it independently. More generally, source misattribution can lead to false memories when, for example, vivid accounts of events described by friends or family are mistakenly incorporated into one’s own autobiographical memory. This is often because the imagined or secondhand memories lack the specific contextual, perceptual, or emotional details that typically distinguish a personally experienced event.

Beyond the cognitive processes that affect everyone, certain personal characteristics can increase an individual’s vulnerability to these memory distortions. Researchers have found that certain individual difference factors may predispose individuals to the formation of false memories. Specifically, persons exhibiting a cluster of traits related to fantasy and absorption, such as high levels of fantasy proneness, imaginative involvement, creativity, and openness to experience, may be more susceptible to memory distortion effects like imagination inflation. This increased susceptibility is often attributed to these individuals being highly capable of creating vivid and elaborate mental imagery, which can subsequently lead to errors in source monitoring – the inability to distinguish between an internally generated image and an externally experienced event.

Similarly, persons with high levels of dissociation – a mental state characterized by a sense of detachment from one’s immediate surroundings, thoughts, feelings, or memory – are also more likely to report false memories. Dissociation, particularly the tendency toward dissociative absorption, appears to impair the cognitive processes necessary for accurate source monitoring, making it more difficult to accurately evaluate the boundary between real experiences and suggested or imagined events. These individual differences are relevant to the criminal justice system as they suggest that some eyewitnesses or suspects may be inherently more vulnerable to suggestive interviewing tactics.

The intersection of memory integrity with the criminal justice system is further complicated by factors related to confidence and social interaction. The confidence-accuracy relationship is a complex phenomenon often misunderstood, which is highly relevant to false memories. While an individual’s confidence can sometimes be a good indicator of memory accuracy, studies consistently show that this relationship is weak or non-existent when the memory itself has been subjected to contamination or suggestion. False memories often feel real and highly believable because they are accompanied by strong emotions, vivid sensory details, and high retrieval fluency – factors that drive a deceptively high level of confidence. Crucially, false memories can be expressed with the same degree of certainty as accurate ones, leading jurors and investigators to place inappropriate weight on an eyewitness’ confidence, especially if that confidence has been artificially inflated post-event by feedback or suggestive questioning by police and prosecutors.

Other factors that can shape memories include social influence and peer pressure. Research shows that in group settings, people are more likely to unconsciously conform to others’ accounts, leading to the creation of shared false memories, a phenomenon known as memory conformity (or the “social contagion of memory”). For example, after the Oklahoma City federal building bombing in 1995, three eyewitnesses identified the suspected perpetrator as Timothy McVeigh. One witness claimed to have seen McVeigh with an accomplice, and after speaking with the other two, they also agreed that they recalled an accomplice. However, the FBI later determined that McVeigh most likely acted alone, and no accomplice ever existed. This case starkly illustrates how peer discussion can cause individuals to incorporate false details into their own recollections, resulting in seemingly corroborating, yet inaccurate, testimony.

In addition to psychological and social factors, an individual’s physiological and physical state can significantly contribute to memory vulnerability. Sleep deprivation does not provide the optimum neurological conditions for memory consolidation; on the contrary, adequate sleep is essential for the consolidation of long-term memories. Sleep deprivation, therefore, impairs this consolidation process. Furthermore, lack of sleep is known to acutely impair the ability to monitor the source of memories during retrieval, leading to an increase in source monitoring errors where an individual confuses distinct memory traces, resulting in false memories.

Similarly, aging can lead to a decline in memory function, particularly in the frontal lobe activity necessary for source monitoring, making older adults more susceptible to forming distorted memories and being influenced by suggestion.

Finally, the impact of high-arousal events on memory is a significant concern for the accuracy of eyewitness testimony. Acute stress and trauma are also known to impair memory. High levels of stress hormones, such as cortisol, can interfere with both the encoding of event details (leading to fragmented or incomplete memories) and the accurate recall of stored information. In the context of false memories, a common psychological theory is that the memory system attempts to reconstruct a coherent narrative, either by filling in gaps or altering details to create a probable event, which is a form of confabulation that may serve the unconscious purpose of helping an individual cope with the traumatic or otherwise stressful experience. These distortions can lead to witnesses or victims reporting details that are subjectively real but factually incorrect.

The mechanisms underlying false memory formation reveal a disturbing truth: the very cognitive processes that allow us to construct coherent narratives from our experiences can betray us, transforming imagination into conviction and suggestion into certainty. From imagination inflation to source monitoring failures, from the contaminating effects of social conformity to the memory-distorting impacts of stress and sleep deprivation, the pathways to false recollection are numerous and often invisible to the person experiencing them. Most troubling is that these fabricated memories arrive with all the subjective markers of authenticity – vivid sensory details, strong emotions, and unwavering confidence – making them indistinguishable from genuine recollections to both the person remembering and those judging their credibility. When an eyewitness declares with absolute certainty that they saw the defendant at the crime scene, or when co-witnesses corroborate each other’s accounts after discussion, the criminal justice system is presented with testimony that feels true but may be fundamentally false.

The implications for justice are profound and deeply unsettling. Jurors instinctively trust confident witnesses, judges weigh the consistency of multiple accounts, and investigators pursue leads based on seemingly credible testimony. Yet, all of these markers of reliability can be present in memories that are entirely or partially fabricated. The Oklahoma City bombing case demonstrates how quickly false details can spread through witness populations, while the Steve Titus case shows how a single suggestive identification procedure can set an innocent person on the path to wrongful conviction. When we add individual vulnerabilities – the fantasy-prone witness, the sleep-deprived victim, the elderly eyewitness with declining source monitoring abilities, the traumatized survivor attempting to reconstruct fragmented memories – the potential for grave miscarriages of justice increases greatly. Without rigorous safeguards that account for the malleability of human memory, the criminal justice system risks building cases not on what actually happened but on collectively constructed fictions that feel real to everyone involved.

How Memory Fails: The Cognitive Science Behind Wrongful Conviction

The professional literature on memory suggests that false memories arise from the operation of normal, constructive memory processes. Traditionally, three foundational principles generally apply to the formation and report of false memories.

The first principle is the Reconstructive Nature of Memory, a concept expounded upon in 1932 by Sir Frederic C. Bartlett. As noted by Melanie Caroline Steffens and Silvia Mecklenbräuker in their paper False Memories: Phenomena, Theories, and Implications, “Events are not stored in memory as are pictures in a videotape.” Instead, events are actively reconstructed at the time of retrieval in light of present knowledge and schemas, which are existing knowledge structures. This is core to Schema Theory, which holds that these schemas guide not only the retrieval of events but also their initial storage, meaning a memory is colored by pre-existing expectations from the moment it is formed.

The second principle focuses on how information is initially processed and encoded. False memories may arise because the inferences necessary for understanding events are often made immediately during encoding, and after a delay, they become indistinguishable from the information that was directly perceived. In other words, “ambiguous situations are interpreted in terms of existing schemas, and they are then stored accordingly. Even in the absence of further distortions, these events would be recalled falsely,” according to Steffens and Mecklenbräuker. This cognitive blending of experienced information and inference is also central to Fuzzy Trace Theory, which posits that people often rely on vague, meaning-based “gist” traces rather than precise “verbatim” traces when recalling an event, making them prone to integrating consistent but factually inaccurate details.

The third general aspect relates to the retrieval process and the difference between potential and actual recall. As memory researcher Endel Tulving explained, a key distinction exists between availability (something stored in memory) and accessibility (the ability to retrieve it under given circumstances). This is captured by the Encoding Specificity Principle, which states that the effectiveness of a retrieval cue in making a memory accessible depends on the extent to which that cue overlaps with information that was encoded with the original event. When a memory cue is misleading or suggestive (as is often the case in criminal investigations), it can powerfully trigger the retrieval of inaccurate or schema-consistent details, thus making an inaccessible false memory accessible and reportable.

        Building upon these traditional principles, modern researchers have put forth more specific accounts of false memory formation. The Fluency-Misattribution Perspective holds that the subjective experience of familiarity results from the unconscious misattribution of fluent processing (ease of retrieval). For example, memory is enhanced when the cognitive process is smooth and easy. After watching blurry photos of famous people slowly come into focus, test subjects were able to recall precisely when they could identify the photo. But the reverse was not true; after watching photos slowly blur, they were later unable to recall as precisely when they could no longer identify them. Because their familiarity with the process of decoding a photo as it comes into focus was much greater than with the reverse process, their memory for the timing of recognition was affected. The general principle is that if a memory is easy and quick to retrieve (fluent), one is more likely to misattribute that feeling of fluency to the memory being true, strong, or real.

The Source Monitoring Framework (“SMF”) Theory provides a more detailed decision-making account of memory errors. As Steffens and Mecklenbräuker wrote, “We need to distinguish between the content of a memory trace (‘what’ we know) and its source (‘where’ we know it from).” Recalling the source involves a complex decision-making process where we evaluate qualitative characteristics of the memory (e.g., sensory detail, emotional reaction, cognitive operations) to determine its origin – did I see it, hear it, dream it, or imagine it?

According to the SMF, memories that are rich in sensory and contextual detail (e.g., clear sights, sounds, smells, and spatial location) are usually attributed to external sources – a personally experienced event. Conversely, memories that are rich in cognitive details (e.g., memories of having thought about it, reasoned through it, or mentally pictured it) are more likely to be attributed to internal sources – imagination or a thought process.

A failure of source monitoring occurs when these qualities are misapplied. For example, a memory that is really a mental picture from an imagination may be so vivid that it is mistakenly attributed to an external childhood memory. This is often the case with stories a family told about a child so often that the individual can no longer be sure whether they actually recall the event or just recall the story. Conversely, when people claim to remember details about events that never happened, they may in fact be remembering features of other, actually perceived events and simply misattributing the source of those features. The SMF explains how a memory of commission (an entirely false event) can arise from confusing an internally generated (imagined) source with an external (real) source.

The Source Monitoring Framework and Fluency-Misattribution Perspective are closely related, as both deal with how people evaluate their memory experiences. However, the SMF is a comprehensive model that emphasizes the evaluation of specific characteristics of the memory trace (like sensory or cognitive details), while the Fluency-Misattribution Perspective focuses more broadly on the global experience of retrieval ease.

Two other key theories further refine the mechanisms of false memory. Both the Activation-Monitoring Account and the Fuzzy-Trace Theory suggest processes that complement the Source Monitoring Framework, but they propose more specific mechanisms for false memory formation, particularly for spontaneous false memories, i.e., those not caused by external suggestion.

The Fuzzy-Trace Theory posits that when we encode a memory, two separate and independent memory traces are established: (1) a verbatim trace and (2) a gist trace. The verbatim trace consists of the exact, detailed information, e.g., a person’s specific facial features, a precise statement. The gist trace focuses on the semantic or general meaning of the event, e.g., gender, hair color, the overall theme of betrayal.

In a criminal investigation, a misidentification occurs because the person relies only on the long-lasting gist information. A face in a lineup may match the relevant gist and “feel familiar, but the details are insufficiently recalled to make a true match.” This can result in confident identification of a suspect who is “gist-consistent” with the true offender, despite the person not actually being the offender. This experience of relying on gist to inaccurately “remember” specific details is termed phantom recollection or “phantom recall.”

The prevalence of these spontaneous errors highlights that memory distortion is not limited to external suggestion. Spontaneous false memory can also arise as a result of source-monitoring errors or a phenomenon known as change blindness, where a witness fails to notice a substantial change in a visual scene, e.g., that the perpetrator and a bystander are actually different people. Research conducted by C. J. Brainerd, V. F. Reyna, and Stephen J. Ceci found that adults are more prone to spontaneous false memories than children. Children, on the other hand, are more prone to suggestion-induced false memories than adults. As researchers Bi Zhu, Chuansheng Chen, Elizabeth F. Loftus, and Qi Dong noted, these differences highlight that susceptibility to memory errors varies significantly across the lifespan based on cognitive factors.

An instance of spontaneous memory errors involves self-serving memory biases. Adults systematically revise their past perspectives to enhance their current self-evaluation. As Eryn J. Newman and D. Stephen Lindsay explained, people are prone to “mistakenly recall some attributes as having improved over time when they have not, thereby misremembering their past selves in a negatively biased way … [and] … producing an illusion of improvement.” This shows that even memories of the self, which are highly rehearsed, are subject to retrospective memory distortion that spontaneously alters factual details to maintain a preferred narrative.

The scientific understanding of memory reveals a fundamental challenge to the criminal justice system: memory is not retrieval but reconstruction, not playback but interpretation. From Bartlett’s pioneering work in 1932 to contemporary theories like Fuzzy-Trace Theory and the Source Monitoring Framework, decades of research converge on a single conclusion – our memories are shaped by schemas, colored by inference, and guided by gist rather than precise detail. When a witness identifies a suspect based on gist-consistent features rather than verbatim recall, when fluent processing creates false familiarity, when internally generated images are misattributed to external events, the resulting testimony can be both subjectively genuine and objectively wrong. These are not failures of honesty but inevitable products of normal cognitive processes that are not well-understood in criminal proceedings.

When the legal system treats confident eyewitness identification as gold-standard evidence and jurors cannot distinguish phantom recollection from genuine recall, wrongful convictions become predictable outcomes. The theoretical accounts of false memory do not merely explain how errors occur; they prove why our approach to memory evidence is fundamentally flawed, prioritizing the subjective experience of remembering over what memory science tells us about the limits of human recollection.

Recovered Memories and the Malleability of Children

A profound chasm in belief exists between researchers and certain therapists regarding the validity of recovered memories. In the early 1990s, a pattern emerged where individuals entered therapy for common issues like depression or relationship problems, but, over the course of treatment, developed vivid memories of severe, often ritualistic, childhood abuse. Therapists claimed these were genuine memories that had been repressed – a defense mechanism theorized to bury traumatic events deep within the subconscious for years or decades, only to be recovered in therapy.

A classic example involves a patient seeking help for panic attacks, disturbed nightmares, and difficulties trusting men. During therapy, the patient “recovers” faint memories of early childhood abuse by the father and subsequently severs all family contact. In such instances, the therapeutic literature sometimes directed practitioners to suspend doubt, with tenets like, “You must believe that your client was sexually abused, even if she sometimes doubts it … No one invents abuse. Neither children nor women make up having been abused,” as documented by Steffens and Mecklenbräuker.

The controversy over repressed memories raises two critical scientific questions. The first is “whether there is evidence that very significant and emotionally upsetting events can be forgotten,” and the second is “whether there is a difference between repression and ‘normal’ forgetting.” After reviewing the relevant research, Steffens and Mecklenbräuker concluded that, first, traumatic childhood events can be forgotten (a phenomenon sometimes clinically termed dissociative amnesia) and second, there is also some evidence supporting a general repression mechanism for memory.

However, the consensus among cognitive scientists strongly rejects the idea that memories of abuse are presumed unlikely to be forgotten in “ordinary” ways, thereby protecting the subject “from the ravages of time and are immune from the normal laws of memory,” as described by Mark L. Howe and Lauren M. Knott. The majority of experts assert that there is no scientific evidence for a special, invulnerable memory mechanism that preserves traumatic memories perfectly. They caution against the use of therapeutic techniques designed to “unlock” these memories, as such methods pose a high risk of suggestion-induced false memory.

This scientific caution is backed by the long-term consensus of major professional organizations. While the debate over the mechanisms of trauma forgetting continued, the scientific community ultimately rejected the notion that repressed memories recovered through suggestive therapeutic techniques should be treated as reliable evidence. Major organizations, including the American Psychological Association and the American Medical Association, released official statements expressing deep skepticism regarding the scientific validity of “recovering” memories of childhood abuse after years of amnesia, particularly when those memories emerge under the influence of hypnosis, guided imagery, or other highly suggestive therapeutic methods. The consensus holds that while forgetting trauma is possible (dissociative amnesia), and true memories can sometimes return spontaneously, memories produced through therapeutic suggestion are inherently unreliable and pose an unacceptable risk of wrongful accusation.

These memory recovery techniques, sometimes used by caring and well-meaning therapists, include highly suggestive procedures such as asking patients to imagine they have been abused and describe what it may have been like (activating imagination inflation), interpreting dreams, using group therapy sessions that provide social pressure and detailed abuse narratives (memory conformity), or the use of hypnosis and drugs such as sodium amytal. Such highly suggestive techniques, along with others like guided imagery and cueing memories with family pictures, have sometimes led people to “recall” extraordinary events like satanic ritual abuse and alien abduction, clearly illustrating the social construction of false memories and the severe unreliability of these memory retrieval methods, as documented by Cara Laney and Elizabeth F. Loftus, in their article, “Past-life identities, UFO abductions, and satanic ritual abuse: the social construction of memories.”

The case of Meredith Maran – an educated and sophisticated adult journalist and activist who ultimately concluded she suffered from a socially reconstructed false memory of child sexual abuse – demonstrates that susceptibility to suggestion is not limited to those with cognitive deficits. Maran, deeply involved in the incest survivor movement, felt intense social and ideological pressure to align her personal history with the cause, which led her to question her past and ultimately conclude she had repressed memories of abuse.

If an educated adult can fall prey to the recollection or implanting of false memories under social and therapeutic pressure, the risk for children is exponentially higher.

Researchers have found that the sophisticated autobiographical memory system – the system necessary for accurate source monitoring and enduring event recall – is still maturing in young children. Experts refer to the period before this system is established as infantile and childhood amnesia, where memories are poorly formed and typically forgotten. As Mark L. Howe and Lauren M. Knott explained, this system generally emerges after the age of five to seven years, and “more stable, adult-like autobiographical memories are rarely seen before the age of 9 - 10 years, and even then, recalling events from this period will only be possible if they are particularly distinctive and memorable.” Due to these developmental limitations, young children are highly susceptible to errors of commission – inserting inaccurate details –even about personally experienced events involving their own bodies.

The dangers inherent in interviewing child witnesses were powerfully demonstrated in the infamous Kelly Michaels case in 1988, which centered on allegations of sexual abuse against a teacher at the Wee Care Nursery in Maplewood, New Jersey.

The case originated when a 4-year-old pupil, after a rectal temperature check by a pediatrician, said, “That’s what my teacher does to me at school.” This single, possibly ambiguous statement led to a highly suggestive investigation. During subsequent interviews, the child inserted his finger into the rectum of an anatomical doll and claimed the teacher did the same thing to other children. Parents of Wee Care children were then told by social workers that abuse had occurred and were urged to question their children, leading to a cascade of allegations. Among the claims were fantastic details that Michaels licked peanut butter off genitals, played piano in class while nude, and used silverware to rape children. Despite a complete absence of physical evidence and no corroboration from other staff members, a jury convicted Michaels of 115 counts. The convictions were ultimately reversed by the New Jersey Supreme Court based upon the unreliability of the testimony and the techniques used to procure the children’s statements.

The amici curiae brief filed in State of New Jersey v. Michaels, 642 A.2d 1372 (N.J. 1994), citing extensive research by scholars like S.J. Ceci and M. Bruck, detailed the investigative practices that lead to false memories and false confessions in children:

Repeated Questioning and Pressure: Investigators who pursue a single hypothesis can inadvertently generate false reports. The amici advised the Court that repeated questions often result in the child changing her original answer, as the child assumes the first answer was wrong or unsatisfactory to the adult. Furthermore, an accusatory tone set by an examiner can lead to the child fabricating reports of past events, even when they have no memory of the event occurring.

Incorporation of Details: Like adults, children subjected to repeated misinformation across interviews can incorporate these details into their memories. A 1997 study involving five- and six-year-olds demonstrated that children’s stories readily conformed to the suggestions or beliefs of the interviewer, even when the suggestions contradicted the staged event they had witnessed. In fact, the amici wrote that up to 75% of children’s remarks “were consistent with the examiner’s view by the end of the first interview,” and 90% answered interpretative questions in agreement with the interviewer’s point of view, as opposed to what actually happened.

Authority, Stereotypes, and Anatomical Dolls: The higher status of an interviewer (e.g., a police officer or social worker) is known to result in children complying with the implicit and explicit agenda of the adult. Stereotype inducement, where a child is repeatedly told that a person “does bad things,” can lead the child to incorporate this belief into their reports. Finally, the use of anatomically detailed dolls during interviews may inadvertently promote sexualized behavior or false reports.

The research is conclusive: suggestive questioning can easily lead children to create and recall false memories. As Steffens and Mecklenbräuker concluded, “There is abundant evidence of false memories, not only in children, but in adults,” and “Experts cannot decide with certainty whether a given specific memory is ‘recovered’ or ‘false.’”

In response to these findings and the lessons from high-profile cases like Michaels and similar day-care hysteria scandals of the 1980s and 1990s, significant reforms have been implemented in child forensic interviewing practices. Developed in the mid-1990s by researchers at the National Institute of Child Health and Human Development (“NICHD”) in collaboration with developmental psychologists, law enforcement professionals, and other experts, the NICHD Investigative Interview Protocol provides a structured, evidence-based framework to minimize suggestibility and enhance the accuracy of children’s testimony.

This protocol emphasizes open-ended prompts to elicit free-recall narratives, e.g., “Tell me what happened,” establishes ground rules such as allowing children to say “I don’t know” or correct the interviewer, and includes phased rapport-building with non-substantive practice interviews to demonstrate expected detail levels without introducing bias. It explicitly avoids leading questions, repetitive probing, and externally derived information that could contaminate recall, shifting from therapeutic or unstructured methods criticized in earlier cases to a forensic perspective focused on reliability.

Research has confirmed its effectiveness. Studies show that interviewers trained in the protocol increase the use of open-ended questions from as low as 5-6% in unstructured interviews to 45-50% or more, yielding significantly more accurate and detailed accounts while reducing false details and suggestibility. For example, field evaluations in locations like Quebec and Israel demonstrated improved information quality and credibility assessments post-training, with sustained benefits when accompanied by ongoing supervision and feedback. Adopted internationally and continually refined based on ongoing research, the NICHD Protocol underscores the criminal justice system’s capacity to mitigate memory distortions through science-informed safeguards, ultimately helping to avert wrongful convictions and ensure more equitable outcomes for both accused individuals and genuine victims.

The evolution from the recovered memory controversies of the 1990s to today’s evidence-based interviewing protocols represents a critical recognition: memory is not a faithful recorder but a reconstructive process vulnerable to suggestion, social pressure, and the passage of time. This vulnerability exists across all ages, but reaches its apex in young children, whose still-developing autobiographical memory systems make them especially susceptible to incorporating false details into their recollections. The Kelly Michaels case stands as a stark reminder of how well-intentioned investigators, armed with flawed techniques and operating under the assumption that children never lie about abuse, can generate testimony so contaminated that it leads to wrongful convictions based entirely on fabricated memories.

The profound lesson is that belief alone – no matter how sincere – cannot substitute for scientific rigor in the pursuit of justice. When therapeutic or investigative practices prioritize validation over accuracy, when repeated suggestive questioning replaces neutral inquiry, and when the absence of corroborating evidence is dismissed rather than given due weight, the criminal justice system becomes an instrument of injustice rather than truth. The adoption of protocols like the NICHD framework demonstrates that we can do better, but only if we remain vigilant against the human tendency to see patterns where none exist and to mistake the vividness of a memory for proof of its veracity. The stakes could not be higher: on one side lie innocent individuals whose lives are destroyed by false accusations, and on the other, genuine victims whose credibility is undermined when the system fails to distinguish reliable testimony from suggestion-induced fiction.

The Misinformation Effect: How a Single Word Can Manufacture False Testimony

Decades of cognitive research on eyewitness testimony have conclusively established the misinformation effect – the phenomenon where exposure to misleading information after an event can impair a person’s memory of the event or cause them to report details they never witnessed.

Professor Elizabeth Loftus of the University of California at Irvine and her colleagues provided foundational evidence for this effect. In a seminal 1974 study, participants were shown a video of a car accident. When asked a follow-up question, the specific verb used to describe the collision significantly biased their recall. For example, when the question used the word “hit” to describe the accident, the average speed estimate was 34 mph. By contrast, when the same question used the highly suggestive word “smashed,” the average response was 41 mph. More disturbingly, in a second experiment, the verb choice biased the recollection of entirely new, non-existent details: 14% of those given the “hit” terminology described broken glass at the accident scene, while 32% of those given the “smashed” terminology reported seeing broken glass, even though no broken glass was presented in the video. This illustrates how a single word can alter both quantitative estimates and qualitative memory details.

The danger of misinformation is amplified by repetition. In a study titled “Repeated Exposure to Suggestion and the Creation Of False Memories,” Kent State University researchers Maria S. Zaragoza and Karen J. Mitchell demonstrated a direct link between the frequency of exposure and the rate of false memory creation. Participants watched a movie of a burglary and were presented with a post-event questionnaire. A control condition presented zero misleading information, while other participants were presented the misleading information once or three times. For example, one question asked about a young man “dressed in jeans, and gloves” entering the house, though the thief had not actually worn gloves. The results were stark: the false alarm rate for reporting the robber wore gloves was 1-2% for the control group, but jumped to 36-38% for the group exposed to the false detail once, and rose further to 52-61% for those exposed three times. This proves how easily wrong details are suggested and incorporated, with the risk dramatically increasing with each repetition.

Beyond altering existing memories, researchers have successfully used suggestive techniques to implant entire false events that produce false memories in up to 25% of participants in general studies. The frequency of false memory reports is often increased by the use of imagery and the presentation of plausible, detailed information, which lends credibility to a non-event. The photoshopped false memory paradigm (a variation of the classic “Lost in the Mall” technique) vividly illustrates this: when researchers manipulated childhood photos to show participants in a hot air balloon, 50% of participants produced false memories and described their non-existent hot air balloon experience.

These false memories of childhood events exploit the fact that “a typical feature of childhood memories [is] that they are part of the family history, revolving around pictures with accompanying stories,” as noted by Steffens and Mecklenbräuker. “Therefore, it must be very difficult to disentangle true memories of the events themselves from memories of pictures and the respective often-told stories, all the more so because childhood memories are often blurred as opposed to rich and detailed.” Memory errors are also common in identification, as research has long established that people have “a tendency to misidentify an innocent bystander to a crime as an offender … or to identify a familiar person from an entirely different context as the offender,” as noted by Helm – a clear example of a source monitoring error.

The body of memory research has established a strong scientific consensus regarding the fundamental fallibility of memory, a consensus crucial for the criminal justice system to acknowledge. A 2011 survey of 4,669 memory research subjects demonstrated this wide agreement:

72% agreed that memory reflects not only what a witness saw but information obtained later on (the misinformation effect).

63% agreed that eyewitnesses sometimes identify as a culprit someone who they have seen in another situation or context (source monitoring errors/familiarity misattribution).

73% agreed that confidence could be influenced by factors other than memory accuracy.

As summarized by cognitive science experts, the five key messages regarding the operation of memory are:

Memory does not work like a video camera or recording device; it is a constructive process.

When we construct a memory, errors and distortions are inevitable.

We not only distort memories for events that we have witnessed, but we may also form completely false memories for events that never occurred.

There is no convincing evidence to support the existence of the psychoanalytic concept of an invulnerable, repressed memory.

There is currently no reliable way to distinguish, in the absence of independent evidence, whether a particular memory is true or false.

The difference between what memory science has established and how the legal system operates represents one of the most consequential failures in modern criminal justice. Despite overwhelming scientific consensus that memory is constructive and inherently fallible, that misleading information inevitably contaminates recall, and that no reliable method exists to distinguish true memories from false ones without independent corroboration, courtrooms continue to treat uncorroborated eyewitness testimony as compelling evidence of guilt.

The misinformation effect is not a theoretical vulnerability but a documented certainty, yet standard investigative practices involving repeated questioning, witness collaboration, and suggestive procedures create precisely the conditions under which false memories are guaranteed to form. When a system knowingly employs methods proven to generate false recollections at rates exceeding 50%, then treats the resulting testimony as reliable proof sufficient to deprive individuals of liberty, it has abandoned any pretense of aligning justice with truth. The question is no longer whether memory evidence leads to wrongful convictions, but how many innocent lives we are willing to sacrifice before we demand that legal standards catch up to scientific reality.

Case Studies in Memory-Based Wrongful Conviction

False testimonial evidence in the criminal justice system creates serious negative impacts for taxpayers, community leaders, the accused, and the criminal justice system. The conviction of innocent persons has allowed serious homicide offenders (“SHO”) to run free and commit additional homicides. Data from the National Registry of Exonerations shows that 62 SHOs were later responsible for another 249 murders, and 114 of those murders were committed after an innocent person was in prison for the SHO’s crime. Serial rapists and other criminals run free when an innocent is misidentified as the perpetrator, creating new victims.

Beyond these systemic harms, the human toll of wrongful conviction is devastating. A false conviction shatters the life of an accused innocent person. The violence and rigors of prison can change a person, not to mention the harms caused by the loss of one’s liberty. Steve Titus only spent a night in jail after he was wrongfully convicted, his sentence delayed pending new evidence his attorney sought that might exonerate him. He was cleared of the crime only after Seattle Times journalist Paul Henderson tracked down a serial rapist who eventually admitted to the rape that was the basis of Titus’ conviction. Despite that happy outcome, the incident created turmoil in Titus’ life. He developed an anger problem from being falsely accused, a situation that resulted in his fiancée leaving him. Titus lost his job. He sued officials connected with his conviction, but just before the trial in that matter, Titus died at the age of 35 from a stress related heart attack. His parents later accepted a $2 million settlement of his case.

The contamination of memory through suggestive procedures can be seen clearly in the case of Anthony Broadwater, who was convicted of the 1982 rape of Alice Sebold, then a freshman at Syracuse University. The identification process, detailed in court records, immediately highlighted the unreliability of the memory evidence.

During the initial investigation, Sebold was shown a police composite sketch that she did not recognize. She was then taken to the Syracuse police station to view a physical lineup, where she viewed several men, including Broadwater. Sebold was unable to make a definitive identification during the lineup itself. Instead, she pointed to two men: Broadwater and the man standing next to him, stating that the two men looked identical and that one of them was her attacker. Crucially, when she was unable to choose between the two, an officer then asked her which man had the “sly eyes,” and it was only after this leading suggestion that she settled on Broadwater.

Despite her initial confusion and the investigator’s suggestive cue, Sebold was certain of her choice at trial. The prosecution then compounded the error by offering a psychologically unsound explanation to the jury: they claimed the man next to Broadwater was his friend and was placed there to intentionally trick her. This explanation provided a rationale for Sebold’s initial confusion, effectively allowing the jury to dismiss her lack of certainty during the lineup and place their full faith in her final, strong courtroom identification.

Broadwater, who was 21 at the time of the crime, was denied parole multiple times before finally being released after 16 years. As a registered sex offender, Broadwater struggled to obtain educational and employment opportunities. Nearly 40 years after the crime, a movie producer working on an adaptation of Sebold’s book, Lucky, which recounts her rape experience, noticed discrepancies in the stories. This led to a re-examination of the evidence and the testing of the original microscopic hair evidence, which excluded Broadwater as the perpetrator. Broadwater was fully exonerated in November 2021. The case stands as a stark example of a source monitoring error – where Sebold may have recalled the face but misattributed its source – that was then fatally cemented by suggestive questioning and a narrative that assured the jury her conviction was correct.

Another case demonstrating how repeated suggestive procedures can transform tentative uncertainty into absolute conviction involved Ronald Cotton, who was identified as the burglar of two apartments and the rapist of two women on a single night in July 1984. Flawed eyewitness testimony led to his conviction, mainly upon the testimonial evidence, despite key missteps in the identification process.

The investigation centered on the testimony of victim Jennifer Thompson, who was repeatedly exposed to suggestive identification procedures that amplified her initial uncertainty. First, Thompson was shown a photo array from which she picked Cotton’s picture, stating, “This looks the most like him.” This tentative identification was immediately reinforced when a detective told her she had done a good job. Second, Thompson identified Cotton in a subsequent physical lineup, where Cotton was the only person present who had been in the initial photo array. This is a highly suggestive procedure known as “mugshot exposure” or “unconscious transference,” which increases the likelihood that a witness will identify the person they saw in the photo rather than the person they saw at the crime scene. At trial, fueled by these procedures, Thompson testified with absolute certainty that Cotton was her rapist.

The only physical evidence was a flashlight found in Cotton’s apartment that was similar to one the perpetrator used and a piece of rubber found at one crime scene that was similar to the rubber on Cotton’s shoes. Cotton was sentenced to life, plus 50 years. DNA evidence eventually exonerated him of both crimes in June 1995, revealing the actual perpetrator was a man named Bobby Poole. Cotton received a $110,000 payment for his wrongful conviction from the state of North Carolina. The case serves as a powerful illustration that a witness’s high confidence does not correlate with high accuracy when the memory has been contaminated by suggestive external factors.

These cases are not isolated incidents but rather symptoms of a broader systemic problem. Despite the fact that eyewitness testimony is unreliable, the criminal justice system allows for the conviction of a defendant upon the testimony of a single witness. At James Earl Ruffin’s trial, Ann Mae Meng confidently pointed Ruffin out as the man who raped her. On May 3, 1982, a jury found Ruffin guilty and the trial court imposed five life sentences upon Ruffin. His initial appeals failed. Then, in 2003 DNA evidence exonerated Ruffin, who detailed his experiences in the book, Why Me? When it Could’ve Been You!

The scope and scale of this problem is staggering. Faulty eyewitness testimony is the single most common cause of wrongful convictions in the United States. This issue directly connects to the inherent fallibility and reconstructive nature of human memory.

According to the Innocence Project, misidentification was a contributing factor in 69% of the DNA exonerations in the United States. Examining these misidentification cases reveals critical systemic problems:

In 38% of exonerations, the innocent person was convicted based on multiple eyewitness identifications of the same suspect, demonstrating that multiple testimonies do not guarantee accuracy.

In 53% of misidentification cases, the error involved cross-racial misidentification, a well-documented cognitive phenomenon where individuals are less accurate at identifying faces of a different race.

In half of these cases, eyewitness testimony was the central or sole evidence and lacked significant corroborating physical evidence.

Understanding the true nature of these errors is crucial to addressing them. The problem, in most instances, is not that the witness intended to lie or deceive. Rather, the witness is the victim of a faulty reconstructive process that is inherent in our malleable, fragile memory system. The critical question for the criminal justice system is whether or not suggestive investigative and interview techniques amplified this inherent unreliability, leading to the creation of false memories or mistaken identifications that a witness genuinely believes to be true.

These systemic injustices caused by faulty testimonial evidence erode public trust in the integrity of the criminal justice system and cause significant fiscal harm to taxpayers, who finance the substantial settlements, attorney fees, and costs that ensue after an exoneration. More profoundly, faulty testimony results in a sham trial that shatters the life of an innocent person and, in 36% of those cases where the true perpetrator was later identified by DNA, allows the guilty party to remain free to victimize society further. Most disturbingly, in just less than half of the misidentification cases, the real perpetrator went on to commit additional crimes of rape and murder while an innocent person was imprisoned.

Evidence-Based Safeguards Against False Memory

Because false memories can feel incredibly real and are shaped by the brain’s natural tendency to integrate new, potentially misleading information, criminal justice stakeholders must take proactive steps to preserve the integrity of the memory process and educate those who test and weigh the testimonial evidence presented to them.

The first line of defense against creating false memory lies in the interview process. As a best practice, interviewers must avoid closed or leading questions. As Professor Annelies Vredevelt noted, rather than probing with specific questions, it is much better to let the person tell the story of their own accord, without undue interruption. Experts refer to this highly effective technique as Free Recall (a core component of the Cognitive Interview).

“Research shows that stories told in response to free-call prompts are much more accurate than stories told in response to a series of closed questions,” said Vredevelt. “So if you really want to get to the bottom of something, restrain yourself and don’t ask too many questions.” At most, interviewers should use an open and general prompt such as, “Can you tell me more about that?” to encourage elaboration without introducing new details or suggestions.

Because memory is highly susceptible to contamination and malleability, experts urge procedural changes, particularly concerning suspect identification. Professor John Wixted of the University of San Diego emphasized the critical importance of limiting contact between the witness and the suspect: “we must avoid repeated identification procedures with the same witness and suspect.” This recommendation applies not only to additional tests conducted by police investigators but also to the final test conducted in the courtroom.

Wixted and his colleagues (including Gary Wells, Elizabeth Loftus, and Brandon Garrett) explained how the memory system processes lineups, revealing why repetition is destructive:

Context Reinstatement and Activation: According to the Encoding Specificity Principle, memory is cue-dependent. When a witness is asked to identify a culprit, the retrieval cue must reinstate the crime context (e.g., “Do you see the person who stole the car?”), which activates the relevant memory content, including the culprit’s face.

The Problem of Elaborative Processing: Upon seeing the lineup, the witness compares each lineup face against the activated memory. This process, governed by Signal Detection Theory, generates memory-matching signals. However, this comparison also creates a new, detailed memory of every face in the lineup via elaborative processing.

Source Monitoring Failure in Subsequent Tests: Because the face of an innocent suspect is better remembered after the first test, this newly formed memory trace is highly accessible in a second test. This powerful signal can easily confuse the witness’s memory of the activation source (a failure of source monitoring), leading them to attribute the strong feeling of familiarity to the memory of the crime rather than to the previous lineup procedure. Several memory tests may, therefore, dramatically increase the likelihood of misidentifications.

Given memory’s fragility, courts must not overlook the importance of the initial identification test. “The first test is the most reliable test,” Wixted said. “The first test probes the witness’s memory but also unavoidably contaminates the witness’s memory. All tests beyond that very first one only serve to test contaminated memory and contaminate it further. And once memory is contaminated, there is no way to decontaminate it.”

Wixted and his co-authors do not argue that eyewitness testimony should be excluded entirely. Instead, they highlight the scientific finding that the Confidence-Accuracy Relationship is only reliable under specific conditions: the confidence expressed at the very first identification (the “first test”) is a strong predictor of accuracy, but only if the identification was made rapidly. This critical information – that a decision made rapidly and with high confidence in the first test tends to be more reliable than decisions made slowly – is often ignored. Greater weight is instead incorrectly placed on second identifications and, especially, in-court identifications, where confidence has been artificially inflated.

It is common procedure that a photo lineup with the suspect does not net an identification, but the same suspect is placed in a second lineup with other persons, resulting in familiarity contamination from the initial lineup. Current jury instructions, which generally instruct jurors on general credibility, are inadequate to overcome lay persons’ common memory misconceptions and educate them on the science of memory. These juror misconceptions about memory profoundly impact how much weight the testimonial evidence tips Lady Justice’s scale.

Because the chasm between what the average person and memory experts believe about false memories is so wide, the best way to educate a jury is often to consult and procure at trial the testimony of a memory expert to educate the jury on the science of false memories.

Stakeholders should be mindful of Loftus’ simple, yet powerful warning: “Just because someone tells you something and they say it with confidence – just because they say it with lots of detail, just because they express emotion when they say it – it doesn’t mean that it really happened. We can’t reliably distinguish between true memories from false memories; we need independent corroboration.”

Conclusion

The criminal justice system rests on a foundational paradox: it demands proof beyond reasonable doubt while routinely accepting as evidence the most unreliable proof imaginable – uncorroborated human memory. Decades of rigorous scientific research have dismantled the comforting fiction that memory operates like a recording device, revealing instead a reconstructive process so vulnerable to contamination that a single suggestive word can conjure nonexistent details, repeated exposure to misinformation can corrupt recall in over half of witnesses, and standard investigative procedures can transform tentative uncertainty into absolute – yet utterly false – conviction.

The science is unequivocal, the expert consensus overwhelming, yet courtrooms continue to operate as though these findings are mere academic curiosities rather than urgent warnings about systematic injustice. When 69% of DNA exonerations involve eyewitness misidentification, when innocent people serve decades in prison while the actual perpetrators commit additional rapes and murders, when the gap between what jurors believe about memory and what scientists know remains a chasm, we are not witnessing occasional failures but the predictable consequences of a system built on obsolete assumptions about human cognition.

The human cost of this willful ignorance is staggering and irreversible. Steve Titus died of a stress-induced heart attack at 35, his life destroyed despite exoneration. Anthony Broadwater lost 16 years and his future to a lineup procedure so contaminated that the witness initially could not choose between two men until an officer asked which had “sly eyes.” Ronald Cotton served over a decade for crimes he did not commit because a detective praised a tentative photo identification and then placed him as the only familiar face in a subsequent lineup. These are not aberrations. They are the inevitable outcomes when memory science is ignored in favor of intuitive but false beliefs about confidence, detail, and emotion as markers of truth. The Kelly Michaels case revealed how easily children’s testimony can be manufactured through suggestive questioning, yet single-witness convictions remain permissible. Maddeningly, expert testimony about memory’s limitations remains the exception rather than the rule.

What makes this crisis particularly insidious is that the system’s failures are largely invisible to those perpetrating them. Witnesses genuinely believe their false memories; investigators sincerely think they are uncovering truth; jurors confidently convict based on testimony that feels compelling; and judges admit evidence under standards that assume memory’s basic reliability. The problem is not malice but ignorance – an institutionalized refusal to align legal practice with scientific reality. When courts treat a confident in-court identification as more reliable than the contaminated first identification from which it derives; when prosecutors fight DNA evidence that would expose misidentifications; and when jury instructions fail to educate about imagination inflation, source monitoring errors, and the misinformation effect, the system has abandoned its stated commitment to truth in favor of procedural momentum and the path of least resistance.

The reforms outlined in this article – evidence-based interview protocols, elimination of repeated identification procedures, recognition that only the first rapid and confident identification has predictive validity, mandatory expert testimony on memory science – are not radical proposals but minimum requirements for a system that claims to pursue justice. The NICHD Protocol demonstrates that we possess the knowledge and tools to dramatically reduce false memories in child witnesses. Professor Loftus’ research proves that we can identify the precise conditions under which memory becomes unreliable. The question is whether the legal system has the courage and integrity to implement what science has already established, or whether it will continue sacrificing innocent lives on the altar of outdated folk psychology and institutional inertia. Every day that passes without comprehensive reform represents a choice – a choice to prioritize convenience over accuracy, tradition over truth, and the appearance of certainty over the messy reality of human cognition.

The stakes extend beyond individual wrongful convictions to the legitimacy of the entire criminal justice system. When the most persuasive evidence is also the least reliable, when confidence masks contamination, when multiple witnesses corroborate shared false memories, and when juries remain ignorant of memory’s fundamental limitations, we have not constructed a system of justice but a theater of false certainty. The real perpetrators remain free, creating new victims. Public trust erodes as exonerations accumulate. Taxpayers fund multi-million-dollar settlements. And innocent people lose decades of their lives, their families, their futures – sacrificed to a system that knows better but refuses to do better.

Until courts require independent corroboration of memory evidence, until judges admit expert testimony as standard practice rather than rare exception, until prosecutors stop fighting DNA evidence and accept that contaminated identifications are worthless regardless of confidence, and until jurors understand that vivid, emotional, detailed testimony can be completely false, the criminal justice system will continue producing wrongful convictions at a rate limited only by how many cases lack the DNA evidence necessary to expose them. The choice before us is clear: align legal standards with scientific reality, or accept that we are operating not a justice system but a mechanism for transforming reasonable doubt into wrongful certainty.  

 

Sources: Brainerd, C. J., Reyna, V. F., & Ceci, S. J. (2008). Developmental reversals in false memory: A review of data and theory. Psychological Bulletin, 134(5), 343–376; “The fallibility of memory in judicial processes: Lessons from the past and their modern consequences” by Mark L. Howe and Lauren M. Knott, published in Memory (Volume 23, Issue 6, pp. 742–759, 2015); Kassin, S. M., Tubb, V. A., Hosch, H. M., & Memon, A. (2011). On the “general acceptance” of eyewitness testimony research: A new survey of the experts. American Psychologist, 56(5), 405–416; Lamb, M. E., Orbach, Y., Hershkowitz, I., Esplin, P. W., & Horowitz, D. (2007). A structured forensic interview protocol improves the quality and informativeness of investigative interviews with children: A review of research using the NICHD Investigative Interview Protocol. Child Abuse & Neglect, 31(11–12), 1205–1231; Loftus, E. F., & Palmer, J. C. (1974). Reconstruction of automobile destruction: An example of the interaction between language and memory. Journal of Verbal Learning and Verbal Behavior, 13(5), 585–589; Newman, E. J., & Lindsay, D. S. (2009). Spontaneous false memories for self-performed actions. Journal of Memory and Language, 61(3), 391–402. Steffens, M. C., & Mecklenbräuker, S. (2007). False memories: Phenomena, theories, and implications. Zeitschrift für Psychologie / Journal of Psychology, 215(1), 12–24. Vredeveldt, A. (2018, October 25). How to interview witnesses to get the full story – and avoid the wrong conviction. The Conversation; Wixted, J. T., Wells, G. L., Loftus, E. F., & Garrett, B. L. (2021). Test a witness’s memory of a suspect only once: The robbing of the “golden thread” of the criminal trial. Psychological Science in the Public Interest, 22(3), 1–50; Zaragoza, M. S., & Mitchell, K. J. (1996). Repeated exposure to suggestion and the creation of false memories. Psychological Science, 7(5), 294–300; Zhu, B., Chen, C., Loftus, E. F., & Dong, Q. (2013). The relationship between DRM and misinformation false memories. Memory & Cognition, 41(6), 863–872.

 

Additional sources: National Registry of Exonerations; innocenceproject.com; American Psychological Association; American Medical Association; nature.com; psychologicalscience.com;  sciencedirect.com; sagepub.com; tandfonline.com; researchgate.com; nlm.nih.org; nobaproject.com; simplypsychology.com; scientificamerica.com; TED Talks; dailyorange.com; abcnews.go.com; cbsnews.com; seattletimes.com; standrewslawreview.com; process.org; verywellmind.com.

Subscribe today

Already a subscriber? Login