In the field of psychology, a false memory is a phenomenon in which someone recalls something that did not actually happen or recalls it differently from the way it actually happened. Suggestibility, activation of associated information, the incorporation of misinformation, and source misattribution have been suggested as mechanisms that underlie a variety of types of false memory.
Early work The first known investigation of the false memory phenomenon was by psychology pioneers Pierre Janet and Sigmund Freud.
Freud was fascinated with the ways memory could be used and manipulated. Some modern day researchers say his studies have been quite influential in contemporary memory research, including research in the field of false memory.Knafo, D. (2009). Freud's memory erased. Psychoanalytic Psychology, 26(2), 171–190. Pierre Janet was a French neurologist credited with significant contributions to memory research, specifically through his ideas on dissociation and memory retrieval through hypnosis.Zongwill, O. L. (2019). Memory abnormality. Retrieved from https://www.britannica.com/science/memory-abnormality
In 1974, Elizabeth Loftus and John Palmer conducted a study to investigate the effects of language on the development of false memory. The study involved two separate but related experiments.
In the first experiment, 45 participants were asked to watch one of three different and randomly assigned videos of a car accident. The videos showed collisions at , and . Afterwards, participants filled out a survey, and one subset of them was asked, "About how fast were the cars going when they smashed into each other?" However, the verb used to describe the collision was different for other groups of participants. Rather than smashed, other verbs used included bumped, collided, hit, or contacted. If the actual speed in the video viewed by a participant was indeed the main factor in their estimate of it, one would expect the participant to give lower estimates for lower speed collisions. Instead, the word being used to describe the collision in the survey question seemed to predict participants' speed estimates better than the speed itself.
The second experiment also involved showing participants videos of car accidents, but the phrasing of one of the follow-up questionnaires was different. One hundred fifty participants were randomly assigned to three conditions. Those in the first condition were asked the same question as the first study using the verb smashed. The second group heard the word hit in the question. The final group was not asked about the speed of the crashed cars. The researchers also asked participants in all the groups if they had seen any broken glass, knowing that there actually was none in any of the videos. The responses to this question for the first two groups showed that whether broken glass was recalled or not heavily depended on the verb used in the earlier question; a larger number of participants in the "smashed" group asserted that there was broken glass than in the other groups.
The first main result here is that the words used to phrase a question can heavily influence the response given. Second, the phrasing of a question can influence the recall of previously unnoticed details. These results appear to support false memory as a true phenomenon.
Replications in different contexts (such as study participants' viewing videos of hockey games instead of car crashes) have shown that different scenarios require different framing effects to produce differing memories.
Manifestations and types
Presuppositions and the misinformation effect A presupposition is an implication resulting from specifically chosen language. If a person is asked, "What shade of blue was the wallet?", the questioner is implying that the wallet was blue. The question's phrasing presents the respondent with an implied fact. This presupposition can generate one of two phenomena: a true effect or a false effect. In true effect, the implication was accurate: in the above example, the wallet had in fact been blue. This makes the respondent's recall stronger, more readily available, and easier to extrapolate from. A respondent is more likely to remember a wallet as being blue if the prompt had indicated that it was blue than if the prompt had not said so. In false effect, the implication was false: from the same example, the wallet had in fact not been blue. This nevertheless influences the respondent's belief that it was true (i.e., that the wallet was blue), which affects their memory. The false effect can also alter a subject's responses to later questions, to which the subject answers so as to keep the responses consistent with the false implication. Regardless of the effect's being true or false, the respondent is attempting to conform to the supplied information.
Loftus's meta-analysis on language manipulation studies suggested that the misinformation effect can influence the recall process and products of human memory. Even the smallest adjustment in a question, such as the choice of a definite vs. an indefinite article, can alter responses. For example, asking someone if they had seen "the" stop sign rather than "a" stop sign invokes a presupposition that a stop sign was in fact in the scene. In the experiment, this presupposition increased the number of people who responded that they had indeed seen the stop sign.
As noted above, the strength of verbs used in conversation or questioning can influence recall; the choice of words among "met", "bumped", "collided", "crashed", and "smashed" all influenced the intensity level of the car accident participants recollected. The words "bumped", "hit", "grabbed", "smacked", or "groped" would each paint a different picture of a potential offender in the memory of an observer of possible sexual harassment if the observer were questioned about it later. The stronger the word, the more intense the incident-as-remembered is. This in turn could generate further false memories to fit the recollection better (e.g. it could change how a person is thought to have looked or how fast a vehicle was thought to be moving before a collision).
Word lists One can trigger false memories by presenting subjects with a continuous list of words. When subjects were presented with a second version of the list and asked if the words had appeared on the previous list, they found that the subjects did not recognize the list correctly. When the words on the two lists were semantically related to each other (e.g., sleep/bed), it was more likely that the subjects did not remember the first list correctly and created false memories (Anisfeld & Knapp, 1963).
In 1998, Kathleen McDermott and Henry Roediger III conducted a similar experiment. Their goal was to intentionally trigger false memories through word lists. They presented subjects with lists to study, all containing a large number of words that were semantically related to another word that was not found on the list. For example, if the word that they were trying to trigger was "river", the list would contain words such as flow, current, water, stream, bend, etc. They would then take the lists away and ask the subjects to recall the words on the lists. Almost every time, the false memory was triggered, and the subjects would end up recalling the target word as part of the list when it was never there. McDermott and Roediger even went as far as informing the subjects of the purpose and details of the experiment, and still the subjects would recall the non-listed target word as part of the word list they had studied.
Staged naturalistic events Subjects were invited into an office and were told to wait there. After this, they had to recall the inventory of the visited office. Subjects recognized objects consistent with the "office schema", although they did not appear in the office. (Brewer & Treyens, 1981)
In another study, subjects were presented with a situation where they witnessed a staged robbery. Half of the subjects witnessed the robbery live, while the other half watched a video of the robbery as it took place. After the event, they were seated and asked to recall what had happened during the robbery. The results showed that those who watched the video of the robbery recalled more information more accurately than those who were live on the scene. Still, false memory presented itself in ways such as subjects seeing things that would fit in a crime scene that were not there, or not recalling things that did not fit the crime scene. This happened with both parties, displaying the idea of staged naturalistic events.Ihlebæk, C., Løve, T., Erik Eilertsen, D., & Magnussen, S. (2003). Memory for a staged criminal event witnessed live and on video. Memory, 11(3), 319-327.
Relational processing Memory retrieval has been associated with the brain's relational processing. In associating two events (in reference to false memory, say tying a testimony to a prior event), there are verbatim and gist representations. Verbatim matches to the individual occurrences (e.g., I do not like dogs because when I was five a chihuahua bit me) and gist matches to general inferences (e.g., I do not like dogs because they are mean). Keeping in line with the fuzzy-trace theory, which suggests false memories are stored in gist representations (which retrieves both true and false recall), Storbeck & Clore (2005) wanted to see how change in mood affected the retrieval of false memories. After using the measure of a word association tool called the Deese–Roediger–McDermott paradigm (DRM), the subjects' moods were manipulated. Moods were either oriented towards being more positive, more negative, or were left untouched. Findings suggested that a more negative mood made critical details, stored in gist representation, more accessible.
Mandela Effect
Specific false memories can sometimes be shared by a large group of people. This phenomenon was dubbed the "Mandela Effect" by paranormal researcher Fiona Broome, who reported having vivid and detailed memories of news coverage of South African anti-apartheid leader and President Nelson Mandela's (after whom the effect was named) dying in prison in the 1980s. (He actually died in 2013, decades after his release and after serving as President of South Africa from 1994 to 1999.) Broome reported that hundreds of other people had written about having the same memory of Mandela's death, some while he was still alive. Her speculation that the phenomenon arises from "movement between parallel realities" is in opposition to more mundane psychological theories.
One well-documented example of shared false memories comes from a 2010 study that examined people familiar with the clock at Bologna Centrale railway station, which was damaged in a bombing in 1980, repaired shortly afterwards, and in 1996 stopped and permanently set to the time of the explosion, as a commemoration. In the study, 92% of respondents falsely remembered the clock as having been stopped since the bombing.
Other examples include:
The Berenstain Bears being spelled "Berenstein" Looney Tunes being spelled "Looney Toons" The logo of clothing brand Fruit of the Loom containing a cornucopia Darth Vader telling Luke Skywalker, "Luke, I am your father" in the climax of The Empire Strikes Back (he instead says "No, I am your father") Mr. Monopoly wearing a monocle The existence of a 1990s movie titled Shazaam starring comedian Sinbad as a genie (which Snopes suggests could be a confabulation of real memories, possibly including Sinbad wearing a genie-like costume during a TV marathon of Sinbad the Sailor movies in 1994,28 December 1994, on the cable channel TNT; the marathon featured movies including Sinbad and the Eye of the Tiger (1977). the 1996 film Kazaam featuring a genie played by basketball star Shaquille O'Neal, and the 1960s animated genie-themed series Shazzan)