The Mandela Effect: Why Thousands of People Confidently Remember Something That Never Existed

Close your eyes and picture Pikachu. You probably see a lightning-bolt tail with a black tip. That tip has never existed.

If you saw it clearly, you are in good company: it is one of the best-documented cases of the Mandela effect, the phenomenon in which many people share the same false memory, with the same details and astonishing confidence.

The interesting part is not that memory gets things wrong; we knew that already. It is that it gets them wrong in the same way in different heads.

The memory that gave the phenomenon its name

The term was coined in 2010 by Fiona Broome, who described herself as a paranormal researcher. At a convention in 2009 she discovered that other people shared her memory: Nelson Mandela had died in prison.

She described it in detail: funeral footage on the news, riots, a speech by his widow.

But Mandela walked out of prison in 1990, served as president of South Africa from 1994 to 1999 and died in 2013.

Broome and many followers took these coincidences as evidence of parallel universes. Memory science proposes something less cinematic, and considerably more unsettling: the certainty of remembering does not guarantee having lived it.

The experiment that turned an internet curiosity into science

For years the Mandela effect was mostly a forum pastime. In 2022, psychologists Deepasri Prasad and Wilma Bainbridge of the University of Chicago put it to a serious test in the journal Psychological Science.

They showed 100 people forty popular icons, each in three versions: the real one and two altered ones. Participants had to pick the correct one and say how confident they were.

Seven icons met every criterion for a shared false memory:

• C-3PO, whom many remember as entirely gold, although he has one silver leg.
• Pikachu, with a black tail tip he does not have.
• The Monopoly man, with a monocle he never wore.
• Curious George, the storybook monkey, with a tail that does not exist.
• The Fruit of the Loom logo, with a cornucopia that is not there.
• The Volkswagen logo and Waldo (Where’s Waldo?, known as Where’s Wally? in the UK).

The key finding was not the error but its consistency: people went wrong by choosing the same false version, and did so with high confidence. As the authors put it, just as we correctly remember the same images as other people, we also hold false memories of the same images.

An independent replication published in February 2024 by Clearer Thinking’s Transparent Replications project, with 389 participants, confirmed all seven cases.

It is not that we fail to look

The most intuitive explanation would be that we never pay attention to those details. The study tested it, and it did not hold up.

In a second experiment, participants first studied the real images while a system recorded which parts of each image they uncovered with the mouse. Minutes later they had to choose between the real and the altered version.

• On control icons they were right 82% of the time.
• On Mandela effect icons, only 53%, close to a coin toss.
• There were no differences in how they had looked at them.

More revealing still: many of the people who got it wrong said they remembered the false detail, such as the Fruit of the Loom cornucopia, which was not in the image they had just seen.

Another hypothesis, that we have simply seen the wrong version many times out there, does not explain most cases either. Searching online, versions of Pikachu with a black tail tip or the Monopoly man with a monocle barely appear, except on pages discussing the Mandela effect itself.

Memory reconstructs, it does not replay

Understanding why it happens means looking back at half a century of research on false memories.

In 1974, Elizabeth Loftus and John Palmer showed videos of traffic accidents and asked how fast the cars were going. Only one verb changed:

• When asked about the cars that “smashed” into each other, the average estimate was 40.5 miles per hour.
• When the verb was “contacted”, it dropped to 31.8.

A week later, 32% of those who had heard “smashed” remembered broken glass at the scene. There was none.

In 1995, Henry Roediger and Kathleen McDermott read lists of words related to sleep (bed, rest, awake…) without ever including the word “sleep”. Many participants recalled it and recognized it with confidence: in one experiment, 55% produced it as if they had heard it.

The conclusion of this work is that memory does not work like a recording. Every time we remember, we reconstruct, and expectations, schemas and words that were not in the original all feed into that reconstruction.

Michael Gazzaniga, who worked with Roger Sperry, described something similar in split-brain patients: what he called the left hemisphere’s “interpreter” prefers a coherent story to a gap, even if it has to invent part of it (Gazzaniga, 2000).

Confidence is not proof

Perhaps the most uncomfortable lesson of the Mandela effect is that feeling sure of a memory says little about its accuracy.

In 2003, Jennifer Talarico and David Rubin asked Duke University students, the day after the September 11 attacks, to describe how they had heard the news. Months later they asked again:

• Memories of 9/11 changed as much as memories of an ordinary day.
• But confidence in those memories stayed high, while confidence in everyday memories declined.

Their conclusion fits in a single phrase: vivid memories are “special only in their perceived accuracy”.

It is the same paradox that shows up with Pikachu: emotional, repeated or highly familiar memories feel more solid, and that very sense of solidity is what makes them so convincing when they fail.

Are some people more prone?

It would be tempting to think there is a type of person with a “bad memory” or who is especially suggestible. Research points elsewhere.

In 2019, Rebecca Nichols and Elizabeth Loftus compared three different false memory tasks and found that people who erred on one did not tend to err on the others. Links with personality were small and inconsistent.

Their conclusion was that there is probably no false memory “trait”: nobody seems especially prone or especially immune.

Put another way, the Mandela effect is not about some people; it is about how everyone’s memory works.

From “Luke, I am your father” to Snow White’s mirror

Outside the lab, popular culture is full of lines that almost everyone misremembers:

• In The Empire Strikes Back, Darth Vader does not say “Luke, I am your father”. The original line is “No, I am your father”.
• In the 1937 Disney film, the queen does not say “Mirror, mirror on the wall”: she says “Magic Mirror on the wall”. The “Mirror, mirror” formula comes from the Grimm tale tradition and later adaptations.

Both cases show a memory tendency that matches the icons: we fill in, simplify and make more “typical” what we remember. A line sounds better with the hero’s name in it; an enchanted mirror sounds better with a catchy repetition.

What the Mandela effect says about us

There is something humbling about this phenomenon. It is a reminder that memory is not an archive but a living process that rewrites itself every time it is used, just as the forgetting curve shows that forgetting is part of learning, or that stress alters what we manage to recall.

And there is something social: if thousands of people agree on the same mistake, perhaps memory is not only individual but also leans on shared expectations about how things “should” be.

Déjà vu makes something new feel already lived; jamais vu makes the familiar strange; the Mandela effect makes real something that was never there. Three oddities that, together, point to a single idea: what we experience as memory is a very convincing construction, and precisely for that reason it deserves a little curiosity, and a little less faith.

Sources

• Gazzaniga, M. S. (2000). Cerebral specialization and interhemispheric communication: Does the corpus callosum enable the human condition? Brain, 123(7), 1293-1326. doi:10.1093/brain/123.7.1293
• Loftus, E. F. and 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.
• Nichols, R. M. and Loftus, E. F. (2019). Who is susceptible in three false memory tasks? Memory, 27(7), 962-984.
• Prasad, D. and Bainbridge, W. A. (2022). The Visual Mandela Effect as evidence for shared and specific false memories across people. Psychological Science, 33(12), 1971-1988. doi:10.1177/09567976221108944
• Roediger, H. L. and McDermott, K. B. (1995). Creating false memories: Remembering words not presented in lists. Journal of Experimental Psychology: Learning, Memory, and Cognition, 21(4), 803-814.
• Transparent Replications (Clearer Thinking) (2024). Report #8: Replication of a study from “The Visual Mandela Effect as Evidence for Shared and Specific False Memories Across People”. replications.clearerthinking.org
• Talarico, J. M. and Rubin, D. C. (2003). Confidence, not consistency, characterizes flashbulb memories. Psychological Science, 14(5), 455-461.