If you've spent too much time on the Internet (or the Internet from a parallel universe before you slipped into this one), you may have heard of the Mandela Effect. Named after people misremembering Nelson Mandela's death, the term is given to any collective false memories.

Examples include films that have never existed, such as Shazaam starring Sinbad, or the fact that Fruit of the Loom has never used a cornucopia in its logo.

A team of psychologists from the University of Chicago – intrigued by the idea – decided to put it to some actual scientific rigor. As well as testing whether there really was an effect, they aimed to discover why visual Mandela effects (VME) occurred. The results were sort of maddening.

In their first experiment – currently available as a pre-print ahead of publication in the journal Psychological Science – participants were asked to look at images of a logo, character, or mascot, including popular examples of the Mandela Effect and others that had been added as controls. As well as the original "real" version of the image, they included several other versions with mistakes on them, made to fit in with the original design as much as possible. This included popular misremembered versions.

The participants were asked to select the image they believed to be the original, as well as rate how confident they were that it was correct, and estimate the number of times they had seen it. The image was only considered to be a potential VME if it was consistently misremembered, people were confident about their choice, and the same wrong image was consistently selected. Interestingly, in the questions which used examples of the Mandela Effect popularly cited on the Internet, the popularly misremembered image was selected "a significantly higher proportion" of the time than the original.

"These results indicate that these seven images (C3PO, Curious George, the Fruit of the Loom logo, the Monopoly Man, Pikachu, the Volkswagen logo, and Waldo from Where’s Waldo?) had not only accuracies below chance, but also specific incorrect versions that were falsely recognized as the original," the team wrote in their paper. "Thus, these seven images were labeled as ‘VME-apparent."

"Furthermore, their accuracy is surprisingly low given the reported familiarity and confidence people had with these images," they added.

So far, so weird. Next, the team showed the participants the correct images and asked them to study them, without explaining they would be asked to recall information about the image they saw. When they were subsequently asked to select between the correct version and a manipulated version, the VME images were still consistently chosen over the correct version they had just studied.

"This low accuracy for the VME image set is remarkable, given that participants had just seen the correct image minutes prior during the study phase, yet still chose the false version to indicate their memory," the team wrote.

When asked about their choice, those who had selected the correct image said things along the lines of they “only saw the fruit, not the cornucopia”, while people who selected the VME image also claimed that they remembered seeing the manipulation a few moments ago (in this example, the cornucopia) even though they had not.

"In fact, incorrect responses to VME-apparent images were more often attributed to memory of the manipulated feature (66.54 percent) than those to matched non-VME images (44.92 percent), which instead tended to be more guess-based."

From this part of the experiment, they concluded that a popular source image (e.g. if there was a popular image of Pikachu with a black tip added to the tail, which it does not usually have but people took as canonical) was not the explanation for the visual Mandela effect "as it is unlikely that the noncanonical version from prior experience is overriding their recent experience of the canonical image."

One idea that might explain why people are making the same mistakes is "schema theory" which suggests that people fill in missing information (say, they can't remember exactly what the Monopoly Man looks like) with information based on our expectations and associations (slap a monocle on him in our minds, as he is known to be incredibly rich). 

However, this theory falls down with many of the VMEs. In one experiment (main image) the participants were asked to select the correct Fruit of the Loom logo from the VME version, the correct version, and one with a manipulation.

“They could have picked the correct Fruit of the Loom logo, the Fruit of the Loom logo with the cornucopia, or the Fruit of the Loom logo with a plate underneath it," co-author Deepasri Prasad said in a press release. "The fact that they chose cornucopia over plate, when plates are more frequently associated with fruit, is evidence against the idea that it’s just the schema theory explaining it.”

Disappointingly, or maybe just intriguingly, the team found no real explanation for the consistent mistakes. In another experiment, participants were asked to draw the logos and characters, and still reproduced Curious George with a tail, Pikachu with a black tail, and the Fruit of the Loom logo with a cornucopia.

"Evidence suggests that some people may be making consistent memory errors, even with extensive visual experience with the icon and without having experienced variants before," they write in their discussion.

"In sum, we revealed a set of images that cause consistent and shared false memories across people, spurring new questions on the nature of false memories. We show that the VME cannot be universally explained by a single account. Instead, perhaps different images cause a VME for different reasons."

Or – hear us out – the participants have shifted universe, and just don't know it yet.