In a recent Public Policy Polling survey of registered Republicans, over one third of the participants believed that thousands of Muslims had been cheering in the streets while the Twin Towers crumbled on September 11th. Donald Trump, as of publication, not only holds this belief himself but asserts that he is “100% right” about it despite being confronted with overwhelming evidence to the contrary. Despite being one of the most well-documented catastrophes in modern times, there is no video footage, photographs, or corroborated reporting of any celebrations happening.

While it’s tempting to believe that Trump and one third of Republican voters are simply delusional, it’s hard not to see the force of their convictions as evidence of a genuine feeling of “remembering” something. But can a group of people really remember something that didn’t happen? Hopes&Fears reached out to psychologists specializing in neurobiology, cognitive science, developmental psychology, and social psychology to find out the answer. It turns out Republicans might not be the only group likely to collectively (mis)remember the past.


Can groups of people "remember" something that didn't happen? . Image 1.

Malcolm Nicholson


Cover Illustration by Leonard Peng




Can groups of people "remember" something that didn't happen? . Image 2.

Stephen J. Ceci

Professor of Developmental Psychology at Cornell University

Off the top of my head the first and best example I can think of that demonstrates mass false memory is the classic social psychology study done in 1954. Princeton and Dartmouth played particularly rough football game in 1951 and then four years later students at both schools were shown a film of the game. Among the many examples of roughness, the All-American quarterback for Princeton got his nose broken and had a concussion, and another player got his leg broken. Everyone agreed that the teams were playing “dirty.” However, the way students from the two schools recalled the infractions was biased. Princeton students were much more likely to recall Dartmouth players committing infractions while Dartmouth students were more likely to recall the infractions being committed by Princeton players.

In some of the high-profile mass-allegation daycare cases, it was customary for one child to initially make a disclosure about his or her molestation by, say, a teacher’s aid, then soon most of the classmates would “remember” similar molestation. The contagion was caused by a combination of social pressures (“Don’t you want to help put the man in jail who hurt your classmate, Isabel?”; “Isabel said you were there and it happened to you, too.”) and various types of suggestion such as inducing the children to engage in imagery about the alleged molestation (“OK, so how do you think you would feel IF it happened to you?”) that subsequently gets incorporated in the memory trace of all classmates subjected to the imagery.



Can groups of people "remember" something that didn't happen? . Image 3.

Kimberly Wade

Associate professor in Psychology at the University of Warwick

There are a few relevant studies that show people can hold distorted memories about significant public events. The best example is an experiment by Stephan Lewandowsky and colleagues in 2005 that shows Americans are more likely than Australians and Germans to falsely remember that weapons of mass destruction had been discovered in Iraq. There is another study by Sacchi and colleagues (2007) looking at how doctored photos of past public events can distort people’s memories. Adults were shown misleading doctored photos of the 1989 Tiananmen Square protest in Beijing and a 2003 protest in Rome against the war in Iraq. The Beijing event was made to look more significant by increasing the size of the crowd. The Rome photo was made to look more violent by inserting police officers and aggressive-looking demonstrators. The results showed that people who viewed the doctored Beijing photo estimated that a larger number of people were involved than people who viewed the original photo. People who viewed the doctored Rome photo said the event was more and negative, and recalled more physical confrontation and damage to property. They were even less inclined to participate in future protests after seeing the doctored photo.

So it’s not inconceivable that this sort of thing can happen in a group setting. There are many studies showing that when two people witness the same event and discuss it, one person’s memory report can contaminate what the other person subsequently claims to remember. This is called ‘memory conformity’ in the literature and it’s a very powerful and well-documented effect.

There are several good reasons to think that childhood memories are more prone to distortion than recent memories. Of course, most people understand that memories fade over time and that childhood memories are more likely to be sparse and lacking in detail. Given this, and some other good reasons, we predicted that it would be easier to implant false memories in people for an event that supposedly happened when they were 2 years old rather than 10 years old. And that’s exactly what we found in an experiment we published in 2008 (Strange, Wade, & Hayne, 2008).

The thing is, false memories can serve a variety of functions. Lots of research shows that some false memories are self-serving— they can make us feel better about ourselves, better about our relationship, better about the country we live in and love.

People are more likely to remember information that is provided if it is in a weird, difficult-to-read font.




Can groups of people "remember" something that didn't happen? . Image 4.

Larry R. Squire

Professor of Psychiatry, Neurosciences, and Psychology at the University of California, San Diego

I am not sure what to make of the idea of false memories in groups and would think the issues are  the same as in individuals. For example, we did a study of how recollection surrounding the O.J. Simpson trial verdict changed over time in a group of students. Was this group behavior or something that happened in many individuals? In any case, I should think the mechanisms should be the same in individuals or groups (if there is a phenomenon that depends in some way on group behavior).

I should expect older memories to be more subject to distortion because there has been more opportunity for them to change as the result of retelling, hearing other versions, and having other (related) experiences.

MIT researchers recently were able to successfully plant false memories in mice




William Hirst

Professor of Psychology at The New School

We know that it’s relatively easy to implant memories. In certain situations you can implant them about 30 to 40 percent of time. We also know that once a memory has been implanted the process continues and there are studies on what’s called updating. Take the case of false news about the Iraq War early on which was discovered to be false; in one study if you were in Germany or Australia you were likely to update the information, but if you were American you weren’t as likely to update the information. So for some people, even if you were told a fact was wrong, you would still remember the previous incorrect fact. This suggests that memory is schema consistent, so if something fits into the way you think things should be it, you don’t easily revise the memory once it’s been formed.

A schema is sort your organized representation of the world. When something fits into your view of the world, you’re primed to absorb that information. If it’s mildly inconsistent, it sort of stands and becomes memorable. But if it’s too inconsistent then it slips away. In terms of groups talking together, studies suggest that false memories are more likely to arise in a group discussion than individually because there is more chance of somebody offering a false memory which can then be implanted. However, if somebody in the group says “no that’s not true”, that will mitigate the influence. But for groups with a strong shared interest, like Trump supporters, they’re less likely to dispute one another and therefore it’s less likely for their memories to be updated. 

I think it’s absolutely clear that our memories fit into a way of viewing and interpreting the world. From all the evidence, memory is not like a tape recorder. There’s not really a “truth” to memory. What Frederic Bartlett said is that memory is a continuous reconstruction. And what guides your reconstruction? Your view of the world. So memory,if you like, has a presentism (to use an ugly word). Your current view of the world, your current attitudes, allow you to reconstruct your own past to be consistent with your present self. We’re constantly reshaping our memory to essentially reinforce our present attitudes.

Also, every time we remember things, we’re selective. There are studies which suggests that not only does silence allow certain facts to decay or be forgotten, but also, because we talk about some things and not other things, the act of talking can actually promote forgetting some things. This is what I’ve called retrieval induced forgetting. Basically, you’re sculpting a collective memory about things which are rehearsed but also around things which we’re silent about. Furthermore, you’re more likely to show induced forgetting if a person you’re talking to is a member of your ingroup than an outgroup.

The basic thrust of my work is that our memories are to a large extent determined by our interactions with others, and by others I mean other people, media, and other external factors. Our memory is not only our own, but also that of all we interact with.

Fake Memories

In 1977, 60 eyewitnesses to a plane crash that killed nine people were interviewed by Flying Magazine. But they had differing recollections. One of the witnesses explained that the plane “was heading right toward the ground—straight down.” Yet photographs showed that the airplane hit flat and at a low-enough angle to skid for almost one thousand feet.




Can groups of people "remember" something that didn't happen? . Image 5.

Craig Stark, Ph.D.

Professor of Neurobiology and Behavior, Fellow at Center for the Neurobiology of Learning and Memory, Facisco J. Ayala School of Biological Sciences

There are two ways we need to think about this.  One is thinking about the group as a collection of individuals. And so, therefore, whatever leads one person to have a false memory-those same things can lead to one person or multiple people having a similar kind of false memory. So a classic example of this, myself included, will give Whether they’re giving demonstrations to audiences whether they’re a class of undergraduates or federal judges in which I plant false memories in 80% of the audience.

But in this case it’s really that you have a technique that’s so good at implanting a false memory—it’s got an 80% chance at working in each individual, so it’s going to work for most of the individuals in the audience. But there’s no interaction amongst that that’s doing anything to alter that it’s really just that each of them have this independent probability. So that’s one kind of group false memory.

The other kind is whether the group itself, or the interactions amongst the group, will do anything to alter the rate of the false memory, so as people start talking to each other, what does this actually do. I’m not a social psychologist who does anything in terms of group dynamics, but there is a bit of stuff out there on this kind of thing. A classic paper is one that surveyed about 3000 people after 9/11 looking at the accuracy of their memories, so comparing their memory versus the original one, and one of the things that they found is that the contents of the memory were really dictated in the end—so first of all one big aspect of the results is when they compared those two accounts, they found that 40% of the details differed from time point 1 to time point 2. What really drove the contents of that was how much they had paid attention to the media or discussed with other people the details of the event in the intervening time.

So as you’re talking with your buddies it can easily become part of your memory. Just as your buddy might be saying “Yeah, I can’t believe it, there were all these Muslims in Jersey who were celebrating,” and then their seeing it becomes your seeing it, so there definitely are these group dynamics as well.

For the September 11th story they looked at things like “did you have personal loss?” How emotionally salient was this for you? Did you live in New York City?” They looked at all those factors and those factors, but they didn’t do anything to alter the odds that those factors would be distorted or changed after a year. But it’s not like “Oh because you had a personal loss, then your memory of those events is protected.” Pitting them against each other is a difficult kind of thing—it’s like saying “Who’s a better athlete? A lineman on the football team, a basketball player?”

I’ve been looking at the neurobiology of memory for a long time; I haven’t seen anything in there that says “You should be able to stick some bit of information into your brain, and have it be free of any sort of distortion or degradation. Nothing that says it’s burned into your memory and you’ll remember it like it just happened. I’ve never seen anything that says you should be able to do that. And there are dozens of mechanisms in which you should be able to change, and it’s good thing that it can change because our memory didn’t evolve so that we would be able to have something happen to us—whether a day, week, or decade later—be able to relive all of that information perfectly from that one individual. We extract information from across events and from our past experiences; they shape we see the world and remember the world, and that shaping is distortion. We call is wisdom, we call it knowledge, we call it experience, and that’s great. But it is also going to be filling them in, and a lot of that filling in is education guesses, but it also leads to filling in of all of these different kinds of distortion. IT’s a much more efficient system, more adaptive system, but it is one that’s not going to be perfect.

Take your digital camera—and it takes these pictures, and we think these pictures are an absolute record of what happened because it’s a picture. Except the digital camera is using a JPEG compression and what the algorithm does it makes these little squares saying “what can I throw away that the eye is never going to see?” I’m gonna throw away the details to get to the gist of what actually matters. Your memory is going a similar thing—it’s called a lossy compression scheme—and what it does is if you want to zoom into, say a license plate, you won’t get that crystal clear enhancement because that information is gone. And our memory is doing the same thing. We have tons of information stored in our synapses and the strength of our synapses is how it’s storing memory but it’s using those same synapses for different memories. So as you expose yourself to similar things it’s using similar patterns.

The mere act of retrieving a memory puts it into a plastic state; you can do mad scientist ways of erasing a memory, or in a more normal situation, it’s in a labile state, so if there’s something new you’ve learned about the memory, you can update the contents of your memory. Just as you might hear your buddies talking about Muslims protesting in New Jersey, you put this new bit of information in your memory.

Frederich Barlett said memory is an imaginative reconstruction; we’re making it up out of our biases, our expectations, and how the world has shaped us.

The strength of true memories 

In 2005, neuroscientists Yoko Okada and Craig Stark used misleading questions in a study to tamper with eyewitness memories. But they used FMRIs to record brain activity while participants viewed the materials. They found that well-formed memories are not easily swayed. True memories were formed when there was high activity in the prefrontal cortex and the hippocampus, while false memories were created when there was low activity in the prefrontal cortex.