How false memories happen

In December 2013, the former President of South Africa, Nelson Mandela, died at his home. That is a fact. It is also a fact that quite a lot of people strongly believe he died in the 1980s, while still unjustly imprisoned. This is one of a number of ‘Mandela effects’, common discrepancies between people’s memories and the real world: Rich Uncle Pennybags the Monopoly mascot doesn’t have a monocle, Sinbad was never in a movie called Shazaam, and the Korean peninsula hasn’t suddenly shifted 1000km north.

Some Mandela effect proponents believe that the world truly used to be different and we have somehow glitched into another reality, Matrix-style. I’ve got a much more mundane explanation for you: we’re not as good at remembering as we think we are, and we can quite easily create false memories of things that never actually happened.

Your memory: nowhere near as accurate as a floppy disk, though at least compatible with modern computer interfaces

What causes false memories?

There are three major procedures that researchers have used to cause people to create a (hopefully harmless) false memory [i]:

Imagination inflation: when we’re asked to imagine that an event has happened, it becomes more likely that we will later think it really did happen.
False feedback: when you’re told by an external authority that it’s likely something happened even when it didn’t, you may well start believing it. For example, in a study where people watched a video and were then interviewed about the events in it, a researcher implying that something happened when it didn’t could later give rise to a false memory that it had happened [ii].
Memory implantation: when you’re told that someone else who was ‘there’ recalls a false event, you’re likely to start believing it happened. In false memory studies, researchers will often lead participants to believe that another family remember recalls an event (classically, being lost in a shopping mall as a child [iii]) that never really happened.

It’s seems like it’s worryingly easy to create a false memory, and though most of us wouldn’t want to do that to another person, there are people who would. However, only a small minority of people seem to develop really compelling false memories, perhaps because the memory of an event and belief that the event truly happened seem to be dealt with separately by the mind.

Old Norse religion was on to something with Huginn and Muninn.

In 2012, Clark and colleagues [iv] asked people to imitate simple actions performed by a researcher, like cupping a hand around an ear. They videotaped these interactions with the researcher and later showed them back to the participants – except that the video had been doctored to show the researcher doing an action in front of the participant that they’d actually done alone. The participants strongly believed they had done these actions, until they were told that the videos were faked. At that point, they still ‘remembered’ having done the actions, but stopped believing that those events had happened.

Okay, so most of us are unlikely to believe false memories. What determines who does have compelling false memories?

What influences false memory formation?

Unsurprisingly, people who are prone to hypnotic suggestion are also prone to finding false memories believable [v]. Perhaps more surprisingly, the same study showed that people who are susceptible to interrogative suggestibility – that is, someone interviewing you about events implies or states that they believe an event happened – are not prone to finding false memories believable. This is sort-of-good news if you’re asked leading questions in, say, a witness interview. It doesn’t stop the problem of being asked leading questions, but it should mean that if there’s later an investigation about those leading questions, even if you gave the desired responses to those leading questions instead of saying what you truly recall, you probably won’t have formed a false belief about your experiences (which would effectively remove some evidence that you were asked leading questions).

Leading questions don’t need to have intent behind them. In some cases, they happen through poor phrasing or inattention, so people conducting taking witness statements need to be very careful. One example of how easy it is to ask a leading question is from an experiment done in the Netherlands [vi] about memories of the assassination of the right-wing Dutch politician Pim Fortuyn. There are no videos of the shooting, yet two-thirds of the people who took part in the experiment said they remembered seeing them. The researchers had asked a purposely ambiguous question, “Did you see the amateur film of the Fortuyn shooting?” so that they could later ask participants to clarify why they had said yes. Only a few really believed that they had seen the footage. Most of the rest thought that the researchers were asking about footage of the aftermath, which does exist, or said that they didn’t want to let the researchers down by saying that they didn’t remember (these are demand characteristics, a known problem in psychology research).

The type of memory you’re being asked about also matters. Most people believe that they have really accurate memories of highly emotional events, so-called ‘flashbulb memories’. These can be events in personal life, like getting engaged or hearing about the death of a close friend, but most research on them is about public events like nuclear accidents or Brexit because it’s much easier for researchers to do studies about one event that happen to lots of people rather than many events that each happen to a few people.

I strongly believe that I remember exactly where I was when I heard about the death of Princess Diana. I was listening to the radio in bed in the spare room at my parents’ house, because my room was being repainted, and my dad came in and told me about the crash in an unusually solemn voice. (My dad disputes this, but there are also elements of his memory of the event that I believe to be wrong.) But that was 22 years ago, and science tells me that though I am confident that I recall accurately, in reality that memory is only a bit more accurate than others of around the same time in my life, likely because it was shocking enough that I repeatedly revisited the memory in a way that I wouldn’t have done with more mundane events [vii].

Bird’s-eye view of an empty volleyball court

Here’s a perfect illustration of the discrepancy between ordinary and flashbulb memories: the same summer as Princess Diana died, I met my best friend for the first time. I can’t actually remember meeting her, except that we were at summer camp and I think we were playing volleyball. I find it very weird that I don’t recall, because her friendship is far more important to me than Princess Di.

There’s probably also a social component to some flashbulb memories: even if we don’t remember an event, it can come across as callous, so we might say we remember [viii]. In imagining that we do remember we then fall victim to imagination inflation.

Lastly, there are a few personality characteristics that can make us more prone to false memories. If any of the following are true of you, then you are more likely to form false memories than people for whom these aren’t true.

You experience dissociation [vii], which is essentially feeling disconnected from your sense of self or the world around you and could make you less trusting of your own memories.
You believe in the paranormal [ix], which might be related to dissociation through acquiescence, the tendency to uncritically accept what you are told.
You have a high desire to gain skills and make significant accomplishment in your life (but only if the false memory is about an achievement) [x].

This link between desire for achievement and false memories about achievements might also help us answer one final question…

What are false memories for?

One of the stranger functions of memory is that it helps us feel good about ourselves. We generally think of how we are now as ‘better’ than how we were before, even if that’s not true [xi]. Creating negative false memories of our earlier selves would therefore help us feel good about our current selves, or create a coherent sense of a continuous self that behaves in certain predictable ways over time.

There are other theories about false memories, too [xii]. These aren’t necessarily mutually exclusive; it could be that we each create false memories for several different reasons. Maybe we ‘fill in the blanks’ in the relatively sparse recollections we have of the past based on what is likely to have happened, but what is likely isn’t always what actually happened. Maybe false memories serve a social function, allowing us to grow closer to others on the basis of (falsely remembered) shared experiences. Maybe mentally revising the past actually helps us imagine and plan for the future. I particularly like this last one because, neurologically speaking, the brain is pretty much sampling the past and remixing it into possible futures.

Close-up on a DJ’s hands at a mixing desk

Like a DJ - but with thoughts!

False memories are strange, no doubt, but a lot of the time we might not be aware we have them because we don’t necessarily discuss memories with others who were there, and even if we do, everyone’s memory is imperfect and we may not be able to tell who’s remembering wrongly. I have a theory, though, that we’ve probably started becoming more aware of false memories over the last few years as many of us record the micro-events of our lives on social media, and especially Facebook. Facebook’s ‘on this day’ function often throws up vaguebooking from 10 years ago that makes me cringe, but it also gives me things that I don’t remember or have misremembered, a message in a bottle about How Things Actually Were, and I am confronted again and again with how fragile a hold I have on the past.

Now, where did I put my keys?