False Memories

Do you remember where you were when you first saw the closed-circuit TV footage of the 7/7 London bombings in 2005?

Hopefully not, else you may be imagining things — no such footage exists. But if you claimed to remember it, you would be in good company. Around 40% of British college students said they remembered such a video, when filling out questionnaires a mere three months after the bombings. It seems as if people had invented a memory to fill in or coalesce the details of an event they had seen or heard described later.

Psychology professor Elizabeth Loftus has studied false memories like these for a while. For example, one study she worked on showed participants a Disneyland ad with Bugs Bunny in it. Almost a third of people who had been to Disneyland at some point in their life falsely reported that they had met Bugs Bunny and shook his hand there — falsely because Bugs Bunny is a Warner Brothers character. The false memory was much less likely in people who were shown the same Disneyland ad without Bugs Bunny in it.

Another recent study by Loftus showed participants a photograph from the 1989 Tiananmen Square protests. However, for some participants the photograph had been doctored, adding a large crowd to a familiar picture of a man standing alone in front of tanks. Asked whether the photograph they saw was familiar, those who were shown the doctored photo responded the same as those who were shown the original image, so they didn’t seem to detect the change. However, those who were shown the fake photograph reported remembering more people being at the protest. A subtle, undetected alteration to a famous photograph easily distorted peoples’ memory of the event.

In all of these cases, people are not necessarily inventing memories out of thin air. Even in the first example — of CCTV footage for the 7/7 bombings — the participants were actively prompted by the questionnaire about footage for a significant event, which itself might subconsciously suggest to them that such footage does exist, and that they should remember it or would be likely to have seen it. In other words, the memory came from prodding or priming, not recalled on its own. Likewise, the Bugs Bunny and protest studies found false memories mostly in those exposed to a fake ad or photograph.

In other words, these studies are not about how generally poor our memory is, but about how susceptible our memory is to cues, suggestion or persuasion.

Unfortunately, our memory problems may go deeper, even when there is no false information (or subtle suggestion of it, as with the 7/7 bombing questionnaire). In 2007, Norbert Schwarz and others showed people a CDC flier listing common views about the flu vaccine, some labeled true, some false [pdf]. 30 minutes after reading the flier, they were asked which statements were true and which false. Young people did pretty good, but older people misremembered 28% of the false statements as true. Questioned again three days later, older people misremembered even more of the false statements as true, and even younger people ended up doing as bad as the older people had after 30 minutes.

Mere exposure to false information — even clearly labeled as such — can cause people to believe it. Trying to correct a misconception may in fact strengthen it, which has disheartening implications. False news stories may stick in the public’s mind as true even if clear evidence corrects it later. Note, for example, how many Americans still believe Saddam Hussein was involved in the September 11, 2001 attacks.

Frankly, our memory is full of biases that make us susceptible to false recall and recognition. Knowing about these may be a first step to minimizing their effects.

I think the bigger lesson here, though, is in coming to understand that our memory does not work how we commonly think. It is not like a computer, where whatever is put in is retrieved in the same form. Rather, our memory is constructive: whenever we retrieve a memory, we fill in the gaps with plausible information that fits our schematic representation of the event in question, or with information that has been cued externally (say, by an altered Disneyland ad). When you recall a vivid experience from young childhood, you may not be directly recalling the event itself, but rather bringing to mind the story you have told (or been told) again and again, and reconstructing the memory out of the oft-repeated story.

Accepting the inaccuracy of our own memory is a hard pill to swallow. Without our memories, a large portion of our personality is gone, so calling your own memory into question is like calling a significant piece of yourself into question. But it’s something we must learn to do if we are ever to understand how memory really works and immunize ourselves as much as possible against the false information of propagandists, marketers, and anyone else who would use our memory biases against us.