“I remember exactly…” – we all know this phrase from various quarrels with friends or spouses, discussing issues like who paid the bill or washed dishes, or whose mother-in-law was the first to act inappropriately. But scientific data from the last three decades show clearly that this half-sentence is misleading — we never remember exactly.
Our memories are not like the recordings of a video camera. What’s stored in memory is never a one-to-one copy of what really happened in the world. The way our brains are built influences what we perceive, and what is perceived biases what will be memorized. The twist is that our brains are structurally different from each other, and these differences mean that even exactly the same event will be remembered differently by two persons.
In addition, we never memorize everything. Even really important events can slip away unnoticed. A classical experiment by Simons and Levin[ref]Simons, D. J., & Levin, D. T. (1998). Failure to detect changes to people during a real-world interaction. Psychonomic Bulletin & Review, 5(4), 644-649.[/ref] found that when the participants’ attention was momentarily disrupted, they didn’t notice that the person they were talking to just a moment ago had changed, although the change had been total — it was now a completely different person. Our brains record in a selective manner.
One can recall events from memory, but it is important to know that the episodes reconstructed in this way never precisely mirror what happened when the content of the memory was stored. That’s because these contents are not being saved into a safe deposit box or a hard drive, but into a network of nerve cells that are constantly at change.
If experiencing other events even partially modifies these very same networks, the original memories are modified too without any option to “undo”. And this is true for events we consider important or noteworthy as well. Storing new memory content changes the old memories. It sucks but that’s how the brain works.
The worst thing is that each time we recall some content from the memory, it will be partially overwritten. Although we intuitively tend to assume that the more one recalls a specific event, the better and more precise the future recall of this event may be, studies show that this overwriting can substantially change the original memory. So the next time you witness a robbery, don’t start discussing and thinking about it, but record the contents of your memory as quickly as possible to a more reliable medium.
Classic experiments demonstrate how the claims of eye-witnesses can change depending on the way the question is asked. For example, Loftus and Palmer[ref]Loftus, E. F., & 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.[/ref] showed an identical film clip about an accident that occurred between two cars to different groups of participants. After seeing the clip, every participant was asked to evaluate at which speed the cars were driving when the collision occurred. Some people were asked “How fast were the cars going when they hit each other?”, while others heard “How fast were the cars going when they smashed each other?”
Those who were addressed with the word “smashed” estimated the speed being greater, although all participants had seen the exact same film. Not only that, the participants whose question included “smashed” “remembered” that the video had broken glass in it, when there wasn’t anything like that in the film.
These old results already demonstrate that the contents of your memories can be changed without you noticing it. A new study by Shaw and Porter[ref]Shaw, J., & Porter, S. (2015). Constructing Rich False Memories of Committing Crime. Psychological science, 0956797614562862.[/ref] shows that even perfectly healthy young persons can, with some clever suggestive questions and too much time to think, be led to believe that they had committed a nonexistent crime in the past. That’s not all: at the end of the experiment the participants claimed that they really did remember such an occurrence (that actually never happened) and describe this false memory as vividly as any real memory. New content had come to life in their brains, and they mistakenly considered it as old.
These facts about the unreliability of memory have substantial, far-reaching importance in court practice, for example, where the eye-witnesses’ claims and the culprit’s confessions play a central role. But this knowledge is actually important to all of us. Tedious quarrels would not take place if the opposing parties would acknowledge that their memory doesn’t work like a video camera. Maybe it really was your mother-in-law who first picked a fight, but who knows – maybe nagging and thinking about it has already changed your brain and your memories? Perhaps one should get a personal video camera to record it all?
Jaan Aru is a researcher at the Computational Neuroscience Research Group and at the Talis Bachmann Lab at the University of Tartu. The Estonian version of this post was originally written for the ‘Psühholoogia Sinule’ magazine and appeared on Jaan Aru’s personal blog.