False Memory

The False Memory Bias has been on my mind recently due to an (un)fortunate binge-watching of Netflix’s documentary series Making a Murderer. The documentary covers the trial and conviction of Steven Avery for a crime he did not commit back in 1985, his release based on DNA evidence in 2003, and his subsequent trial for the murder of Theresa Halbach in 2005. What made this documentary different from most others that cover high-profile trials is that it is not about the actual case; it’s intended as a commentary on the legal process itself.

Watching this documentary was a grim reminder that most portrayals of court rooms in popular media fail to understand how memories work. And it turns out, the same is true in the actual legal system.

That’s scary to me. More specifically, an uneducated jury that doesn’t understand that memories are fallible is terrifying. (And prosecutors, judges, and defense attorneys that ignore the research are bad at their jobs.)

How Memory Works Link to heading

Let’s step back for a minute and review how your brain actually works. When we talk about neurology it’s important to remember that our cognition is not yet completely understood, but a lot has been fairly well-established. I will try to speak only about the generally accepted models, but the exact mechanisms for how some of these work have not been proven.

That being said, here is the basic process by which memory works:

You experience something.
You store this memory in your brain. It’s stored in multiple places.
When you try to remember that something, your brain pulls those pieces together and reconstructs the memory from those pieces. Memory is not like a video where you just ‘play it back’, rather we collect all the associated fragments and feelings that are associated with that memory and reconstruct it on the fly.
After recalling it, you then rewrite the memory back to your brain.

It sounds simple and straightforward, but it causes all sorts of problems.

Why Your Memory Sucks Link to heading

Your memory sucks because of two separate bugs in the formula that are highlighted above. When you remember something you reconstruct it from its pieces and after recalling something you then rewrite that memory back to your brain. Memory doesn’t work the same way as a DVD, it works more like a jigsaw puzzle; you grab a bunch of disparate pieces and fit them together in a way that seems to make sense, then put them back in their box.

This leads to two really big problems with memory.

First problem, you don’t know for certain if your memory is something that actually happened or if you are putting the pieces together incorrectly. Link to heading

Numerous studies have shown that it’s possible to create totally false memories in people. In one well-known study, the researchers were able to trick a quarter of their subjects into recalling a fake memory about being lost in a shopping mall as a child. They didn’t even need to work very hard.

Working with the family of their subjects, they presented them with 4 events from their childhood, one of which was the fake memory of being lost. The subjects were asked to write down what they remembered about those events and rate how confident they were with the memory being real. This was repeated a few times to see if they remembered additional details, and to further encourage their memories.

Without going into too much detail, 25% of the subjects recalled several details about getting lost in a shopping mall when they never actually did, and several more thought they remembered getting lost but were unsure. And when asked how confident they were of it being real, some of them were very confident.

It makes you wonder how many of your early childhood memories actually happened, and how many are there just because your parents told you about them. When I was 4 my family went on vacation, and while there I got sick and had to be taken to the hospital. For a toddler this is a fairly traumatic memory so you’d think I’d remember it, and you’d be right. The only thing I recall about that trip is a distinct memory of being driven to the hospital.

The only problem is, the streets I remember driving down are in my hometown, which is obviously wrong.

I’ve heard this story told to me so many times that I recall something that I know is wrong, that I know didn’t happen like I remember, but I still remember it.

Second problem, it’s possible to modify the memory as you recall it and ‘save’ it back again. Link to heading

Let me take an example from Making a Murderer. I’m only going to deal with the uncontroversial false conviction of Steven Avery in 1985. The documentary covers this event in the first episode of the series, which you can watch in its entirety on YouTube.

For those that haven’t watched it, I’ll summarize:

In 1985 Penny Beerntsen was raped by a blonde haired man with a beard that bore a slight resemblance to Steven Avery, a man known to the police department at the time, but who had a very strong alibi for the time of the incident. Despite this, Steven Avery was convicted of the crime. Why? Because Penny was convinced that he did it.

Of course it’s not remotely her fault, her memory was effectively altered by the police’s evidence gathering process. Here’s the relevant sequence of events that happened after she began to give her statement:

She describes her attacker roughly as blonde haired and with a beard, along with other descriptors.
Someone at the Sherriff’s office says “That sounds like Steven Avery.”
Penny works with a sketch artist to produce an image of her attacker.
The drawing produced looks exactly like an old picture of Steven Avery that the police department has on file, even though his current hairstyle (and that of the real attacker) was very different. (The filmmakers argue that the police department did this intentionally to finger Avery. For our purposes it doesn’t matter how it happened, just that it did.)
She picks Avery out of a line-up that includes him and other men that all appear dark haired.
She continues to be convinced through the trial that Steven Avery was her attacker.
18 years later, DNA evidence proved it was a different individual responsible.

Before you start thinking that it would be really easy to confuse two men who look vaguely similar, realize that the real attacker was almost a foot taller than Steven Avery, who is 5’2″.

As you can likely tell by reading this cherry-picked sequence, the problem is that the police department influenced Penny’s recollection of the event. They failed twice here in ways that reinforced each other. First, the sketch they made for her is clearly of Steven Avery. That helped solidify in her mind what her attacker looked like even though it was not entirely accurate. Whether by coincidence or not the sketch produced looks like Avery and not the person responsible. Second, they placed him in a line-up for her to pick out her attacker from. There was one man in the line-up that matched the picture the sketch artist made, and vaguely fit her description, so she was forever convinced that that was the guy.

Think about it in terms of what we know about how memory works. Each time she was introduced to the image of Steven Avery, she was recalling the memory and trying to fit the face she saw onto her attacker. In the line-up especially, she knows that the police think they have the right person, so she picks out the person that looks close to her attacker. It’s not her fault that she associates the image of Steven Avery with her attacker, the police as much as told her that he did it. But when she re-wrote that memory back to her brain, her modified memory now showed her that Steven Avery was the one doing it.

Her memory alone was enough to convict despite the fact that he had 19 witnesses and several store receipts as proof that he wasn’t there. Unfortunately the prosecutor, judge, and jury did not understand how inaccurate memories can be.

How We Can Use This Link to heading

At first there doesn’t seem like a whole lot of useful takeaways from knowing about this effect. You almost certainly won’t change anything about how you see the world, but it’s very good to be aware of. I’m not saying don’t believe witnesses, just keep in mind that someone can honestly be 100% certain about what they saw but also be 100% wrong. This issue comes up a lot with regards to early childhood abuse cases; it’s very easy to accidentally ‘implant’ a false memory of an early childhood event.

One way I do know how to use this has been explored by Jane Mcgonigal, a video game researcher (sort of? I’m not sure how best to describe her). Basically, research has shown that it’s possible to use Tetris to treat PTSD. I highly encourage you to follow that link and learn more, though the gist of it is this: If you experience or witness a trauma, play a pattern-matching videogame such as Tetris or Candy Crush Saga as soon as possible, ideally within the first twenty-four hours after the event. The research shows that the visually absorbing games can prevent your brain from concentrating on the event, so your brain is prevented from storing the memories as vividly as they normally would. You can still recall the events afterwards, but you are less likely to suffer unwanted flashbacks.

Potentially you could try it after the event as well, by recalling the memory and attempting to modify it, but that is not as effective as disrupting the storage of the memory in the first place. Interestingly, if you’ve seen The Hunger Games movies they provide a reasonable description of doing this to one of the characters, though in the movie they do the opposite and transform someones happy memories into traumatic ones…

Other than being aware of this while dealing with your own and others memories, I don’t really see many everyday takeaways from knowing this even though I’m sure there are some. Let me know in the comments what you come up with.

(Pulling your hair out over the witness treatment in Law and Order doesn’t count.)