In this Tech Talk, Nancy Hawa, Support Team Lead here at Fog Creek, explains why you can’t trust your own memory. She explains a number of ways in which we change our memories, lie to ourselves and convince ourselves of things that never happened. It gives us to pause to reflect upon our decision making and why we really do the things we choose to do.
About Fog Creek Tech Talks
At Fog Creek, we have weekly Tech Talks from our own staff and invited guests. These are short, informal presentations on something of interest to those involved in software development. We try to share these with you whenever we can.
- Flashbulb Memories
- Memory Game
- Selective Attention
- The Weapon Focus Effect
- Cognitive Dissonance
This talk is an expansion of a talk that I gave at the Fog Creek off-site called Why You Can’t Trust Your Own Memory.
We’re going to start by talking about flashbulb memories. Flashbulb memories are memories that have fascinated psychologists for a really long time. These are memories like where were you when Kennedy was assassinated? Where were you when the Challenger exploded, or where were you when 9/11 happened? They’re the kind of memories that you tell over and over again. Where were you when this happened? We tell stories.
I heard the story of where my Mom was when she heard that Kennedy was shot maybe hundreds of times. It feels like a part of my own history. She was in class, a social studies class, and her social studies teacher made the announcement and then immediately read the class the Constitutional Order of Succession, which was his way of telling the class there is a plan. Something horrible has happened, but there is a plan. Then another teacher came to the door, a female teacher, crying and he went over to the door and hugged her. I have like such a visual memory of that that it feels very real to me, even though I wasn’t there, I’ve just heard my Mom tell that story over and over again.
We have those kinds of memories for other things as well. One thing that’s an example was the explosion of the Challenger, where it had a teacher named Christa McAuliffe on it. It was a big thing for people who lived through it, and this is the kind of thing where they told that story over and over again.
“They were in different places, doing different things, with different people”
Some researchers at Emory University, the day after the Challenger exploded handed out a questionnaire that said, “Where were you, who were you with, and what were you doing when you heard that the Challenger exploded?” The students answer those questions and then the researchers collected them and put them away for two and a half years. At the end of two and a half years, they bring the same students in and pass out the same questions, where were you, who were you with, and what were you doing? The answers varied wildly. They say that they were in different places, doing different things, with different people.
What’s more, many of those students when presented with the piece of paper that they filled out two and a half years earlier disown it. They wrote it, it’s in their handwriting, and they say, “No, that’s not what happened. I must have been confused,” and realistically that’s not the case. The memory that they wrote down 24 hours after this event is the one that’s right, but they’re so sure of this new, fabricated memory that they’re willing to disown the one that they wrote down immediately after the event.
The even more important thing is that of course there’s some variance in the degree to which students report they are certain of their memory and there is some variance in how accurate the memories are, but there’s no correlation between those two things. The degree to which they were certain of their memory is completely unrelated to the degree to which it’s accurate, and that’s pretty horrifying if you think about it, that how sure you are of a memory in your brain doesn’t relate to how accurate it is.
This is something that’s good for us to keep in mind in many areas of our life. If you think about a fight that you’ve had with the family about some disastrous family vacation or the fight you had with your boyfriend, where you can’t believe he doesn’t remember that thing you know he said and you can’t tell if he’s crazy or lying, this is what’s at play.
What researchers have found is that this is particularly likely to be true in memories that have high emotional resonance for us. It’s really hard to study this in a lab because you can’t knock over the Twin Towers or blow up a space shuttle to be able to study it, so what they do is they bring people in and present them with things that they are meant to remember and either deliver electric shocks or don’t while they’re looking at those things in order to increase the emotionality of the memory and they find the same thing.
The memory of the actual events is crystal clear. People remember that Kennedy was shot, they remember that the Towers fell. The students remembered without a doubt that there was a teacher on the space shuttle, that was a very clear memory, but memories of the peripheral details are inaccurate, but their certainty of the central details bleeds out to the peripheral details, even if those peripheral details are not as accurate as the central ones.
That’s one way that our memory can fail us, because it’s highly emotional and we become certain of details that we weren’t really paying attention to. There’s another way that our memory can fail us as well, and to do this I’m going to ask you to watch a video, so if you’re not looking at the screen I’m going to ask you to look at the screen and I want you to count how many times the players wearing white pass the basketball, so you’re watching this and I just want you to count the number of times players wearing white pass the basketball. It should be taxing. Try paying attention.
Some of you probably already got the trick. Just watch the video again now, the same video. Watch the video again. Don’t pay attention to who’s passing anything. This is the same video.
Is there anyone who will admit that the first time they watched this video they did not notice a gorilla beat its chest and walk off screen directly in their line of vision? This experiment has become to be known as the invisible gorilla. Lots of people don’t notice the gorilla. The reason why is that as human beings we’re very bad at attending to two different things. I told you to pay attention to who’s passing the basketball. You were paying attention to that and you don’t notice something else.
“How sure you are of a memory in your brain doesn’t relate to how accurate it is”
This is actually not considered part of memory research, but the reason that I think it’s related to memory is that many people, if you ask them did a gorilla walk into the middle of the game, beat its chest, do a little dance and then walk off, they would say not only no, but they would say, “I would remember that. I’m certain that that didn’t happen because that, I mean for sure I would if that happened, I would have noticed.” They’re wrong and they’re becoming sure of a memory that’s inaccurate. That’s called the invisible gorilla experiment.
I had a conversation with Emanuele recently. Emanuele is on the Gomix team, and reminded me of this. This is a screenshot and there’s this red sign above this thing and we were having a problem. I was talking to Emanuele about it and I said I want this to beat me on the head because I don’t notice it, and he said, “Yeah, I get what you mean. The red bubble is really hard to see apparently”. I don’t know how it could be hard to see, but it is. The point I want to make here is that I think that red bubble could literally be a gorilla beating its chest and people would still not notice it because you’re attending to something else. It’s a similar thing with Coding. Coding is probably more mentally demanding for many, if not all, people than counting, but it is the case that we know that many people have trouble dividing their attention between two different things and noticing something like this.
I remember having… I think I remember. Everything is uncertain in this context, but I think I remember having a conversation with Emanuele eight months ago where I was saying I had this problem, like I’m trying to report a bug, and he said, “Was it working? Did you get the red bubble?” I am fairly certain that what I said is, “No. I would have noticed. I would have remembered that,” which means that my bug report was completely meaningless because I might not have noticed. I might not have remembered, and I discovered later that there were many times that I was seeing this and not noticing it.
Selective attention is what the field of research is called. It’s something that really does affect us every day. As a teacher, especially early on, if I was working one-on-one with a student, the other kids could have been burning down my classroom and I wouldn’t have noticed. My attention was on the one kid.
I remember in my first year of teaching there was a week where I caught like eight kids with cell phones. It was before I worked at a school with a metal detector, and I was really annoyed. I said, “I don’t understand what’s happening. All year we haven’t had this problem and all of a sudden you’re all using your cell phones all the time.” One of the kids said, “No. We’ve been using our cell phones from the beginning of the year. You just started noticing.” It’s because if you’re doing something that’s demanding of a lot of your attention, like a first year teacher who’s overwhelmed, you just can’t pay attention to the peripheral stuff. Once I got my feet under me a little bit, I started noticing that all the kids were texting through class the whole time and probably not learning very much for the four months that preceded that unfortunately.
The Weapon Focus Effect
This is something called the weapon focus effect. We know that witnesses to crimes… If there’s a weapon at the crime scene, they see only the weapon and they’ll kind of make up details and fill them about the gender or race or what the perpetrator was wearing, but really what they are looking at is the weapon and that’s the only thing that they’re attending to, but we trust what they say about the other things. Eyewitness testimony is considered the gold standard. If you’re a prosecutor, you can’t do any better than an eyewitness, but they’re notoriously unreliable.
Both for that reason and because it’s possible for people to develop false memories based on suggestion. Elizabeth Loftus, a famous researcher who did research where she could implant a memory of a stop sign where there had not been one before and witnesses in this very experimental setting would report that they saw… I think actually it was a yield sign, when there had not actually been a yield sign. Then she took that step further and… She did this because at the time recovered memory was a big area of psychological research and psychologists were helping patients to recover memories of abuse from their childhood. The theory was that someone might have been sexually or physically abused as a child and they repressed that memory and then through therapy came to recover that memory and be able to confront that situation.
“If you’re a prosecutor, you can’t do any better than an eyewitness, but they’re notoriously unreliable”
What Elizabeth Loftus wanted to know was are these people recovering those memories or are those therapists implanting those memories in those patients’ minds. It’s an important question, because in theory if you’re going to go to therapy having never been the victim of abuse and then end up walking way feeling that you are the victim of abuse. It tears apart families. It ruins people’s lives, because the trauma is there, even if it’s false.
What she found… Again, you can’t study this directly because you can’t try to implant in people a memory of having been sexually abused [for ethical reasons], but she found that she could through techniques similar to therapy implant in people a memory of having been lost at the age of six in a crowded shopping mall and having been rescued by an elderly person. Because she went down this line of research, she was very, very personally attacked by the community of psychologists who believed in recovered memory. It became a very personal thing. They accused her of having fabricated her data, which she did not. Her career was on the line for a long time. She is very respected now and she held to her guns, but it was very, very personal… People on both sides felt so personally about it because the psychologists who were recovering these memories felt certain that they were helping people and the psychologists on the Elizabeth Loftus side of it felt certain that they were actually doing harm.
Another thing that I want to talk about is cognitive dissonance. We just talked about other people doing things that will change your memory of something that has happened. Cognitive dissonance, the dissonance refers to like there being some kind of tension in your mind between two things. There’s this classic study that’s done. They have people come into a lab and do this for 40 minutes. I’m just going to show you a little bit of this. They did this for 40 minutes. They were just turning these spools.
It’s designed to be extremely, extremely boring, like if you can imagine just turning those knobs for 40 minutes in a row, and then they were probably paid a nominal fee. I used to do these things in College. You go and you do a study like this and you walk out with a $5 bill. The researcher, who’s not actually a researcher, he’s an actor who doesn’t know what they’re trying to study, but this guy who’s here in the image now says to them, “Listen, we really need people to participate. It’s really important, so we’d like you to go outside and tell the person in the hallway that this was really fun so that we can get them to participate. If you do that, I’ll give you either $1 or I’ll give you $20.”
They all do it. They all lie and say this was super fun. Then they bring them back into the lab and they say, “Can you tell us what you thought of the task?” The people who were paid $20 said, “The task was really boring. It was horrible,” and this was a long time ago. $20 was a lot of money at the time. The people who were paid $20 said, “This was really, really boring. I hated it.”
The people who were paid $1… Now there’s no money on the line at this point. They know that they’re talking to the researcher. The people who were paid $1 said, “I really enjoyed it. It was pretty fun. I found it to be really engaging.” The theory is that the reason why is that the people who were paid $20 don’t have any tension in their brain. That lie was worth $20. It was a lot of money. Yes, they lied, but they don’t feel bad about it because they got enough out of it.
The people who were only paid $1 feel bad. There’s a tension, like was it worth it for me to lie and steal 40 minutes of this next person’s life when all I got out of it was a dollar? They have more of a tension in their brains that they have to reconcile, and the way they reconcile it is by convincing themselves that they had fun during those 40 minutes, so when they say, “Yes, I had fun,” they mean it. They’re telling the truth, but it’s like they’ve lied to themselves so that they can still feel like good people.
There’s another study like this where they take non-science majors and science majors and they are either wearing lab coats or not, so there’s four groups, non-science majors not in lab coats, non-science majors in lab coats, science majors not in lab coats, science majors in lab coats. They say, “Could you give us a list of your top 10 favorite record albums in order from your favorite one to your 10th favorite one?” Then they say, “Okay, great. You can have your sixth favorite one.” That feels kind of crummy because they’ve written their top five above that.
“They’ve lied to themselves so that they can still feel like good people”
Then they move on. They do something to kind of distract them. Then 10 minutes later they say, “Could you just write down for us one more time what your top 10 record albums were?” For the people in these three groups, who I’m calling fools, the number six moves up, so they say that it’s their number six album, they’re told that that’s the one that they get to keep and that feels kind of crappy, and then 10 minutes later when asked the same question, they’re like, “Actually that’s a fantastic album. That is now in my top three,” and they’re just reporting what their top 10 record albums are.
These people, who I’m calling not fools, don’t report any change. They still get to keep the same album. They still are keeping number six, but it stays at number six on their list.
The theory for why is that the science majors who were given lab coats have something else making them feel good about themselves while this is happening. They feel really cool already because they’re getting to wear a lab coat, and that means something to them, and it doesn’t mean anything to the English majors or Humanities studies majors. Because they have something that is affirming their self-worth they don’t need to tell this lie in their minds about which is their favorite record album. Again, this really isn’t in the field of memory. It’s sort of like you can’t… Not just you can’t trust your own memory, but you can’t trust your own thoughts about the world because something as simple as you getting your fifth favorite album or sixth favorite album, is enough to catapult it to the top because you want to feel good about that.
We do this with lots of our decisions. We make a decision and all of a sudden it feels like the right decision. It feels like fate, and this is something that we tell ourselves. We suddenly forget all the reasons there were to not have made that decision. It means that people are not very good at being reflective about decision making because they’re not good at reflecting on what the other options really were and it means that there can be a little bit of denial about that because you’re able to fool yourself without knowing it.