The Fallibility Of Memory
Memory is a constructive process. Many of us believe that memories work like cameras, where we take a photo and have that file forever. Yet that's not quite true. Whenever we recall memories, we engage in a process of reconstructing that memory. And that has real consequences on what we believe, how we behave and how we think.
But first, what is memory? Is it possible to plant false memories? And what are practical strategies for drawing on accurate information when we make decisions?
What Is Memory
Memory is the encoding, storing and retrieving of information from the past.
When we break it down, this means:
Encoding: The process of getting information into our memory system
Storing: The ability to hold the memory over time
Retrieving: The act of looking inward and accessing stored memories
So there are three stages of memory, and room for error in each one. Ever wonder why you and your friend have completely different accounts of the same event? Whenever we commit something to memory, we store it as "bits" of information. Then, when we call on the memory, we stitch together the "bits" and voila we have an event. Every time we recall that event, we add more "bits" to it. So, we not only remember the event itself but we now associate information from the last time we remembered the event. It gets pretty meta!
Basically, the problem is that there are two types of "bits" we stitch together: 1) the things that really happened and 2) the things we think could have happened. We end up with both types of information seamlessly stitched together and we swear it's what happened. To see this in action, let's look at the Deese/Roediger-McDermott Effect.
Memory Illusions In Action
The Deese/Roediger-McDermott Effect is an associative memory illusion, where people falsely remember words that were not presented. In this situation, people are presented with lists of semantically related words e.g. nurse, hospital, bed etc. After a short delay, people are asked to recall or recognise these words. Research shows that people often misremember words on the list by adding things they think should be there. For example, they might remember 'doctor' even though 'doctor' is never mentioned.
You can test this with a friend in 2mins. Show them this list of words, and give them 60secs to remember it. Then, ask them to write down how many words they remember. See if there are any "associated" words that slip in there, but are not actually on the list. And don't tell them about the effect until after!
So if we're reconstructing memories, is it possible to plant a completely false memory?
In her pioneering work on false memories, psychologist Elizabeth Loftus asked: is it possible to implant a false memory? To test this out, researchers told participants three true events and one false event. They asked family members to supply three true stories about the participant and help construct the details of a false one e.g. "being lost in the mall". This was followed by a series of interviews across a few weeks. In the first interview, participants were "reminded" of the four events and wrote everything they could remember. Two weeks later, they were asked to remember the events and identify which ones were false. Out of 24 participants, 5 failed to identify the lost in the mall event as false and picked a true event instead. This demonstrates that over time, people find it difficult to differentiate between what actually happened, and what they imagined happened.
In another study, Loftus found that the language we use to interview people influences what they remember. In this study, participants were asked to watch a video of a staged car accident and estimate the speed of the cars. They were split into two groups: one was asked "About how fast were the cars going when they hit each other?" and the other "About how fast were the cars going when they smashed into each other?". The results? Estimations were higher when "smashed" was used, with some even recalling false events that took place like seeing broken glass. So there's some evidence that memories across time and by the language we use.
So what can we do to manage memory biases better?
Always take notes. Don't rely on memory for important events. Take notes for anything that's important to you, whether that's a team meeting, an interview or for an idea.
Gather feedback quickly. Make sure you give and gather feedback while it's still fresh.
Use neutral language. When you're collecting information, use neutrally worded questions. e.g. If you're interviewing customers, don't use highly emotive words and try to ask the same questions. Don't wing it, it will skew your results!
Put it into perspective. Understand the perspective of the person providing the information.
Elaborate. Think of a time when you misremembered something. Elaboration is the process of changing or adding to material, or making associations. It's a neat trick to makes it easier to remember things! Ask yourself: When did I last misremember something? How does it help me to know this?
Tying it up
Memory is a paradox: incredibly powerful but full of holes. Like many biases, it's impossible to entirely avoid. By applying simple strategies, you can manage some of the harmful effects. Whenever you make big decisions, try not to rely on memory and look at the information you've recorded.