Memory Is (Partly) Organized Around Time
As you remember things, when they happened becomes important.
Posted February 25, 2019
One of the most important functions of human memory is to give you information you need when you need it. If you’re programming a computer, you’d like to use any knowledge you have about programming to solve problems that come up. Your knowledge about cooking or driving is less relevant in that context.
As a result, one of the major factors that influence the structure of your memory is what psychologists call semantic similarity, or similarity in the meanings of things. If you see a duck, you will remember things that are related to ducks rather than to things very dissimilar to ducks.
A second factor that affects memory is association. If you encounter a particular concept in the world, not only do you care about things that are similar to that concept, but you’re also likely to be interested in things that are often associated with that concept. If you’re thinking about ducks, you are probably also going to care about things that are associated with ducks like lakes.
A third factor that influences memory is time. Things that happen at about the same time are more likely to be related than things that happen quite separated in time. So, you also tend to remember things that occurred close together in time. If you are thinking about something that happened when you were in middle school, then you are likely to be reminded of other things that happened to you in middle school rather than jumping to things that happened to you much earlier or much later in life.
This idea that memory is organized around time has been studied in experiments in which people learn lists of words. Recalling one word tends to be related to recall of other words that appear close to it in a list.
Does it happen in people’s memories for facts they encountered as well?
This question was explored by Mitchell Uitvlugt and Karl Healey in a paper in the January 2019 issue of Psychological Science. They did two studies in which people had to recall news stories they had heard in the past. One study was done soon after the 2016 Presidential election in the United States and asked people to remember stories they heard related to the election. The second study was done in May 2018 and asked people to recall news stories that they had encountered in that calendar year. Participants recalled the story by writing a headline that corresponded to that story.
Participants each recalled several different news stories (roughly 11 on average in the first study, roughly 6.5 on average in the second). In the first study, the experimenters then identified the first date on which that story appeared in the news. In the second study, after recalling headlines, participants were given an internet search on the headline they created and were asked to identify which story they were referring to.
The experimenters were interested in the order in which things were recalled. If you recalled a particular story, was the next story likely to be one that appeared close to it in time or far away? The experimenters looked at the number of days that elapsed between the two stories throughout the sequence of stories people recalled.
In both studies, participants often recalled pairs of stories that appeared for the first time on the same day. This bunching of stories on the same day happened more often than you would expect by chance.
However, it is possible that the stories that appear on the same day are semantically similar to each other. For example, on the day following the Academy Awards, there are a number of stories about movies in the news media. On the day following a historic treaty signing, many news stories will focus on that treaty and that area of the world.
In each study, the experimenters used a computer measure of the similarity among words to determine how similar the news stories were to each other. Then, they used statistical techniques to determine whether the tendency to recall stories that appeared close in time could be explained just by the similarity among the stories.
As you might expect, similarity did influence recall. However, the whole pattern of what people recalled could not be explained just by similarity. Closeness in time also mattered.
This study suggests that we associate things with each other partly by when they occur. This organization is helpful for establishing the context of stories you hear about the past. It is one reason why movies that want to establish a particular era will often have characters listening to news stories that happened at that time.
Associating stories that occur at similar times is also helpful because they may ultimately be more deeply related than just a similarity in time. When you first hear about events, you may not have a chance to analyze them to know whether one event caused the other. By storing the events based on similarity in time, it provides a way of helping you evaluate later whether the two events share a deeper connection. In this way, the structure of your memory helps you to focus on which events you might want to consider when trying to understand the reasons why things happen.
Uitvlugt, M.G. & Healey, M.K. (2019). Temporal proximity links unrelated news events in memory. Psychological Science, 30(1), 92-104.