After taking my kids to see Moana recently, I was struck by how differently they relayed the story to their dad than I did – different parts stood out to them as being especially funny or scary, and they used very different words to describe the plot and characters. Thus, it is mind-blowing to consider a new study that suggests that people share remarkably similar patterns of neural activity when they recall the same event. Granted the study was with adults only, but it found that an individual’s neural activity while retelling a story was more similar to another person’s activity during recall than it was to brain activity when the person first experienced the story.
In the study, published in Nature Neuroscience, researchers asked 17 adults to watch and listen to a 50-minute show, BBC’s Sherlock, while undergoing an fMRI scan. They then had unlimited time to speak about any aspect of the show. The researchers’ goal was to compare activity between the brains of different people during movie-viewing and spoken recall, as well as to identify transformations of brain activity between when they watched and recalled the show.
Led by Janice Chen at Princeton University, the research team found that that neural patterns when someone is experiencing something are altered systematically across people into “shared memory representations for real-life events.” Writing about the study in a News & Views column in Nature, Eva Zita Patai and Hugo Spiers of University College London said: “Their findings imply that humans share a remarkably similar neural circuit for transforming experience into memory.”
I spoke with Chen about the study, why shared memory studies are important, and how the findings could shed light on why we remember certain things better than others.
CNS: Why is studying shared memory and individual differences in memory important?
Chen: The goal of science is to describe the whole world. That means we are not only interested in how one individual person’s memory works; we also should be explaining how people compare to each other, and how their interactions with each other shape their memories.
CNS: What have we known previously about neural activity as it relates to shared memory? How did your study approach the topic differently?
Chen: Many prior studies have shown that when an individual remembers a prior event, activity in the brain is reinstated: for example, similar activity is found when studying a picture and later remembering it. However, those analyses were always done within an individual, while in our study we show that you can compare directly between the brains of different people. In other words, people’s brain activity during remembering is a lot more similar to each other than we expected. On top of that, we used a very realistic type of perception and behavior: watching a movie and then speaking freely about it for many minutes.
CNS: Why and how did you choose Sherlock for the movie clip?
Chen: From previous experiments, we knew that only movies with complex narrative content would drive across-subject synchrony in the high levels of the brain that we were interested in. For example, a boring video of people walking around in a park with no storyline will only synchronize low-level brain regions (like early visual cortex); we needed a richer narrative like Sherlock to drive the higher-level areas (like the default network).
CNS: What were your most excited to find? Were any findings surprising?
Chen: First, we were surprised at how great people’s memories were. Many people talked for over 30 minutes to describe the 50-minute movie. This is one of the consequences of using a highly realistic task: It would be nearly impossible for people to view and then report back a 50-minute list of random words, but a 50-minute story is easy to retell.
Also very interesting was the fact that each movie scene seemed to generate a distinct pattern of brain activity, meaning that activity was relatively stable within each scene, and we could discriminate the scene-patterns from each other very well.
I was also very surprised at how similar brain activity was across people while they were speaking aloud from memory. Even though every person used their own words to describe a given movie scene, we could still see that the distinct fingerprint of activity was similar between brains.
There are fundamental similarities between the brains of different people as they perceive and remember the real world. That gives us a good groundwork for understanding each other.
CNS: Would you please discuss the transformation of memories from perception to recall and what you learned about it?
Chen: Memory is not perfect, as everyone knows. Imagine that you met a new person, Jack, in a coffee shop this morning, when the barista got your drinks mixed up. When you think back on that episode now, you might remember it in a lot of detail, but it isn’t like literally reliving the experience in real-time. The memory is a lossy, compressed, or otherwise altered version of the original experience (perhaps you remember that Jack was wearing a blue coat, but you don’t remember his last name). Similarly, the brain pattern during memory is similar but not identical to the brain pattern from the original experience. This transformation between the original and the memory is what we were interested in.
What we found is that, in some regions of the brain, the transformation happens in the same way across different people: Just as you and I might both have forgotten Jack’s last name, our brain patterns during memory both changed in a similar way. We may have both remembered some of the more important elements of the scene (like Jack’s face) while forgetting unimportant elements (like his shoes). This was important because we showed that the memory brain pattern wasn’t just a noisy version of the original experience; instead it was altered systematically, in a way that might actually be beneficial for behavior.
CNS: What do you mean by that? How might it be beneficial for behavior?
Chen: In the paper, we found that the movie scenes for which the neural pattern was altered more between perception and recall tended to be remembered better – specifically, they were recalled by more people. One possible interpretation is that people had common knowledge of how certain events are structured – for example, what typically happens in a car chase scene – and these existing schemas guided recall, for example by allowing them to think of likely scene elements (dodging through traffic; pedestrians jumping out of the way) to trigger their own memories.
CNS: What do you most want people to understand about this work?
Chen: There are fundamental similarities between the brains of different people as they perceive and remember the real world. That gives us a good groundwork for understanding each other.
CNS: What’s next for this line of work?
Chen: There are two follow-up projects that I’m very excited about. In one, we use a model-based approach to identify stable patterns and change points during movie-viewing, which means we can show how different parts of the brain segment the movie in different ways. In another, we add another group of people who listen to a recording of the spoken recall from this study, and look at how simply listening to someone describe their memory might generate brain activity that resembles watching the original movie.
-Lisa M.P. Munoz