Time Warped: How Repetition Distorts Our Sense of Duration

Why do familiar experiences sometimes feel like they happened longer ago than they actually did? 

In this episode of Under the Cortex, host Özge Gürcanlı Fischer-Baum speaks with Brynn Sherman from the University of Pennsylvania about her recent study published in Psychological Science, the flagship journal of the Association for Psychological Science. Sherman’s research uncovers a surprising illusion: Repeated experiences, which are more vividly remembered, are often perceived as having occurred further in the past than they did. 

Through a series of experiments, Sherman and her colleague Yousif demonstrate that this distortion in time perception is both robust and consistent, shedding light on how our memories can mislead us about the timing of events. Tune in to explore the mechanisms behind this illusion and its implications for our understanding of memory and time. 

Send us your thoughts and questions at [email protected].

Unedited Transcript

[00:00:09.140] – APS’s Özge Gürcanlı Fischer-Baum 

Have you ever felt like time was slipping by faster than usual, especially when your days seem repetitive? Today, we are diving into new research that explores how repeated experiences can alter our perception of time. This is Under the Cortex. I’m Özge Gürcanlı Fischer-Baum with the Association for Psychological Science. I’m joined by Brynn Sherman from the University of Pennsylvania, the lead author of a recent article in Psychological Science that investigates the Illusion of Time and how familiarity might actually compress our mental time. Let’s get into it. Brynn, welcome to Under the Cortex. 

[00:00:46.720] – Brynn Sherman 

Thank you so much for having me. 

[00:00:49.280] – APS’s Özge Gürcanlı Fischer-Baum 

Can you tell us a little bit about yourself? What type of psychologist are you and what drove you to questions about time and memory? 

[00:00:58.220] – Brynn Sherman 

I would refer to myself as a cognitive neuroscientist. I’m really fundamentally interested in questions about human cognition, human behavior, and in particular, memory. But I take a multi-modal angle. I use a lot of different tools from human neuroscience, from some computational modeling, and then, of course, just from measuring human behavior to try to get a comprehensive understanding of human cognition. 

[00:01:25.440] – APS’s Özge Gürcanlı Fischer-Baum 

Let’s start with something familiar. Why does a memory from last week sometimes feel like it happens month to year? 

[00:01:33.560] – Brynn Sherman 

Great question. In part, I don’t think we totally know the answer to that. There just seems to be a ton of ways that our memory for time, our perception of time can be distorted. And what I would argue is that we don’t really have good memories for time at all, and we’re often having to rely on other cues that we can use to pinpoint time in our memory. So we’re thinking, Oh, that happened after this other event happened, and so it must have been around this time. I think when you’re relying on these external cues rather than having an actual sense of when something happened in time, that can cause these feelings like something Something that has been happening occurred forever ago, even when it actually just was last week. 

[00:02:19.180] – APS’s Özge Gürcanlı Fischer-Baum 

Yeah. In your paper, you explain this using the term temporal repetition effect. Can you talk a little bit about that? How did you first start thinking about it. 

[00:02:31.940] – Brynn Sherman 

Yeah. The temporal repetition effect refers to basically the finding that we found, which is that when information is repeated many times, and then you ask someone, When did you first encounter this piece of information. People misremember that as having occurred earlier in time than it actually did and earlier in time than something that occurred right around the same time, but which was never repeated. So specifically, repeating an experience causes to be remembered as pushed back on your mental timeline. And we were really inspired by just our own subjective experience. This was very much inspired by just thinking about hearing the same headlines, seeing the same tweets day after day after day. And when you’re bombarded by all this repetitive information, we just had this feeling like things actually seemed… It seemed like we first encountered that months ago when actually it might have just been that week that that headline happened. 

[00:03:30.000] – APS’s Özge Gürcanlı Fischer-Baum 

That’s very interesting. It is like the frequency expands that time in our minds, that representation. Your research shows that repeated events can feel older than non-repeated ones like we just talked. That feels backward. Can you explain why? What is the mechanism behind that? 

[00:03:49.780] – Brynn Sherman 

I definitely see why you would say that it feels backwards. It actually is backwards to a lot of findings in the literature, which have found that when you repeat information, it actually seems more recent. The idea behind that work is that when you repeat something, each time you repeat it, you’re increasing the strength of that memory. Then when you’re trying to remember when it occurred, you have this really strong memory, and a strong memory is going to seem recent. And so you would actually see the opposite result, which is that repeated things would seem more recent in memory. But in those cases, they’re always asking about, When did you see this most recently? Whereas what we’re doing in our study is asking people, When did you see it first? And so it seems like maybe there are these separate memory traces for when you first encountered something, which we’ve shown now is pushback earlier in time, versus when you most recently saw something which seems to be even more recent. It seems like you said that there’s a stretching along our mental timeline where repeated things are remembered as more recent and also remembered as earlier in time. 

[00:04:58.620] – APS’s Özge Gürcanlı Fischer-Baum 

Yeah, that’s fantastic. And across six experiments, you found this illusion in different contexts. How did you measure it in a way that felt natural and realistic? 

[00:05:09.860] – Brynn Sherman 

Yeah, so our base experiment was pretty pared down. We were just showing people’s sequences of objects. They would just see an object on the screen one at a time just over the course of a few minutes. We would show some of those objects multiple times to induce that repetition structure. But our most naturalistic experiment, we were really trying to understand whether this temporal repetition effects that we found could explain the types of behaviors that we were interested in that evolved over the course of days, weeks, months. We ran a version of the experiment where instead of just having people view this sequence of images over the course of a few minutes, we broke up that encoding session over the course of a week. In this case, the repetitions that people were experiencing were happening across days. So more at the time scale of the behavior we’re trying to We found evidence for the temporal repetition effect there, even when people were doing the study over the course of a week. 

[00:06:07.060] – APS’s Özge Gürcanlı Fischer-Baum 

What did you find basically in very short terms? 

[00:06:10.660] – Brynn Sherman 

Basically, we showed people this sequence of images. After they viewed it, we just gave them a timeline, a very open-ended response timeline, where the left hand of the timeline said beginning of the experiment, the right-hand of the timeline said end of the experiment. We just asked people to place these objects on that timeline. What we What we found is that people would place the images that were repeated as having occurred earlier in time. They’re actually placing them as closer to the beginning of the timeline. We were asking people to indicate when they first saw that image. 

[00:06:44.080] – APS’s Özge Gürcanlı Fischer-Baum 

That is like an indication for expansion. They think, Oh, yeah. I first saw this object really when I started this task, in a way, right? 

[00:06:55.920] – Brynn Sherman 

Yeah. So not necessarily expansion because we weren’t measuring people’s memory for every single time they saw it, but evidence for a pushing back in time. They’re remembering the first thing as having occurred even earlier than it did. 

[00:07:08.920] – APS’s Özge Gürcanlı Fischer-Baum 

Pushing back in time. That’s what I meant, but you said it better. Yes, that’s very clear. Thank you. I would like to get a little bit deeper for the experiment. You controlled for things like event boundaries and memory strength. What was your motivation for that? 

[00:07:30.340] – Brynn Sherman 

Yeah. Like I said at the start, there are so many different factors that seem to influence our memory for time. Event boundaries have been a really… There’s been a lot of recent work on how the way that we structure our experiences into events influences time. And one key aspect of our study design is that participants would do these little mini blocks. They would see a subset of images, and then they would have a little break, and then they would see another subset. And the repetitions would occur across the different subsets. One concern that we had is that maybe this is all just about event boundaries, the fact that the repetitions are happening across these different events. And so that was the motivation for trying to control for that. So we just got rid of the subblock structure and had participants view one long sequence. And in terms of the memory strength, like I said, that has been used as an argument for why you see the opposite effect, which is that when you have repeated experiences, then you have a stronger memory trace and that could render it to seem more recent. And so we did find evidence for that. 

[00:08:37.240] – Brynn Sherman 

People were better at recognizing the repeated images. So they did seem to have stronger memory traces for those repeated images. But We actually saw that the temporal repetition effect was stronger when we only looked at those trials that were remembered correctly. It doesn’t seem like memory strength is only explaining the fact that memories can seem more recent, but it also, in part, might be explaining our effect, which I think is interesting to this dual memory trace idea. 

[00:09:06.480] – APS’s Özge Gürcanlı Fischer-Baum 

Yeah, I want to say to our audience that you have really very creative experiments, six experiments and they are a really complementary set. We won’t have time to go into the details of all of them, but I highly recommend the audience to go back to the article and read the details of these creative experiments. I would like to talk about the follow-up survey. What did your follow-up survey show about people’s strategies? Were they aware they were using repetition to infer time? 

[00:09:39.880] – Brynn Sherman 

We designed this follow-up study because we were thinking, Is this repetition effect, or is it just that people are aware that information is repeating and they’re thinking to themselves, Oh, if I saw this five times in order for me to have seen it five times within this span, the first presentation must have been very early on. We were thinking, do people have insight into the ways that they could have used time to have a strategy in the task? We had this follow-up survey where we just asked them first and just describe what strategies you were using. Then we asked some more targeted questions to try to get at whether they really had insight into this. We just saw a huge range of responses. In the free response, I think there were maybe two or three people who identified using the repetition structure to guide their judgments of of time. Again, when we asked these more targeted questions, it was really just a huge spread. I think the range was truly from zero to 100. It’s hard to say across the board whether people did have insight into it. But importantly, the extent to which people did have insight into it didn’t seem to explain our effect. 

[00:10:47.980] – Brynn Sherman 

It wasn’t just that the people who were using this strategy were the ones who showed the effect. 

[00:10:54.140] – APS’s Özge Gürcanlı Fischer-Baum 

This is really interesting. Self-awareness of the temporal repetition effect varies use this, hugely, among participants, and it is not a predictor of how they use it in your experiments. 

[00:11:07.140] – Brynn Sherman 

Correct. Yeah. 

[00:11:08.820] – APS’s Özge Gürcanlı Fischer-Baum 

That is quite interesting. I wonder why this metacognition about this doesn’t affect it. What do these results mean for our everyday lives? How we remember routines, habits, or conversations? 

[00:11:24.640] – Brynn Sherman 

I think things like routines and habits, these are obviously very repetitive things. I think our memory for them is not so much affected by the temporal repetition effect. But I do think that there are a lot of other ways that very repetitive sequential structure can influence our memory for time. But I’m not sure that the temporal repetition effect necessarily speaks to that. But I think that in conversations, in reading headlines of news articles, seeing information on social media, I think these are very pervasive ways where we really can fall prey to repetition affecting memory for when something occurred. 

[00:12:03.740] – APS’s Özge Gürcanlı Fischer-Baum 

Would you say your findings challenge traditional models of memory? 

[00:12:09.580] – Brynn Sherman 

I think so. I think especially with what I was talking about before with the fact that a lot of people have shown this opposite effect and related that to this memory strength idea where the stronger your memory is because it was repeated, the more recent you’re going to think something occurred. Our results really cannot be fit into that theory at all. I think that what our work suggests is that there are many different routes to remembering temporal information in memory. And even for a given item, there can be distortions in the opposite directions where you might remember it as the most recent thing is being more recent and the earliest repetition as being even earlier. I think there needs to be more theories to accommodate the bi-directionality of that effect. 

[00:12:57.260] – APS’s Özge Gürcanlı Fischer-Baum 

If someone asked, Can I trust sense of time? What would you say, based on your findings, this bi-directionality effect? 

[00:13:07.660] – Brynn Sherman 

I personally don’t think we can trust our sense of time too much. I think that probably most of the time, this basic mechanisms that we have to sense and remember time are probably going to get us mostly to the correct answer. But I do think there are so many factors that can really influence and distort our sense time, and I think that’s very interesting. 

[00:13:33.880] – APS’s Özge Gürcanlı Fischer-Baum 

What is next for your research? Are there new illusions of time you are curious about? 

[00:13:39.560] – Brynn Sherman 

Yeah. Some work that I’m doing right now is trying to understand how our memory for time basically within an experience can be distorted. So a lot of work, including some of my previous work, has suggested that if you have these, like I said before, if you have these event boundaries that can influence your sense of time. So if you just break an experience into two discrete events. In the case of my previous research, we were literally just showing people a colored square, and the color of the square would change color halfway through. And if you just had that color switch, that massively caused people to underestimate how much time had passed. There’s a lot of work that these event boundaries can influence our sense of time. But we don’t really know much about how our sense of time unfolds within a single event. One thing that I’m trying to understand now is basically as an event unfolds, are there distinct contractions and dilations of time within a single experience? 

[00:14:37.120] – APS’s Özge Gürcanlı Fischer-Baum 

Well, Buren, this was a wonderful conversation. Thank you for joining us today. That wraps up our conversation for today. 

[00:14:45.820] – Brynn Sherman 

Great. Yeah. Thank you so much for having me. This is a lot of fun. 

[00:14:51.560] – APS’s Özge Gürcanlı Fischer-Baum 

This is Özge Gürcanlı Fischer-Baum with APS, and I have been speaking to Brynn Sherman from the University of Pennsylvania. If you want to know more about this research, visit psychologicalscience.org. Would you like to reach us? Send us your thoughts and questions at [email protected].