I have this one particularly odd memory (let’s just call it a memory). There isn’t much to it in terms of detail. There’s this image of some shapes and lines floating around a white surface (think of abstract design you would come across in the early 90’s).
Besides these visual features, there is also this peculiar feeling of calmness. That’s pretty much it. But there are two last components that give me reasons for thinking this is a memory. It’s that it strikes me as both incredibly familiar and as presenting me with something from my distant past.
Not all of my memories are this strange. Most are pretty mundane. I have a spotty childhood memory of fumbling to put on my glove as I almost catch a foul ball at a Yankees game. I clearly remember my first Sainsbury’s delivery while quarantining after moving to the UK. I also have the distinct memory of seeing my partner in the crowd when we first met.
While these memories have varying levels of detail, and concern different subject matter, they all share two features with my odd memory. They all strike me as familiar and as presenting something from my past. Yet even here there are differences. Each of these memories locates events at different distances in the past. Some seem older than others.
Sometimes memories only locate events in time in the vaguest way (e.g., sometime in the past), while other memories might not locate events in time at all (see recent work from Alexandria Boyle). In these cases, I might have to exert some effort and engage in some reflective detective work to figure out when some event occurred.
However, in the cases I just described, I have what seems like an automatic sense of how far in the past the event occurred. I can even be surprised by this sense. My memory of the Sainsbury’s delivery strikes me as being incredibly old, and I find it surprising when I recall that this only happened four years ago. The memory seems like it’s from some far-gone era when I know it’s not.
It’s this sense of pastness that I want to focus on. Where does this come from? What explains how some memories strike us in this way?
Maybe this sense of pastness is the result of something that happens when memories are formed. Maybe memories are given a timestamp when they’re formed and this grounds our sense of pastness. While this would be a nice explanation, it has its shortcomings.
Not all memories have a sense of pastness, so we would have to explain why some memories are timestamped and others not. Some memories also seem to be more precisely located in time than others. Why would that be the case? Finally, why do some memories strike us as being much older (or more recent) than they actually are?
Some have attempted to account for the feeling of pastness in terms of some non-temporal features of recollection. One proposal of this sort is found in the recent work of Denis Perrin and colleagues. They have attempted to explain the sense of pastness in terms of “epistemic feelings” – feelings that are related to our knowing of facts and events.
For these approaches, the sense of pastness is grounded in the feelings of familiarity, fluency, and knowing that often accompany recollection. However, while these epistemic feelings might track a general sense of pastness, since anything I’m familiar with must be from my past, they seem unable to explain the sense of distance into the past that varies between memories.
Consider familiarity alone. My odd memory, that of meeting my partner, and of almost catching a foul ball are all very familiar. Yet, each of these memories seems to be located at very different times in my past. It’s this fact about distance into the past that familiarity alone can’t explain, and my hunch is that epistemic feelings in general can’t do this either.
Something else must be going on here. Maybe the sense of pastness is due to a temporal feature produced at the time of recollection rather than formation. However, there’s nothing specific about the visuals in my odd memory that places it so far in the past. The mere imagistic components of memory (e.g., what we see and hear with the mind’s eye) don’t explain this.
Perhaps instead there’s some form of inference at work that takes us beyond the mere imagery of the remembered event.
But if there is an inference, then how do we explain how our sense of pastness and temporal distance in memories is often at odds with the conclusions of explicit inferences we draw about the timing of remembered events? How is it that my memory of my Sainsbury’s delivery seems much older than I judge it to be after some reflection? And why doesn’t that feeling of pastness go away once the judgment is made?
Here’s where I think progress can be made by looking at how time is represented in another psychological faculty of ours: perception.
While scientists and philosophers have discussed temporal perception for centuries, it’s only recently that we’ve begun to appreciate how complex this timekeeping capacity is. There is a large literature on the complexities of temporal perception, but here, we’re going to focus on one aspect – temporal order perception (a lot of this material is covered in a piece I wrote with Kris Goffin).
We often perceive events as standing in temporal order relations – e.g., earlier than, later than, and simultaneity relations. A simple story for how we come to perceive events as standing in these relations appeals to the temporal sequence of stimulation or processing.
When I experience a strike of lightning as occurring before a crash of thunder, the light from the lighting often arrives at the retina and is initially processed before the mechanical waves of the thunder arrive at the eardrum. The lightning crosses the perceptual finish line first, and therefore, we perceive it as occurring first. However, the mere timing of sensory stimulation appears to be only one factor in temporal order perception.
Consider what happens when you see and feel something touch your foot. You often experience these as simultaneous, yet tactile signals take significantly longer to travel through your body to your brain than the visual signals do. Or consider what happens when you watch a movie where the audio and visual tracks are slightly out of sync. As long as the delay isn’t too long, and the movie isn’t dubbed, we tend to get used to the delay and after a short time we don’t even notice the asynchrony.
What’s happening here? Well, it seems like our perceptual systems are engaging in a complex causal inference.
In both cases, perceptual systems conclude that the various incoming sensory signals are likely due to a common cause, and therefore, we perceive these events as being simultaneous.
Similarly, when we have cues for perceiving one event as causing another, then these causal cues influence us to perceive the cause as occurring prior to the effect, even if their temporal order is reversed.
Importantly though, while causal cues can influence temporal perception, our perceptual systems still treat causal and temporal relationships as distinct. In fact, in a series of interesting studies, subjects can be made to have peculiar illusions in which effects seem to occur prior to their causes (see Stetson and colleagues ).
So, what’s going on here? Our perceptual systems use causal cues, along with a range of other cues, in the service of complex inferences regarding time.
The idea that I’m suggesting is that our sense of pastness in memory is due to a similar inferential process. We use a host of cues, including epistemic feelings as well as causal reasoning principles, and facts concerning the specific events being remembered, to represent via inference the pastness of our memories.
But now we’re back to our earlier worry. If this is an inference, then how do we explain how our experience of time can contradict our informed reasoning? This is a key feature of perceptual processing. Our perceptual inferences are “encapsulated”, or insulated, from much of our thinking. Consider the classic Muller-Lyer illusion
The two lines look to be of different lengths. The illusion persists even when you know that the two lines are the same length. Our perceptual processing is “modular” in this way. And so here I suggest that recollection involves a complex inferential process that locates events and memories in time.
Yet, while recollection is often thought to be a form of thinking, it shares a commonality with perception in that these inferential processes are encapsulated and hidden from introspection. These processes are hidden from our introspective capacities and only their output, the sense of pastness, is what shows itself to us.