Dan Schacter on the memory and the imagination.

When looking through research into imagination, memory and the brain, one name that keeps appearing is that of Dan Schacter.  We recently interviewed Donna Rose Addis on this podcast, and she and Dan often collaborate on research.  Dan has been a Professor of Psychology in the Psychology Department in Harvard University since 1991, focusing mainly on memory, and more recently on the relationship between memory and imagination, using techniques from cognitive psychology and neuroscience.  We chatted via Skype and I started by asking him if he might firstly introduce us to the hippocampus.  If he was introducing it to someone at a party, who knew nothing about it, where would he start?

“The hippocampus is a structure buried deep in the middle of the brain that came into prominence in neuroscience really starting in the 1950s with the famous case of the amnesiac patient HM, that was a young man who was in the hospital as a result of epilepsy and was having surgery to remove the hippocampus and a couple of other parts of the brain that are buried deep in the inner or medial part of the temporal lobe.

It wasn’t known at that time exactly what the hippocampus does, and in the case of HM, the removal of the hippocampus and several other structures produced a severe deficit in the ability to establish new memories of ongoing experiences that he could recall consciously.  So starting with that, that didn’t really specifically establish a link between the hippocampus specifically and memory, because there were other parts of the medial temporal lobe that were also removed, but the hippocampus was what people focused on.

That suggested an important link between the hippocampus and our ability to establish memories of ongoing experiences that we can recall, call to mind, consciously at a later time.  That really initiated 50 plus years of work on the hippocampus that has highlighted that it is critical in the ways that were initially thought by people like Brenda Milner, who was the neuropsychologist who tested initially patient HM and many others since.

There’s also been a separate line of research that’s emphasised the importance of the hippocampus in spatial processing, and spatial cognition.  Our ability to find our way around and remember where we’ve been.  This was work that was among others, spearheaded by John O’Keefe, who won the Nobel prize for this work a number of years ago, based in England.  O’Keefe and his colleague Lynn Nadel wrote a very influential book called The Hippocampus as a Cognitive Map.  It was published in the late 1970s.

It’s been these two themes of the hippocampus, its involvement in memory, and involvement in spatial processing, and more recently those have come together to suggest a role for the hippocampus in imagination as well as memory and spatial processing.  Those three things are very tightly intertwined.

As I understand it, the thinking up until 15, 20 years ago was this idea that different parts of the brain did different things, and the hippocampus was the place where those things happened, and now people think much more in terms of networks.  Scott Barry Kaufmann, refers to the Default Mode Network as the Imagination Network, but the impression I get from your work is that the imagination in its healthiest form depends on several different networks, and can be activated by different networks which sometimes only operate when the other ones aren’t there, or sometimes they’re working.  Could you talk a bit about that?  How the imagination networks?

Yes, so some of our work, the work that Donna Addis and I and others did about 10 years ago, pointed toward this network that is broadly known as the Default Network, or Default Mode Network, as a collection of regions that comes online when people remember past experiences, and also when they imagine future experiences, and other kinds of experiences.  For that reason it seemed to have a role both in memory and in imagination and some have referred to it as the Imagination Network.  It definitely does play a key role in various kinds of imagination.

If you ask people to imagine some event that has not taken place, compared to appropriate control conditions, if you’re doing a neural imaging experiment, you’re going to very reliably see the Default Mode Network come online.  Now at around the same time, there were other studies suggesting that this Default Mode Network may not really be all that functionally useful, in that when you give people difficult cognitive tasks to do, that require a lot of attention and cognitive effort, default network regions tend to go down.  Tend to show decreased activity.

They show increased activity when people are just left alone in a scanner, and really not doing much of anything, perhaps because they’re engaging in forms of imagination.  But these kinds of observations led some people to suggest that the Default Mode Network is what they called a ‘test negative network’.  In other words, it’s not clear exactly what it does, but we can be pretty confident that it doesn’t play a role in goal directed, functionally useful cognition.  Maybe the kinds of imaginations that it supports are things that are more akin to day dreams or idle fantasies, or just internally focused thoughts that really don’t do much for you.

Around 2010, in my lab, a post-doc, Nathan Spring, was doing some work on autobiographical planning.  In our initial studies of memory and imagination we were giving people word cues and asking them to either remember a related past experience, or imagine a future experience.  So I might give you a cue word such as ‘vacation’ and say, “Just imagine something that might happen related to this word in the next few months, and you might say, “Well, I imagine I’m going on a vacation to California and I’m going to enjoy the warm sunshine.”  Something of that nature.

In the Spring studies, people were given a more challenging goal directed imagination task, if you will, that we refer to as autobiographical planning.  Here, instead of just imagining any old thing that you might want to imagine, you would be instructed to try to come up with an autobiographical plan to achieve a specified goal. The goal might have something to do, let’s say in an educational context, with graduating from college.  You would have to come up with a plan that would integrate several steps that would allow you to achieve your goal, and also consider obstacles that might get in the way.

People were scanned when they did this more goal directed, complex, autobiographical planning task.  What was interesting here is that compared to a control condition, the Default Mode Network came online, and that was consistent with previous observations, but another network also came online and coupled its activity with the Default Mode Network, a network known as the Frontal Parietal Control Network.  That’s a network that typically comes online when you give people cognitively demanding tasks to do.

Here in these studies, and the first one was published in 2010, and there have been several since, what we were seeing is a coupling of the Default Mode Network and an Executive Control Network when people are engaged in this more demanding form of autobiographical planning that involves considerable imaginative activity.  That line of work has really provided strong evidence against the view that the Default Mode Network is a task negative network, that it doesn’t really support useful goal directed cognition.  It clearly does but it supports goal directed cognition of a particular kind.

When we’re focusing in on the internal landscape, but in this case, doing important cognitive work.  Interestingly, in more recent work that’s focused more on creativity, obviously closely related, has shown that in several different situations during creative cognition, you see this kind of default-executive coupling going on.

How do stress and anxiety and trauma impact, firstly the hippocampus, but also on those wider networks and our ability to be imaginative?

Generally negatively.  The hippocampus has long been thought to be affected negatively by stress, and of course anxiety has an important relationship to imagination, because it can lead us to a position where we’re focused on imagining negative future outcomes that may further cause us to become even more anxious.  So we may be anxious, we may start looking at the future in a negative way, constructing simulations of negative outcomes, and that may make us even more anxious.  So those two things are related in important ways.

One of the questions that seems to come up a lot in the things that I’ve been reading is the question of whether the imagination is a completely value neutral thing.  You know, Hitler was as imaginative as Picasso, for example, so the imagination can go one way or the other, and both are manifestations of imagination.  Or there’s another perspective that says actually the imagination in an untraumatised, unstressed, undamaged person, leans much more to imagining positive scenarios and being a positive imagination.  So if we see people who are imagining really dark visions of the future – I don’t know, Donald Trump always comes to mind – but if you have somebody like that, who is imagining dreadful things, and inflicting them on people, is that just the imagination, or is that the imagination being filtered through a damaged, anxious, stressed mind?

It really depends on the situation and what’s being imagined.  I would try to avoid really broad statements like the ‘imagination is this, or the imagination is that’.  So for example, we know that there’s generally an optimism bias when we think about and imagine the future.  People tend to be biased optimistically, and that may serve some useful functions.

However, that can also get us into trouble if there are real obstacles in the way that we need to take into account.  We don’t think about those because we’re overly optimistic.  So a realistic incorporation of potential negative future scenarios into imagination can be productive, when it’s appropriate.

What you’re talking about is situations maybe where the negativity is overdone or the negativity is over-emphasised.  That’s the kind of thing you do see in anxiety disorders, and anxious conditions.  So it’s not strictly speaking one thing or another, ‘negativity is bad, positivity is good’.  Neither are necessarily the case in absolute terms.  It very much depends on the particular situation I think that you’re imagining and what kinds of strategies and behaviours are necessary to achieve a future goal.

The research that Kyung Hee Kim did, The Creativity Crisis, she looked at divergent thinking as a proxy I suppose for measuring the imagination of a population.  So her thing was that when you look at it like that, that imagination and IQ rose together until 1990, and then they separated.  Do you think that divergent thinking, although imperfect, gives us some sense of the imagination level, and would you have a sense of the state of health of our collective imagination in 2018?

Divergent thinking is one important index of imagination.  We’ve studied that some because one of the focal points of research in my lab has been the extent to which the capacity for what we call episodic retrieval, or episodic memory, impacts imagination.  We know for example that amnesiac patients, who have a very poor ability to remember their own experiences, also have difficulties imagining the future and have been shown also to have decreased divergent, creative thinking.

We’ve done some research in my lab that has shown that divergent thinking, or measures of divergent thinking, can be increased by a procedure that we’ve come to refer to in my lab as an ‘episodic specificity induction’.  This is basically brief training in recollecting details of a recent experience.  After you’ve had this brief training in just remembering in detail, unpacking in a lot of detail what you remember from a video that you just saw, a lot of my work in my lab suggests that that gets you into kind of what we call an ‘episode retrieval mode’, where you’re really focused on the details of people, places, objects, that constitute events in one’s life.

What we’re interested in in these experiments, is after you’ve had this episodic specificity training, how does that carry over to a later task that you might perform?  The idea is that if we see an impact of that episodic specificity training, or induction, on a subsequent task, that tells us that that task is some circumstance does and can draw on episodic retrieval.  To bring it back to divergent thinking, we studied this with a classic divergent thinking task, known as the alternate uses task.

This is where you’ve got to come up with novel uses of familiar objects.  I might give you as a cue the word ‘brick’ and your task in the alternate uses task is to generate unusual, but appropriate, uses of a brick.  So the number of unusual, novel, but appropriate uses that you come up with is roughly speaking one index of divergent thinking very commonly used in the field.

What we found in these studies is that if you’ve just had the episodic specificity induction, then you come up with a few more novel uses of items such as a brick, or a paperclip, or whatever, than if you’ve had a control induction where you’re thinking in more general terms about the properties of a video you saw as opposed to unpacking the specific episodic details.

We’ve got several studies on this from my lab.  They were done with a graduate student in my lab, Kevin Madore, and also with Donna Rose Addis was involved in some of this work.  That suggests to us that the capacity for constructing a detailed episodic simulation, a mental event that’s populated with a lot of people, places and objects, when you’re in a set to do that, that actually seems to enhance divergent thinking.  We’ve done some very recent work in the fMRI scanner that shows for example that after you’ve just had an episodic specificity induction, and now you do the alternate uses task as divergent creative thinking, a couple of interesting things happen compared to a control condition.

Number one is that you see increased activity in the hippocampus, which we know comes on line during episodic memory retrieval, imagining future events and so on.  That shows a boost during the alternate uses task.  And we were talking earlier about interactions between the Default Network and Executive Control.  You see some evidence of increased interaction between components of those networks during divergent creative thinking on alternate uses tasks when it follows on from an episodic specificity induction.

So that induction, to get you in a mode of detailed episodic retrieval, seems to impact the way, at a neural level, that you carry out divergent thinking, with a little more activity in the hippocampus, a couple of other regions, and an increase in functional connectivity between Default and Executive networks.  Those are some things we’ve learned about divergent thinking.  I think it’s important.  Whether there’s a societal crisis in divergent thinking, I can’t really speak to that.

One of the things about the Default Mode Network is that it needs space, it needs time, that kind of space where you’re not doing something, that day dreaming kind of space.  But it seems like so often in the world around us, the experience is that whenever we have time when that might happen, we get our phones out.  I interviewed Larry Rosen and Adam Gazzaley recently with all their distracted mind work.  Adam was talking about he says we have a cognition crisis because of these different technologies.  If we have very little space for the Default Mode Network, if we fill all that space up with Facebook and so on, what are the impacts of that?  Is that a problem?

Yeah, I think it is potentially a problem, just because there’s less time spent in this more inner focused, imaginative mode.  There’s also another side consequence of that, which is again, for example, in a learning context.  If students are distracted into their phones, it’s more of a potential learning deficit that can arise in addition to what you’re talking about.

What you’re talking about is private time that might have been once devoted to focusing inward and imagining the future and thinking about the past, now you’re just hooked into your phone and more into an external mode.  That potentially could affect the richness of one’s imaginative life for sure.  We’ve also been interested in other ways in which distraction can impact learning.

One of the questions I’ve asked everybody that I’ve interviewed is, “If it had been you who had been elected as the President, and you had run on a platform of ‘Make America Imaginative Again’, what might you have done in your first 100 days in office?”

I’m not sure what I would do. I might try to create incentives for things, like getting people to read more, focus less on the internet and their phones.  Time away from social media.  Reading is a time when you can really go inward, when some of these constructive aspects of imagination are at work.

If I could find some way to incentivise people to do more of that, that might be something I would do, related to that ‘Make America Imaginative Again’ programme.  Encouraging also finding incentives for creative and artistic activity, things of that nature.  I would have probably supported the national endowment for the arts, things of that nature.