221 | Adam Bulley on How Mental Time Travel Makes Us Human

0:00:00.4 Sean Carroll: Hello everyone, welcome to the Mindscape podcast. I'm your host, Sean Carroll. Welcome to the first podcast of 2023. And to start out, I want to do a little thought experiment here. Imagine that I told you that if you go to patreon.com/seanmcarroll, you could pledge a small amount of money to support Mindscape and in return, you would get the ability to listen to ad free versions of the podcast as well as to ask questions at the monthly AMA Ask Me Anything episodes. Now that's true. I've said that to you before, you probably already know it. But I'm not actually pitching the Patreon right now. Instead, I want to think about what is going on in your mind when I say those words. We're conjuring up an image, right? You can imagine doing that, going to Patreon, kicking in a dollar per episode or what have you, or you could imagine not doing it. You can contemplate different hypothetical futures. This is an ability that human beings have to think counterfactually, to not just say, this is happening, therefore that. In other words, not to just be automatic or unconscious about our reactions to what is going on, but to think through different hypothetical possibilities, even hypothetical possibilities that involve things in the future, not just right now, right here, right this moment, but at some undetermined time in the future.

0:01:31.3 SC: There's different hypothetical possibilities and we can sit down, imagine them, and choose which one we want. This is a remarkable ability and it's not at all an obvious one to evolve over biological time. In fact, arguably, it is only human beings who have this ability. Of all the different animal species or plant species, for that matter, here on earth, the ability to what we call mentally time travel, to put ourselves in hypothetical future situations, as well as of course to remember real past situations or even to remember fake past situations, for that matter. This ability to imagine ourselves at different moments of time in different situations seems to be uniquely human. Now of course it's a spectrum, right? I mean it's not a hard and fast thing. There are plenty of animals that have plenty of impressive abilities to do other things, but we certainly do it way more in much more detail and it's kind of implicitly crucial to who we are as a social species, right? It allows us, the reason why I chose the Patreon example, not just to pitch the Patreon, but also because it's not just imagination in the sense of, you know, artistic imagination or even scientific imagination.

0:02:47.6 SC: It enables us to make agreements with other human beings. It opens a new social dimension to our lives. We can make contracts, explicit contracts or implicit ones. We can agree if you do this, then at some other future moment, I'm going to do that. That ability is absolutely crucial to make us human beings, certainly to make this kind of society and cooperative culture that we do have and we do enjoy. So today's guest, Adam Bulley, is a psychologist who recently collaborated with two other psychologists, Thomas Suddendorf and Jonathan Redshaw, on a book called The Invention of Tomorrow and Natural History of Foresight, where they talk about exactly this. All three of them have done research in exactly this area, how the ability to mentally time travel evolved over evolution, and what role it's played in human history, how important it is. There's certainly a strong relationship between this and another episode we had with Malcolm MacIver sometime back, where we talked about a much earlier moment in evolutionary history where the first fish climbed onto land, and that might have been the very beginning of this ability to mentally time travel, but human beings have developed it much later in evolutionary history to an extraordinary degree, and that really is very, very crucial.

0:04:13.5 SC: That's the point that they're trying to make in this book, that it's not just going along for the ride. This ability to put ourselves in the future has made us human in a very, very real way. So this is something, you know, combining humanity with time travel, things like that, this is exactly the kind of thing I love to talk about in the podcast, a good way to start the year 2023. So let's go.

[music]

0:04:56.0 SC: Adam Bulley, welcome to the Mindscape podcast.

0:04:58.3 Adam Bulley: Fantastic to be here. Thanks for having me.

0:05:00.9 SC: So I wanted to start with a provocative factoid that you mentioned in your book or you've shared around, which is that chimpanzees will say hello to each other, they'll greet each other when they meet, but they do not say goodbye. Is that accurate?

0:05:14.2 AB: Yes, that is accurate. They tend to greet each other very affectionately. They'll sometimes even give each other a little high five, a hug and a kiss, but Jane Goodall frequently remarks that they do not bid each other farewell. And one of the sort of reasons for that is that they do not really have the same kind of capacity for foresight. They can't imagine themselves meeting again in the future. And so there is not really any real point in bidding one another goodbye if, of course, there is no anticipated reunion in the future.

0:05:47.7 SC: I mean, I guess we'll get into this in great detail. But how do you know that? Is this built up in a whole bunch of different ways or is there some direct way in which we can say that chimpanzees cannot anticipate meeting each other?

0:06:00.9 AB: Well, it's obviously a very deep question. It's a complex question about exactly how you ascertain the abilities of other animals in any cognitive domain. You obviously use various tools. There's comparative psychology, there's ecology, looking at the behavior in their natural environments. There's a lot of observational work done. But at the end of the day, one of the key things required is systematic scientific studies of animal competencies. Obviously, you can't read their minds and they can't talk to us to tell us what they're thinking about. No chimpanzee is going to say, oh no, I'm not saying goodbye because I'm not going to see him again. And so what you need to do is systematic controlled scientific studies. And many of those have been done over the past few decades because the question of whether other animals can think ahead in time is obviously very important when... If you're trying to make some kind of claim like we do in our book, that thinking ahead is one of the key things that makes us human.

0:06:58.4 SC: Good. So I'll give you the opportunity to define exactly what it is we're talking about. I know there's all these subtleties between like planning and anticipating and thinking ahead, mental time travel. So what are we focusing on here?

0:07:14.5 AB: Okay. So, yes, very good point. It's a complex, multifaceted topic. We're talking specifically, at least in this sense, about the ability to specifically imagine, think about the distant future and to organize your behavior accordingly. That's the kind of foresight that we have in mind here. Obviously, there's other kinds of future-oriented behavior, and we go through some of them in the book, that include things like planning, as you just mentioned. So the ability, for instance, for an animal to face a maze and navigate through that maze requires some kind of planning capacity because there's a requirement to find the way through the maze to a goal, to a target.

0:07:57.5 AB: And other animals have goals. They pursue things. They navigate through space and obviously through time as they're doing it. And what you find in humans, though, is a capacity to imagine ourselves in a remote future time, vividly pre-experiencing what might happen to us. And you also see this ability to anticipate mutually exclusive possible future events simultaneously, as opposed to, for example, cycling between them, as a rat might do while going through a maze. And we can dive into any of these different aspects of future-oriented cognition. And at the end of the day, we don't really think of foresight as just one thing. It's not like a module in the brain. Instead, it's a suite of interacting capacities that kind of work together to enable complex future-directed actions.

0:08:47.0 SC: Well, I like the word you use, pre-experiencing. I hadn't heard that one before. So is that sort of at the heart of the distinction we're drawing here? Like, famously, squirrels will gather nuts to tide them over the winter, okay? But that could arguably be said to just be like an automatic thing, and they'll gather things and it will work, but they're not sitting there imagining the future winter and visualizing it in any sense. Is that okay?

0:09:14.3 AB: Yeah, that's a fantastic example. I mean, so, as you say, the squirrels that will hoard nuts in anticipation of future winter food shortages, they will do that even if they've never experienced a winter before. They obviously have no way to tell each other that there's a forthcoming frost. And so we know just simply from that, that their behavior is actually driven by instincts. It's a response to particular kinds of cues in the environment that will trigger that behavior. And they're not actually imagining themselves in a possible distant future situation where they don't have the food they need to sustain themselves.

0:09:51.5 SC: And one argument for that is just that squirrels that are less than one year old do this.

0:09:58.3 AB: They'll do it even if they've never experienced any winter. And you can... People have done experiments on this kind of behavior, and you can manipulate it fairly easily. In the same way that if you imagine various other kind of fixed action patterns that animals will do, famously, Niko Tinbergen, if you do know the work with, for example, the graylag goose.

0:10:22.5 SC: I do not.

0:10:23.9 AB: The goose has a fixed action pattern where it will try and roll an egg back towards its nest, if the egg falls out, and that seems very future directed, it seems very intentional, because you might imagine, oh, no, the goose is thinking that the baby is potentially going to be damaged, and it's a very dangerous situation. But if you simply replace the egg with something else, or even remove the egg entirely, once the behavior pattern is actually initiated, the goose will enact the entire sequence through to completion, regardless of whether the actual object in question, which is the egg, is present or not.

0:11:00.1 SC: Right. Okay.

0:11:00.2 AB: And so that's just an example of a kind of fixed action pattern or an instinctual behavior that you can see in animals that it's revealed to be fairly blind to the future, simply by the fact that it can be manipulated by changing the cues that initiate the behavior, if that makes sense.

0:11:15.8 SC: Good. And in contrast to that, we're talking about the human capacity to imagine that we are doing something different or will be doing something different. So I guess one question is, what is the relationship between our imagination in the sense of just visualizing something different in our heads, and literally indexing that with, "oh, this is the future, that was the past," something like that?

0:11:43.0 AB: Okay, yeah, that's another really good question, because it's actually one of the kind of key topics in the scientific study of memory and imagination. And the real question, kind of just to almost reformulate what you said, and you tell me if this is on the right lines, is how does the brain tag an imagined episode as being fictional, or a memory, or imagined future, or any other kind of context? That comes down in some sense to our ability to meta represent, or to have some reflective capacity to represent our own lower, lower order mental capacities. So for example, in theory of mind, the key test there is often the ability to represent that another person has a false belief.

0:12:29.2 SC: Oh, yeah.

0:12:29.5 AB: And in much the same way, I might need to represent that I could be wrong about the future, I need to represent that this is just a memory, or that this is fictional. For example, if I imagine the same event, for example, myself eating a slice of pie, I can also reflect on the fact that that imagined episode in my mind's eye, it either was a memory that happened to me, I'm remembering the last time I ate a slice of pie, or it's completely fictional, because I don't eat pie, or it's in the future, because I'm imagining eating pie for dessert tomorrow. And so that metacognitive capacity is actually a very essential part of the puzzle.

0:13:08.5 SC: And I guess so, you reformulated a more interesting question that I asked, but let me ask the question that I actually had in mind, which is, so it seems like there are two parts. One is like, putting on the play in our heads, right? Like literally constructing the scenario. And another part is thinking of it or associating it with past, present, future, fantasy, whatever. Like, do these necessarily go hand in hand? Are they separate parts of the brain doing different things?

0:13:39.4 SC: Yeah, good. They are probably, like I said earlier, foresight is we don't tend to think of it as just one encapsulated module. And in fact, it is possible to lose various aspects of the ability to think about the future while keeping the other parts intact. And so maybe it'll be helpful if I kind of break down some of what we know about how you put this mental play on as you sort of refer to it. And we can go from there. So one of the key distinctions, first of all, is between what scientists would call semantic memory and episodic memory. And so semantic memory, for those who don't know, is our memory for pieces of information, bits of knowledge, facts. So for example, that the capital of France is Paris. On the other hand, you have episodic memory, which is our memory for the episodes of our life. And these are the kinds of episodes, for example, remembering my 10th birthday party. And I can vividly recollect that experience and I can walk through it in my mind's eye and I can remember all of the details. Now, those two things can actually be dissociated. So they are intertwined, but you can lose access to one while keeping the other intact.

0:14:51.1 AB: And so for example, in particular kinds of amnesia, this famous case of amnesic patients who have particular damage to the hippocampus, and that can impair the ability to actually vividly imagine future events and remember specific past episodes while leaving semantic information intact. So for example, having the knowledge that climate change might affect low-lying coastal regions in the future without being able to imagine oneself actually being subject to that kind of damage. And so that's just a first level distinction. But there's many other aspects to this mental time machine that we can get into.

0:15:31.8 SC: But something just happened there that maybe you did say and I missed it, but we're talking about episodic memory, but then you related it to the ability to sort of imagine that we will be subject to climate change. So is that the same part of the brain that is putting on the play and we use both in memory and in foresight?

0:15:53.4 AB: One of the most interesting discoveries of the last 10 years in this area has been the fact that it is indeed the case that the brain uses much of the same neural circuitry to remember past episodes and to imagine future events that might happen. And there's been a lot of neuro imaging work on this particular question, which has found that if you ask people in an fMRI scanner, for example, to imagine something happening to them in the future, and you ask them to remember things happening to them in the past, you see a very, very similar analogous set of neural structures activated in both of those conditions. And that work has been led by Donna Rose Addis and Daniel Schacter, collaborators, and they've done many interesting variations on those kinds of studies. And the picture is emerging that, yes, there is a kind of mental time machine which uses the same cognitive resources to go both backwards and forwards in time.

0:16:50.0 SC: And how closely connected is it to visualization? Could we have this kind of imagination without painting a picture in our heads?

0:16:58.1 AB: Yeah, that's a very interesting question. There's been studies done where you ask people how they're imagining the future because people differ in this. You might say someone might just verbally think about what might happen to them in the future, while someone else might vividly imagine it. As it turns out, the vast majority of people, when they are asked to imagine something happening to them in the future, and you modify the wording and stuff, they're doing it visually. It's not a necessary prerequisite. And so, in fact, you wouldn't necessarily equate episodic, which can be defined more specifically with regards to the where and when details of an episode on the one hand, and on the other hand, the visualization capacity to actually bring an image into the mind's eye. For example, if I ask you to picture an apple, you can do that. You can bring it up into your mind's eye and sort of rotate it. You can imagine how it smells. And in that case, the brain is sort of reactivating sensory cortices in order to bring about that mental image.

0:18:00.0 SC: Except some people can't, right? Some people lack that ability, but presumably they can still remember and imagine the future.

0:18:07.0 AB: Yes, except there's been some interesting studies done recently looking at aphantasia, which is this inability to generate mental images in the mind's eye, showing that these people have a harder time imagining the future and planning for events that might happen.

0:18:26.0 SC: And when we're talking about this kind of foresight that you're studying, I guess, where does the idea that it is hypothetical or counterfactual or there's more than one possible thing come in? Could we imagine the ability to imagine ourselves in the future, but still think that it was a definite thing? Or did it come hand in hand with this idea of different possible worlds, as the philosophers would put it?

0:18:52.5 AB: Right, possible worlds. Yes, so my collaborators and co-authors on the book, John Redshaw and Thomas Suddendorf, have done some really elegant studies on the early cognitive development of the ability to represent the future as a branching world of possibilities, rather than just a single definite single track vision. Maybe I'll just describe one of the studies which can elucidate this a little bit. So if you imagine dropping a grape through a tube, and the tube is just a straight pipe, and at the bottom there's an exit, and you ask a child to catch the grape, what they'll do is they'll put a hand out and they'll catch it as it comes out the other side. Now imagine we make the future a little harder to predict, and you turn the single tube into a forked tube, kind of like an upside down shaped Y. And so now there's a single entrance point at the top, but there's two exits at the bottom, it splits into a fork. When you give that task and you drop a grape through the top or a prize of some kind or a ball and you ask a child to catch it, when they're two and three years old, what they tend to do is only put a single hand out and they'll move the hand back and forth between the two exits as if they're guessing where it's going to come out.

0:20:07.9 AB: They're making a prediction, they can anticipate that the ball is going to drop out of one or the other exit, except once they get to about age four from the first trial of a task like that, children will actually put both hands out and cover both of the exits from the tube. And what that seems to imply is that the child not only makes a prediction about the future, but also represents that they could be wrong about that prediction and that therefore they need to make a contingency plan. I don't think they're going through it in all that much detail. That's the experimenter interpreting it. But what it seems to represent is that they have the basic cognitive capacity to represent the future as having mutually exclusive possibilities rather than just a single track outcome. And so that ability kind of comes online very early. And what we argue in the book is that it's actually one of the most fundamental abilities that kind of comprises foresight. Like I said, it's not one thing. But this ability to represent the future as a world of possibilities is really fundamental to what makes foresight so powerful and effective. If you only had the ability to represent it as a single outcome, it wouldn't be anywhere near as useful.

0:21:24.1 SC: And one important note about the relationship to memory is that our memories are terrible, right? Human beings are not very good at remembering things. Does that mean that we're not very good at imagining the future, or is that actually a feature that we make use of in imagining the future?

0:21:42.4 AB: Yeah, the errors of memory are probably, one way of thinking about them is that they're actually a kind of... They're either a by product or they are a design feature of the more fundamental and powerful ability to imagine the future. I'm going to unpack that slightly.

0:21:57.4 SC: Please.

0:21:58.1 AB: If you have a memory system that is flexible, where the various bits and pieces from memories, so for example, that the man had a hat on, and that the guy had a mustache, and that there was a coffee cup present, and so on. All those bits and pieces from an episode can be combined and recombined in flexible ways to enable us to remember a vast multitude of previous events that have happened to us by using constituent elements. So that's an effective way for the brain to reconstruct what has happened to it before, without having to store all the specific details from every single event that has happened. Now, what that's really useful for though, is when you want to imagine the future, which of course, by definition, hasn't happened yet. And what the brain does, seem to do is combine and recombine the elements that we've accrued in memory in order to imagine possible future events as well. So for example, now I can combine and recombine that, okay, there was a mustache and there's a coffee cup, but now I can invent a new fictitious scenario using those basic elements, and then anticipate how certain changes to those elements might change what eventually happens in the end.

0:23:18.9 SC: So one way I've heard it described, and I wasn't ever sure whether this was an exaggeration or not, so I'm glad that I have an expert here to ask, but the memories that we have are really not like a recording, not like a videotape recording. It's more like what we store is a screenplay, which of course requires far fewer gigabytes than a whole video file, right? And then our brain has the production capacity to take in the screenplay and put on a little play for us. And that's the same kind of parts of the brain that might be used to imagine something that hasn't happened. Is that an even close to accurate picture?

0:23:57.4 AB: Yeah, that's very accurate. That's great. I mean, and another way, it's just more metaphors, but another way of thinking about it is like that the brain stores a kind of recipe for the event, and then it uses that recipe to reconstruct the neural activity that did occur when the event was originally perceived and encoded. And so that's a very effective way. It's kind of like an index, which is used to retrieve the various details and combine them in the particular configuration or the particular pattern that they were in when they were first encoded. And so understanding memory as a kind of pattern reconstruction system, in computational terms, actually makes sense of a lot of kind of how it works and then how it can be used to imagine the future. Although I will make a side note there that the search for the memory trace, the actual where the index is stored in the brain has been a very, very difficult search and it's still ongoing. I don't think it's necessarily settled by any stretch of the imagination exactly how the brain stores these traces and then uses that trace to reconstruct the whole pattern. A lot is understood, but it's still not, you know, the details are not worked out.

0:25:11.0 SC: Well, this sounds interesting and important. I want to hear more about this. So what do you mean by the trace that we don't know where it's stored? Is that like the data about what happened at that particular event that we're trying to remember?

0:25:21.3 AB: Yeah, if you think about it theoretically speaking, what's probably going on is that the brain is using... When you're perceiving something, when you're encoding some event, there's a whole bunch of details and you need to not only encode what they were, but the relationship between all the details. There's a certain pattern there about how these details are combined. And what the brain is probably doing is storing a trace or an instructional kind of, I don't want to say a guide, but for metaphorical sense that enables that pattern to be reconstructed later at will, when you volitionally attempt to remember what happened to you in the past. And this might be a good juncture actually to talk a little bit about the kind of brain mechanisms of memory and imagination. And obviously I think we mentioned briefly the hippocampus earlier, but the hippocampus is really very heavily implicated in all of this. So if these traces are to be found, that's a good place to look for them. And the hippocampus is a brain structure in the medial temporal lobes. It's also the structure that if it's destroyed by some kind of injury or insult to the brain, that's the part that will be removed that will leave you unable to imagine yourself in the future in specific circumstances and the inability to remember specific past events as well.

0:26:48.1 SC: And there have been patients to whom this has happened, so we have the data.

0:26:52.1 AB: We have the data. There's a very famous patient who goes by KC in the literature who this patient had damage to the hippocampus and the medial temporal lobes. And then despite retaining semantic knowledge and particular knowledge about particular facts, he could no longer remember any event from his personal life and he could not imagine what would happen to him in the future. So he's in no position to plan a party or imagine what he might do in the future.

0:27:24.0 SC: So we have to...

0:27:25.4 AB: It's a very interesting patient because if you start to investigate precisely what he lost, the picture becomes quite complicated. So I'll just give you one little tidbit. This patient KC, he would still discount rewards in the future according to the same hyperbolic pattern that most people do. And maybe this is something we could talk about, but I just put that on the table because it just reveals that the kinds of imagination of the future and the other kinds of decision making about the future, they're not necessarily in a one-to-one mapping. It's not like if you lose one, the whole thing falls apart.

0:28:03.7 SC: Well, part of what I wanted to ask next was it sounds like you mentioned the hippocampus and its memory functions, but it sounds like there's different parts of the brain that are getting into the game, right? With like putting on the play or anything like that. Have we located which specific parts of the brain are involved in both episodic memory and or in foresight?

0:28:26.1 AB: One important thing to note on this is that it's become increasingly, I think, common to think about the brain instead of in localizationist terms where you've got a particular part that does X, a particular part that does Y. Instead, you've got these kind of wide scale networks which are many brain regions kind of communicating and interrelating together. And it's the pattern, the overall pattern of that network activity that can perform various functions. So that's just to say, just as a slight caveat, but there have been various lines of evidence to try to isolate other parts of the brain and what they're actually doing when it comes to putting on this mental play. So for example, we already mentioned a little bit about metacognition and reflection and executive control. People might be aware that that kind of higher order cognition is often associated with the prefrontal cortex, because it's reliant on executive functions. For instance, reflecting on other parts of cognition or keeping balancing the demands of multiple competing tasks. Obviously, that's something you need to do. If you're going to imagine the future, you need to somehow balance that daydreaming with the goings on around you in the world.

0:29:43.4 AB: There is a tension there as well, which the brain has to navigate. It has to, "when do I think about what's going to happen next and when do I focus on the here and now?" And you have to have some kind of mechanism that performs that switching.

0:29:56.9 SC: I did do a podcast a while back with Robin Carhart-Harris about psychedelics and their effect on the brain. And one of the things I learned there and elsewhere is that in terms of vision, at least, there's a whole bunch of things that are always going on in the brain and our brain naturally sort of tamps them down. Like the vacuum state of the brain is actually abuzz with activity. And what the psychedelics do is not increase activity in those visualizations, but decrease the effectiveness of the damping mechanism. So I'm wondering if... Is there something like that for imagining the future? Are we like secretly, constantly imagining a whole bunch of things that our conscious brain just doesn't have access to?

0:30:40.4 AB: Well, if you try and meditate for even just a few minutes, you'll discover pretty quickly that the brain does tend to drift into thoughts about past and future. And this has been... There's been studies done on experience sampling where you tell... You just probe people throughout the day. You say, "what are you thinking about?" And you have them give you an answer. And unsurprisingly, vast majority, well, not, I don't know the specific number off the top of my head, but a lot of the time people are thinking about the future. Is my gym membership going to expire soon? What am I going to have for dinner tomorrow? How long until XYZ, you know, they're worrying about things. They're thinking about what might happen. And this has led to many people arguing that the... It's almost a, it's, it's part of what comprises the brain's default mode. And as you just mentioned, Robin Carhart-Harris, he's done some fascinating work on the effect of psychedelics on the brain's kind of default functions. And just to clarify for people who might not be aware, if you put people in a brain scanner and you don't tell them to do anything, the question is, well, what, what does the brain default to?

0:31:48.7 AB: What are people doing when they have no task? And it is exactly what I just mentioned. It's remembering the past, it's imagining the future, it's thinking about the minds of other people. It's daydreaming, it's exactly what you would, you would discover if you tried to meditate and you just see, just observe for yourself what your brain is doing, one of its own accord. And so I think to answer your question, yes, people very, very commonly think about the future. And of course, as I mentioned earlier, balancing the demands of the present with the benefits of planning is tricky. And obviously, if you spend way too much time in imaginary worlds, you'll get nothing done in the present. And so there is, there is a kind of balancing act that has to happen in order to make foresight powerful.

0:32:32.3 SC: Good. Okay. I mean, plenty to still say about foresight. There's one thing about memory that I didn't want to quite let go, which is the idea that if it's true, that rather than storing a video recording of what happens, we store some much tinier script than we can replay. Does that help us understand why faulty memories can be equally vivid to real memories? You know, all we have to do is change a little bit of the script, and then we're running it as just as vividly in high definition as if we'd actually been there.

0:33:03.2 AB: Yeah, I think that's probably fair to say that the specific reasons that memories can go awry, they are varied. I mean, there's there's many, many reasons why... And there's many ways that it can go wrong as well. So you could, you could miss that misattribute something that you've heard to someone else. So let's say, I incorrectly on this podcast, I say, "Oh, I've heard such and such a piece of evidence from such and such a person," and I've got it wrong. Well, I'm misattributing the source. That's an example of a memory error. And that reveals that it's not necessarily just that the reconstruction can go awry, which it can, as you just mentioned, but it's also the these other aspects of memory, these complicated tagging mechanisms and so on. And often the kind of metacognitive ability to say, Okay, where did I hear that? Identifying the source that kind of comes along for the ride, but it can also be... It can be it can be disrupted, it can, you can get it wrong.

0:34:00.6 SC: For completely other reasons, just earlier today, I was reading about the fact that there were two Beatles songs that John and Paul completely disagreed about who wrote them. John claimed that he wrote, he wrote most of the lyrics for Eleanor Rigby. And Paul said, No, he wrote almost nothing. And Paul said he wrote a lot of the tune for In My Life. And John said, No, he didn't. And every other song they agreed on. But still, I mean, I know I should know better than this. But I couldn't help but thinking like, how can you not know that? These are famous songs.

0:34:28.6 AB: Yeah, how can you not know? Yeah, it's really funny. Once you start thinking about memory systems as a kind of computational system for achieving particular kinds of goals, the errors, they do crop up everywhere. And maybe you get a little bit more forgiving of them. Because you start to say, Okay, this is a kind of, this is a system that is like a robot or a computer or something trying to achieve some particular outcome. And it's going wrong in a kind of well described way. And you can sort of see where the errors are occurring and try to correct for them.

0:35:02.0 SC: Well, and one of the arguments that I get from your book, and again, you'll correct me if I'm misinterpreting it, but the fundamental purpose of the brain in some very simplistic sense is to predict the future is to help us anticipate what's going to happen next. The brain is a prediction machine. And maybe it's okay for memory to be a little bit faulty if the overall goal is to predict the future, not to reminisce about the past.

0:35:27.6 AB: That's exactly it. And if you think about even the sort of simple types of memory that take place in the course of learning, so for example, in classical conditioning, or in operant conditioning, where an animal learns to associate particular behavior with a particular kind of outcome, that is memory, in the sense that the animal is kind of building up a repertoire of behaviors and outcomes that it can sort of store those represented their relationships. But it's really about the future. Because, even even Pavlov's dog, when he salivates, it's in anticipation of the coming food. It's only a memory in a kind of in a certain sense, really what it's about is preparing the animal to face an upcoming opportunity in this case, but also, it's all about preparing for threats as well. And so you have these finely tuned systems in the brain, that many animals have, and that humans have taken to a really just an outrageously developed degree, that enable preparation for the future from from very, very fine timescales, like the piece of food that's about to reach my mouth, all the way to planning for the afterlife, or the the end of the universe as we know it. And so it's a it's an interesting system in the sense that it does seems it cuts across all those different timescales.

0:36:47.8 SC: So I'm curious about how it all developed over the course of evolution here. So here's a simple question. Which came first memory or foresight in evolutionary history?

0:37:03.0 AB: That's an interesting one, I think I'd probably say that they come on board together, at least in some sense, like if you if you imagine some of the the kinds of memory that that I just laid out, like classical conditioning, very basic kind of learning. It also has a preparatory function. And so these two things, they do sort of come hand in hand, even really simple organisms, like we lay out in the book, an example from E. Coli, is incredibly simple, but it seems to switch on digestive genes for the digestion of maltose hours before it reaches the maltose rich tracts in the digestive system of the host animal that it's infecting. And what it's probably doing there is it's it...

0:37:52.6 AB: There's no need to think that, okay, the E. Coli is envisaging the how good is that maltose going to be when I get there? It's not doing that, instead, what's happened is over evolutionary time, the variants that did that were more likely to survive and and pass on those beneficial genes to the next generation. And so over time, you can get very functioning forms of preparation evolving without anyone really having to develop a considered plan about how it all is going to work. Yeah.

0:38:21.2 SC: Does the E. Coli have a memory correspondingly?

0:38:26.7 AB: Well, I mean, the E. Coli has it's got a memory in in the sense that it's storing... That it's evolved to represent the relation between I mean, it's not representing anything consciously or in any kind of mind, but there is a genetically hardwired relation that is remembered "between the time course of events that it needs to correspondingly prepare for," and for the people at home can't see the scare quotes I'm putting around all of the everything in that sentence, because obviously, the E. Coli is not thinking about anything.

0:39:03.0 SC: Well, but it Yeah, so you're saying that the E. Coli is sort of an even more primitive version of the squirrel just doing something automatically, right? And it's not envisioning the future ahead of time. But if it's true that over the course of evolutionary history, there are all sorts of these capacities that develop in animals or even pre animals to do something to prepare effectively for the future. Did our imaginative capacities latch on to them and take advantage of them once they did spring up in this more human, visualizable way?

0:39:39.0 AB: Yeah, I think it's an extremely gradual process. The ability to imagine the future as humans do it was actually something that gave Darwin pause because he said at one point in his autobiography that he was most inclined to believe in God when contemplating how such a capacity to traverse time in your own mind could have possibly evolved through blind chance or necessity. And so when you just look on the face of it at the final product, which is this quite remarkable ability to span the ages in your mind's eye, that seems very difficult to try and understand in terms of how it could have evolved. But I think as you just said, the right way to think about it is to kind of chart the trajectory and think about all the small incremental steps that would have led to that ability. And in the book, we kind of take as our starting point the split that led to modern chimpanzees and modern humans. But of course, you can go much, much earlier. I know you had Malcolm MacIver on the podcast, and he's done some... Only relatively recently found his work, but arguing that the transition from water to land was a pivotal moment.

0:40:55.1 AB: So maybe that's one of the steps that you can zoom in on, and then maybe you can go back to the beginning of the book and think about the processes that led to an increasing imagination of the future in those particular species and then kind of go from there.

0:41:10.3 SC: So you're saying on the one hand that this imaginative capacity that we have grew up gradually over evolutionary time.

0:41:19.4 AB: Yeah.

0:41:20.3 SC: But you also do want to say, or at least tell me if this is right or not, or say it as clearly as possible. We can do it better than anyone else can in the set of animals and vegetables here on Earth.

0:41:33.2 AB: Correct.

0:41:35.0 SC: So is there...

0:41:35.0 AB: Both things can be true. You can have a gradual increase in some competency that ends up developing a particular capacity way better than other animals can do it. So we're particularly good at bipedal locomotion as well. And some people are very interested in that. And so it's really, it shouldn't be particularly controversial. You just look at the human ability to travel mentally in time, and it is just obviously way, way better. At least it's more articulated. You can look further into the future. You can anticipate possibilities, contingencies. You can go into remote, distant times. You can anticipate drives that you don't currently possess. All of those things, they are much more articulated than other animals, but that is not to say that other animals cannot, they have no ability to engage with the future, they have no future directed capacities. And that is a gradually acquired trait.

0:42:34.5 AB: One aside here is important for people to bear in mind is that if all the other hominins that humans once shared the planet with were still alive, then the gradually acquired trait would look more gradually acquired. But all of the intermediates have gone extinct. And so now there looks to be a sort of, as Thomas, my collaborator, has argued in his previous book, The Gap, there does now appear to be a chasm between the mental abilities of humans and all of the rest of the animals that are currently alive. But part of that is an artifact, is an illusion based on the fact that all of the intermediary species have gone extinct.

0:43:12.4 SC: Okay, perfectly fair. And so what is it that in the book you talk about the differences between chimps and humans? I mean, how specific can we be? Is it something that a physicist would call a phase transition along the way where there's a relatively quick sort of snowballing effect that changed things relatively rapidly?

0:43:30.2 AB: I think that's probably fair. But at the same time, once you dive into the process by which this ability evolved, you can start to see precisely how it might have come about. I mean, this is not completely well understood by any stretch. Obviously, the archeology of all this is very murky. It's very far in the past. But just in the same way that you can have gradual changes, for instance, as the temperature changes on some water, and then it becomes ice, a similar kind of thing can happen here where all of the different component pieces come about. And then when they're combined in this particular array that we have, it enables some of this extremely complex and powerful behavior and cognition.

0:44:15.7 SC: How close can chimps come? How good is their capacity to imagine the future?

0:44:21.4 AB: Okay, yeah. So maybe we could just describe some of the behaviors that are particularly interesting. So for example, chimpanzees will... Many people know by now they use tools. But they will sometimes even pick up a tool, for example, a rock that they're going to use to crush a nut, a few hundred yards away from where they end up using it. So they have a drive, they're acting on that drive, they're using a tool to solve some problem. And during my PhD, I worked a little bit with chimpanzees. And if you drop a grape outside their enclosure, they will leave the enclosure and, they'll go into the other part of the area, they'll get a stick, they'll peel off the leaves from the stick, and they'll use that sticker to kind of fish the grape from underneath the. So clearly, they can represent a problem in their mind's eye sufficiently to seek an immediate solution to that problem. That's future directed. They also will, they they have various other kinds of future directed behaviors in the sense that they'll navigate.

0:45:27.3 AB: And they seem to do route planning. There's even reports of them hunting, for example, and doing border patrols and this kind of stuff. What what one of the things they don't seem to be able to do, or at least a kind of hypothesis, and I will say a contentious hypothesis, is that one of the things they can't do is actually anticipate a drive state that they do not currently possess. So that could be a, you know, this is drilling down a little bit into the details now, but the Bischof-Köhler hypothesis, as it's so called, and Thomas was one of the people that put this forward, is that they cannot represent, they cannot act upon a drive state they do not currently possess. To make that more concrete, let's say the chimpanzee is not currently thirsty, it will not go and gather water in anticipation of being thirsty tomorrow.

0:46:18.2 SC: I see. And this has to do with the recursion and metacognition we talked about before. We have to be able to run a little virtual machine inside our head that imagines ourselves being thirsty in the future.

0:46:29.4 AB: Yeah. And that's kind of difficult because you, you have to, well, it's very difficult. You have to disengage from the current drives that you have to represent your own preferences in the future. So it requires both a kind of prospective thinking, imagining a situation removed in time, but it also involves representing your own preference function changing in that future.

0:46:57.2 SC: How good are chips...

0:46:57.5 AB: And so that's just an example of one of the things that has not been comprehensively falsified. And so there's no good evidence that they can achieve that.

0:47:07.4 SC: How good are chimps at understanding the state of mind and knowledge of other minds, other chimps, other human beings?

0:47:16.4 AB: Yeah. Well, it's good you raised that because it's obviously very related. In some sense, when you are imagining your own preferences changing in the future, you're using a very similar kind of cognitive capacity as you would use to represent the mind of another person. So if I'm trying to anticipate what you're going to ask me next, what I'm doing is, is doing some kind of mind reading on you, but I might also just apply that same kind of mind reading to myself. Chimpanzees lack the theory of mind capacities that we have. At least they are nowhere near as articulated. There is some tantalizing evidence that they track the false beliefs that other agents, other chimpanzees or other humans have, and some really interesting work by Chris Krupenye and his colleagues, they've done eye tracking experiments with chimpanzees to basically run a version of a false belief task where the chimpanzees eyes are tracked to see if they look at where they think a person falsely believes an object is hidden, even when the chimpanzee themselves know that it's hidden somewhere else. And so if they... And they do do that.

0:48:24.3 SC: And they do. Okay.

0:48:25.7 AB: They look longer at a place where they have reason to suspect that another individual believes something is hidden. But that's a pretty low bar in some sense. I mean, it's an impressive ability, but it's not the kind of, it's nothing near like what we can do where I can appreciate that, you know, that I know something and that meanwhile, my friend Josh doesn't know that either of us know that thing, and it gets a little tricky once you get, you know, too embedded, but the fundamental ability s very sophisticated.

0:49:02.9 SC: Yeah. And so your argument is basically that there is a continuum, it's gradual, it wasn't magic or anything like that, but still there's a big difference in this particular capacity between a modern day human and our closest competition here on earth.

0:49:16.7 AB: Exactly. And I think... And that has various consequences for trying to understand what it means to be human and also trying to understand how to use that ability wisely. Because it's something that other animals don't possess and as you say, it's not magic. It evolved gradually. We don't know exactly how, but it's like any other problem in science where if you accumulate more and more data and you run and you try and falsify the hypotheses as they come up, you could imagine a full picture, at least in principle of how this ability to travel mentally in time evolved through natural selection.

0:49:55.4 SC: Yeah. And I think it does sound to me, smell to me, whatever, like a phase transition in the sense that once you do develop that capacity enough, it just opens up a whole bunch of other things that are important. So I'll ask you some and feel free to mention others that are in there, the ability to make contingency plans, right? The ability to not only make a plan, but to have a backup plan if it goes wrong. That sounds like a pretty human ability.

0:50:22.6 AB: That is probably one of the key defining things that makes it unique. As I mentioned earlier, with the tube task, when young children are given that upside down Y shaped tube, before around age four, they only cover one or the other exit. Well, when you give that exact same task to chimpanzees and other great apes, they also tend to only ever really cover one or the other exit and they'll move their hand back and forth between the two exits. They'll even sometimes switch hands, and they don't seem to represent the fact that they could be wrong about their prediction and therefore that they need to cover both exits in order to guarantee that they're going to get the grape falling out of the tube. They do occasionally use both hands, but then they go back on the very next trial to only using one hand. So the penny hasn't really dropped there. They don't seem to actually understand that in order to guarantee the prize 100% of the time, which can be done, if you set the right contingency plan by putting two hands out, you can't guarantee that you're going to get the prize.

0:51:22.5 AB: That's a rudimentary form of contingency planning in humans. We use that same basic capacity for contingency planning in a huge variety of different important ways. For instance, you might... We all have plan A for career or relationship or anything else in life. And, but you recognize that things don't go according to plan. Reality has a way of getting in the way of your best laid plans.

0:51:49.5 SC: Not to me. I have no idea what you're talking about?

0:51:51.5 AB: You know, and we have plenty of poems songs all about this. Yeah. You know, the possibilities being, being obliterated as reality gets in the way. So contingency planning can be extremely useful there, because for example, you know, this underpins the entire insurance industry because this... And the fundamental cognitive insight that the future might not turn out as anticipated drives us to do something about it. Yeah, so that's just an example of where as you, maybe you're right. Exactly. It could be a phase transition in the sense that if you don't have all the component capacities working together to generate that kind of foresight, well, none of the other implications follow.

0:52:25.9 SC: And the downside of this is that we can have regret, right? We can imagine, oh, I could have done something differently and that would have been better. This seems like a... Is it a uniquely human capacity also?

0:52:39.9 AB: Probably. Yeah probably. I think it's hard to tell. It's one of those, it's another one of those difficult ones where you obviously can't read their mind. You need careful behavioral studies. The studies are difficult to do in this case because you're trying to tease apart the reaction to just not getting what you want from the reaction driven by a contemplation of a counterfactual of what I could have got instead.

0:53:03.3 SC: Sure. That sounds hard.

0:53:05.1 AB: And so those two things are very closely related, but they're not the same. So it's one thing to be frustrated when things don't go according to plan because you had a goal and it was not realized and other animals will rarely get frustrated when their goals are not realized. But what we do is languish over a long ago committed sins or embarrassing things we said a long time ago. We're kind of taking this mental time machine for a spin into an alternative reality and thinking how much better it could have been if I just said something slightly better or reacted in a more cooler headed way to some problem. This mental time machine is a blessing and a curse actually, I think it's not all hunky-dory.

0:53:51.4 SC: But like you say, these are... We have to be able to draw some pretty fine distinctions to say these clearly. I mean, we all have seen pictures of dogs looking guilty after they've done something bad but it's not regret, right? It's a little bit more automatic.

0:54:06.8 AB: I don't know really exactly what's happening in those cases. And I mean, you can imagine a situation where you condition an animal to experience highly negative emotion in response to some stimulus because in the past it has been associated with that negative outcome. So for example, if every time the dog jumps on the couch gets a smack or gets yelled at or something and it jumps off the couch. Well, now when it's on the couch, it might've anticipated the negative outcome. But exactly what's going on in those cases, I don't really know. I mean, there's, fascinating work recently, on dogs, but it's funny enough, they really, the cognition of dogs was ignored for quite a while in comparative psychology, relatively speaking. But recently there's been a kind of a big push towards trying to understand canine cognition, because they are extremely clever animals and they have a long co-evolutionary history with humans. But maybe that takes us too far afield, but anyway, it's a very interesting kind of research area because these are highly social pack animals that have co-evolved with human beings for thousands of years.

0:55:20.1 AB: And which have been now selectively bred for all these traits that we find desirable. And so the cognition there, it's actually a very interesting test bed for understanding some things about the processes of domestication or what does it mean for something to be a threat to these animals and so on. And so it's an interesting area.

0:55:40.0 SC: Is that one of the reasons why they haven't been studied because the co-evolution sort of complicates the whole thing. Like they've been around human beings for so long, we've contaminated them?

0:55:49.0 AB: Oh, they're so contaminated. They're completely... If you're studying domestic dogs, it's extremely difficult. I'm blanking on the name right now, but there's some work being done at the Harvard looking at the brain structures and the cognitive capacities of different domestic dog breeds. And the brains of these animals have also been radically altered by the domestication process just as their bodies have been. So the domestication produces different ears and tails and coats and everything, but it also obviously radically changes their behavior. Unsurprisingly, that is reflected in the brain. The actual brain structures, uh, involved in different kinds of behaviors in different animals and in different dog breeds are correspondingly changed on the basis of what they're good at or what they're scared of or whatever. So it's an interesting one.

0:56:39.5 SC: I will spare our audience my dispositions on cats because I've already... I had Karl Friston on the podcast and I gave him my theory that one of my cats has the ability to imagine the future and the other one does not. The other one just lives absolutely in the moment, without regret or fear. But the first one is sort of always fretting about something or another. So good. Plenty of work to be done by future generations of graduate students here.

0:57:04.5 AB: Yeah, exactly. You just need some dogs and a multimillion dollar fMRI scanner and you'll be set to go.

0:57:10.9 SC: Yeah, exactly. This sounds like one of the more fun things you can do within this subfield. I guess I'd be remiss if I didn't bring up the marshmallow test, right? Because is our ability to restrain ourselves and exhibit willpower, is that related to our ability to imagine the future?

0:57:28.6 AB: So this is close to, very close to what I do kind of on a day to day basis in my personal research work, which is kind of the behavioral economics and the psychology longterm decision making. And so the marshmallow test for people that don't know, I mean, many people have seen cute videos of children trying to resist the temptation and experimental will place a marshmallow in front of the child and say that if you wait until I come back, you can have a second marshmallow. And then the children are left in the room to commence their struggle with temptation. If they eat the marshmallow before the experiment comes back, well then, okay, that's it. If you come back and this it's still there, you get another one. Now, what's pretty interesting there is that, when that study... I mean, there's a lot to say on this. The first thing might just be that as people will know, there's a kind of whole cottage industry in psychology looking at what the marshmallow test performance predicts later. And so for example, it's been found that if people children wait longer for the second marshmallow, well, then they also have higher SAT scores and they perform better in various other ways in later life.

0:58:38.2 AB: There's various longitudinal studies that have been done there. However, the story really is, uh, is actually more complicated than that, than it seems at first blush. So recent attempts to replicate the marshmallow study correlations with later life outcomes has found that it's hugely contaminated by things like SES, socioeconomic status.

0:59:00.8 SC: Okay. Surprise surprise.

0:59:00.9 AB: So for example, if you're in the kind of environment where delayed rewards do not actually manifest as promised, well, then you're also in the kind of environment where it makes little sense to delay gratification, right? So, and so it is... A lot of stuff that I've done, research I've done is looked at the kinds of environments, the kind of situations where it actually makes more sense to prioritize the immediate reward. And that's kind of put the marshmallow test on a slightly different spin on it, which is to emphasize the flexibility of knowing when it makes sense to wait and when it's better to just go for it.

0:59:39.1 SC: And okay. But then what is the connection to just foresight as an ability? Is there any way that we can measure that there's a relationship between one marshmallow versus two marshmallow and one's ability to vividly imagine the future consequences?

0:59:54.5 AB: Yeah, great. So that's an ongoing question. It's whether... To what extent does this ability to actually vividly imagine yourself in the future influence the kinds of decisions where the outcomes play out only over time? And as I mentioned earlier, one of the things that really throws that relationship into question is that patient KC who'd lost his ability to imagine himself in the future, nonetheless discounted rewards and outcomes in the future normally. And so for just for your listeners, if you want to study something like the marshmallow test in adults, they don't respond quite as well to marshmallows as children do. And so, but what they do respond to is money. And so some of the common tasks that you can do, which are kind of analogous to the marshmallow task, but they're not the same, is you ask people to make various decisions about different amounts of money distributed at different points in time. So for example, would you rather have $10 now or $15 in one month? And then if I ask you many questions like that and I change both the amounts of money and the distances in time, then I can compute for a given person the extent to which as rewards move further and further away in time, how do they lose their subjective value?

1:01:09.5 AB: And you can fit preference functions to those, preference to those curves. And you can say, okay, for a given person, how steeply do they discount rewards as they move further way in time? Now that discounting curve can then be used. You can say, okay, well how is that affected by the ability to think about the future? And as I already mentioned, this patient KC who cannot imagine the future discounts normally, so this has led to some suggestions that maybe in order to make flexible decisions about time, you need the ability to imagine the future. And so for example, what we found in some studies and a lot of other people have found now as well, is that if you ask people to imagine future events while they're making these kinds of decisions, you can shift their discounting curve. You can make them more willing to wait for delayed payoffs, probably by making the future more vivid, more concrete and more immediately relatable to the present.

1:02:09.6 SC: Okay. So that's the relationship. So you can shift the discounting curve based on how vivid that imagination is.

1:02:19.0 AB: Yes. Yes, exactly. And there's been some work done by Hal Hirschfield showing that if you show people, for example, virtual reality age progressed versions of their own face, then they are more willing to save for retirement, for example.

1:02:37.0 SC: But also you point out that people are wildly irrational. Like there's the people who would rather have a $5,000 vacation than a $10,000 wad of cash.

1:02:48.4 AB: Yeah. There's some interesting situations where the discounting curve gets a bit funky. So for example, people might sometimes actually prefer to, as you just said, they'll prefer to take a voucher for like a nice massage instead of an equivalent cash prize, even if the cash prize is bigger. And on the face of it, that seems kind of strange because well, couldn't you just buy the massage with the money and then on the way home, have a nice cup of coffee with the change. And what it turns out people are doing in those kinds of situations is actually anticipating that if they take the cash, they'll use it for boring essentials. So if they take the cash, they know their own preference function, they know what they're like, they know that what they'll probably do is get home and they'll end up putting it just alongside the rest of their money, they'll end up buying a toothbrush with it or something. And so instead of that, what people do is they kind of pre-commit to indulgence, which is to say that they intentionally choose a lower value reward that they know is going to be used for a pleasurable experience.

1:03:51.2 SC: This seems like a very, very deep fact about human psychology, which is that we can only imagine the future, but we have a somewhat adversarial relationship to our future selves. We try to trick it and get around it.

1:04:03.0 AB: It's a funny way to put it. Yeah, yeah, yeah, exactly. We're not on necessarily on great terms with our future self all the time or with our past self for that matter. It's a very funny situation. It's almost like there's a kind of game theoretical tension if you think about the mind as being made up of many selves distributed across time, the relationship between all of those selves can actually be modeled in a kind of game theoretical sense as if they were all individual agents. And that leads to some very interesting work on what leads people to pre-commit to decisions where they intentionally tie their own hands in advance of some upcoming challenge, because they know that their future self can't be relied on. Take a classic examples of like hiding the donuts at the back of the fridge or flushing your cigarettes down the toilet.

1:05:01.6 AB: I mean, one thing I do is like most days, if I really need to get some writing done, I now have a website blocker on my laptop.

1:05:07.0 SC: Oh, yeah, yeah, that too.

1:05:09.2 AB: Yeah, it's just turn that thing on. It's like, okay, you are not going on Twitter now for eight hours. You know, this is the workday now. And so understanding that behavior is quite difficult. It's almost paradoxical in a sense because people are in that says intentionally restricting their own freedoms to win a battle against a future version of themselves.

1:05:33.9 SC: Well, because we're self aware enough to know that we're weak. I mean, it makes me wonder whether or not dogs or chimpanzees can have imposter syndrome. Can they like feel that they're not up to a task or just be embarrassed? Like is this ability to imagine all these different hypotheticals really weighing us down with anxiety as human beings in a way that other animals don't suffer?

1:05:55.4 AB: I think so. I think one of the things, one of the interesting things that Endel Tulving, who was responsible for making that semantic episodic distinction that we talked about earlier, one of the things he said about KC, the patient who lost access to the future is that he's in many ways, he's happier than the rest of us because he doesn't worry about the future. And he doesn't worry about the fact that he's going to die because he has no mental access to that, to the fourth dimension. So he's not plagued by that. And I think on a moment's reflection, people will recognize that their mental time machine will often dredge up really negative possibilities, on the face of it, you could imagine a functional evolutionary explanation for that, which is that it's very useful to anticipate threats in your environment, but it's a double-edged sword.

1:06:50.7 SC: It is.

1:06:51.1 AB: It's simultaneously extremely functional in the sense that it lets you prepare for upcoming dangers, but it also plagues you with all the worries about all the possible things that could go wrong. And that's especially problematic because of the mutually exclusive nature of our mental time machine. The fact that I could say, oh, this could go wrong or, or maybe this could go wrong, you just go through the list. It's like, here's a thousand different ways that this thing could go wrong. And you have mental access to all of them.

1:07:21.0 SC: Lucky. Well, I do want to... Now that we've gotten that on the table, we can speak up for some of the positive aspects, right? I mean, there's a relationship that there's a million relationships, maybe pick out your favorites between the ability to imagine the future and developing tools, social cooperation, language, right? Like all of these things that we associate specifically with humanity and its social structures in some way, they're super reliant on this ability.

1:07:51.6 AB: Yeah. Fundamentally reliant. And actually this, this is nice because it brings us to kind of come full circle on the evolutionary question, which we opened that door and we didn't really finish the kind of point there, which was how did this thing evolve? This ability to imagine the future evolve in our hominin ancestors. And in the book we kind of, we, like I said, we start at the split leading to modern chimpanzees and we kind of go through how this happened. Maybe I'll just take a waypoint, which directly addresses your question about some of the key technologies to illustrate. So something like a bag, I mean, I know it sounds funny on the face of it, but a mobile carrying container is actually one of the most important fundamental innovations that human beings have ever devised.

1:08:37.3 AB: Forget your fire and your axes and stuff. The mobile container is actually extremely important and often overlooked in archeology, but it's foundational because it represents this ability to anticipate that what I currently have, I could bring with me into the future. And not only that, but I can now store multiple objects that I might need given what the future could hold. And once you have a mobile container that actually produces new incentives, new selective pressures on preparation, because only now does it make sense to prepare multiple tools in advance. Before mobile containers you basically, if you make a tool, you can carry it with you, but that's it. You only have the carrying capacity of your hands. And that's not enough. If you want to actually become a species that has now found its way to every ecosystem on the planet, including some extremely hostile ones, you need a kind of mobile life support system. You need various tools, you need clothes, you need all these different bits and pieces, and that's unlocked by the ability to carry them with you.

1:09:49.0 AB: And so this actually illustrates something kind of fundamental, which is, there's actually a feedback loop between the technology that we create using our foresight and the evolution of foresight itself. Because if I use my foresight to create a bag in collaboration with many other people, well, now that bag itself feeds back to create new pressure on the next generation of people who are now growing up in an ecosystem where bags exist, in a society where that can be taken advantage of, it's a new feature of the environment. It's a constructed niche. And so now that is going to feed back to actually increase the far sightedness that brought the thing into existence in the first place. And now you can imagine how that might ratchet up.

1:10:33.8 SC: Do we know when the first bag was ever used?

1:10:37.1 AB: We don't know when the first bag was ever used. And the reason is that they're made of materials that don't tend to last the ages. There's evidence from around a hundred thousand years ago that, hard evidence that people are using shells to store things like pigments for creating markings. And then even more recent, 50,000 years ago, there's evidence of... There was a great excitement in the scientific community when even a scraggly little piece of cord was found from a Neanderthal site, because it's like, well, that's amazing. It's just literally like a tiny little piece of cord, but it implies that they were creating, possibly slings or straps or some kind of a carrying device. I mean, who knows what they were actually using it for.

1:11:22.4 SC: And is this kind of tool use come before language as far as we know?

1:11:30.7 AB: Well, the evolution of language is tricky to pin... Very tricky to pin down. Tool use and language probably co-evolve in some very fundamental ways. So for example in a similar way, as I just laid out with the other feedback loop with bags, as you become increasingly reliant on tools, it also becomes important to be able to pass those tools to other people and to teach them how to use them. And so we've already addressed it, one of the technological aspects, mobile containers as an example of a piece of technology that foresight enables, but as you mentioned as well, there's also social institutions and teaching is one of them. So in order to teach, you represent the mind of another person and you anticipate what they need to know in the future. And then you attempt to shape their mind towards the goal of that expertise. And so teaching is actually really fundamentally reliant on foresight as well, because it involves people who are in the know transmitting lessons to people who are not in the know in order to, to pass those on.

1:12:40.8 SC: It just seems like there's a very natural opportunity at least for co-evolution between language and foresight. For one thing we can tell stories, right? I mean, storytelling, I don't know what people think about the origin of language. Did we tell stories about the origin of language? Did we tell counterfactual stories as soon as we started constructing complex sentences? Probably not, but I honestly don't know.

1:13:03.4 AB: Yeah. The argument has been made that ultimately language evolved at least in part and maybe primarily to share our mental time travels.

1:13:12.7 SC: Oh, okay. There you go.

1:13:15.3 AB: And, yeah, so exactly. As you sort of insinuated there, like there is a very deep relationship between these capacities and it might even be the case, although it's hard to prove that the reason we even have language in the first place is because it enables us to transmit these lessons about things that are not here and now. And linguists will point to displacement, what they call displacement as a key feature of what makes human language different from the communication systems of other animals. Displacement just refers to the fact that our language systems can represent or refer to things that are not in our immediate perceptual environment. I can I can use my language to refer to things that are distant in time and space.

1:14:00.7 SC: And what's the relationship of that between that and the ability to use language to refer to hypotheticals, to say... Or to make a contract, say, if you do this, then I will do that. That seems like something we couldn't have done without language and the ability to add some foresight.

1:14:18.8 AB: Yeah that's a great example. So in in the book we talk about various things like even exchange systems and bartering, contracts are a big part of that. Relying on foresight, again because you're representing the fact that in the future you owe me something and I expect to get it back and that might not be for a long period of time between now and and then. And so accounting systems, there's a good good reason to believe that those essentially first arose because they enable you to off-source the difficulty of keeping track of who owes what to whom, and when the bill is due across time. And so writing itself provides a tool, a cognitive offloading tool for externalizing representations of future events that need to be kept track of. Bearing in mind, as I said, this kind of contingency planning thing, you represent so many mutually exclusive possibilities that there's too many to carry. In the same way that you need a bag to carry all of your tools, you might need external artifacts where you can encode and represent all of the possible futures that you need to keep track of.

1:15:27.9 SC: I mean, it does seem like a bunch of things happened at once. I don't know if there was a single event, but they do co-evolve. I'm going to start using the word co-evolve in all my sentences from now on because it's so very helpful. But yeah, foresight, tool use, social cooperation, contracts, language, storytelling, and we still haven't equilibrated. We're still feeling out what that means and what we can do with all these capacities. And I'm not even going to mention Twitter drama, but it's an example of all the ways that we're still learning how to communicate and what it means for us.

1:16:10.4 AB: Yeah, exactly. I mean, learning how to communicate and what it means to us. And I think also in some way learning how to use our abilities to think about the future and plan for it. This will come as no surprise to anyone that our ability for mental time travel is not clairvoyance. It's definitely not perfect by any stretch. And many of the errors of planning and anticipation are some of the most catastrophic that we make because many reasons. We can go into some of the details about how the psychology can let us down in the modern era. But many situations where you invest in some vision of the future and you believe that this is going to work and it doesn't work and you fail. And so trying to get to grips with how to use foresight, especially in terms of its relationship to our technologies, is I think one of the key challenges that we're facing now. I mean, because if we're going to navigate into a better future, one way to do that is to understand the mental machine that actually enables you to think about the future at all in the first place.

1:17:22.8 SC: Well that's perfect because it leads in, it's sort of the flip side of what I was going to do for the very last question to wind things up. You just mentioned how there's sort of systematic ways in which our predictive abilities fall short sometimes. But on the flip side, sometimes you make pretty good predictions, but we don't take them seriously enough, right? If they're too long term or too outside our scope, we're seeing this with various issues like climate change and other big picture risks to the earth. Is there some sense in which we can be better at actually believing our own abilities in this sense? Is there something that we can hope for for the future as we imagine possible futures?

1:18:04.9 AB: Yeah, I think we need to be optimistic about our abilities to think about the future. I mean a slightly whimsical way of thinking about it, but it is true, is that because we are the only animals with this access to the fourth dimension that we do have, that also burdens us and us alone with the responsibility to steward that future and to get it right. No one else is going to do it. And so on the one hand, it's responsibility, but on the other hand, it's a nice way of saying that actually we are the most far sighted creature that has ever existed. And we have an incredibly sophisticated and powerful mental time machine in our skulls that we could use to really create a better future. So that's one optimistic way of thinking about it. But just to flip the question on its head a little bit, I think if anything, we need to believe less in our abilities for mental time travel. And what I mean by that is just that we need to adopt that sort of skepticism that we talked about earlier, the kind of metacognitive appraisal of our own abilities when you say, okay, how could this go wrong?

1:19:08.4 AB: Much in the same way as the young children with the fourth tube task, it will not work to just put one hand out underneath the tubes. We need to realize that we could be wrong. And so I think many of the problems that we really face are actually people being too gung-ho in their predictions and getting too invested in what they believe the future will be like. But when you realize that the future is actually just a, like we talked about already, a mental theater, a conjuring trick in a way where we've created this mental representation, it's actually very similar to just making something up completely fictitiously. And so realizing that puts a kind of a fine point on the skepticism that we should apply to our own predictions, and obviously science is all about doing that. It's looking at your predictions and attempting explicitly to falsify them.

1:20:00.2 SC: These are very good things to think about about the future. I'm not sure whether it's an optimistic or pessimistic conclusion right there, but anyway, we should take seriously our ability to make contingency plans if that's one of the things that evolution has given to us. So Adam Bulley, thanks so much for being on the Mindscape podcast.

1:20:15.0 AB: That was great talking to you. Thank you very much for having me. I really appreciate it.

[music]