Discovering this human-primate economy

https://www.parsingscience.org/wp-content/uploads/2021/03/ParsingScience096-JB_Leca.mp3#t=02:26,07:11

Leca: As I was doing my second postdoc in Japan, we heard that stone play – the object play behavior – that we were studying in Japan was also occurring in Bali, in another species of macaques: long-tail macaques, Balinese long-tail macaques. So we decided we would take a short trip to Bali, to check out this population. So we did, and that was in 2008. And we found out that this population of monkeys were indeed engaging in this stone play behavior. So we spent several weeks at this field site in Bali. And then we heard about another population of monkeys. And we heard that apparently, they were not engaging in this cultural stone play behavior. So we [said], “okay, we’re going to look at this population to have some kind of control group.” So that’s when we visited, for the first time, this field site at Uluwatu on the southern tip of the island. And when you arrive at Uluwatu, it’s a Hindu temple, and the monkeys are free-ranging: they live around the temple grounds. They are free ranging but provisioned every day by the temple staff with extra food, because they can spend a lot of time in the forest and on the cliffs. The temple overhangs the ocean. But as soon as you enter the temple grounds, you are warned – right from the start – that you should remove all your valuable belongings. There are signs everywhere in the staff let you know that that’s a practice of these monkeys. They steal objects from tourists, and then they engage in something that looks like a bartering interaction.

The little story behind the science here is that when we arrived – Mike Huffman and I at this field site – we are both short-sighted, so we wear prescription glasses. So, we took off our glasses, put them in our bags, and we entered the temple ground, but after a few minutes, and Mike had a terrible headache – couldn’t see anything from far – so he said, you know, “I’m a professor in primatology, I will see … I will see the monkeys coming.” The problem is the location of the temple – really it overhangs the ocean, and the monkeys spend a lot of time on the cliff and sometimes they come up suddenly from the cliff – and if you just stand on the fence, looking at the ocean, you can be surprised that’s exactly what happened. And he got his eyeglasses snatched in a split second. Very swift action on the monkey’s part. He said, “That’s it. I’m done, if I don’t have my prescription glasses, my visit in Bali is over.” But the monkey was not leaving. The monkey was staying. It was sitting there looking around, and we had no idea with what was going on. And after a couple of minutes a member of the temple staff came and started to engage in what looked like a bartering interaction: offering a number of food rewards to the monkey, who was still holding the prescription glasses. The monkey was doing some gestures with its arms, as if he was declining some of the first food rewards. And at some point he accepted the food reward and returned – more or less gently – the prescription glasses. So we were … we were … we were stunned. We were stunned by what had happened. But then we learned that daily occurrence over there. It happens all the time; many, many times everyday, many monkeys engage in this behavior. So that was 2008 and I kept the event in mind, and I came back with my wife – who’s also a primatologist – in 2010. We were engaged in a completely different research project – conservation project, actually, in the national park – and, I said, “Let’s go back to Uluwatu, because I want to see this behavior from closer.” That’s when we started collecting our first data with video cameras. And I realized that there was the potential for a real research topic there. So 2010, we collected our first data. And then I hired a postdoc when I got a position here at the University of Lethbridge. And this postdoc collected 12 months of data on this behavior in this population.

[ Back to topics ]

Robbing/bartering

https://www.parsingscience.org/wp-content/uploads/2021/03/ParsingScience096-JB_Leca.mp3#t=07:10,09:35

Leigh: After stealing objects from temple visitors that are pretty much otherwise worthless to a monkey, the macaques use them as tokens: returning them only in exchange for a certain quality and type of food rewards that the temple staff offered to them. But Ryan and I wondered: why don’t the victims just do the trading themselves?

Leca: The temple staff are the main human barterers. When the tourists try to barter with the monkeys, it doesn’t go well. Because tourists visiting the temple, they don’t have the array of food rewards that the temple staff always have handy. And we’ve seen many times a tourist – who’s not aware of of the deal and how it goes – being so surprised and pissed off with having their valuables stolen … they’re gonna grab anything around them. They’re gonna grab a twig. They’re gonna grab a bunch of leaves, and literally hand them in to the monkey, and clearly the monkey’s going to decline them. Because obviously that’s not … even though they feed on leaves sometimes, that’s clearly not the kind of food reward that they expect from the exchange. So you have to picture these events as being highly emotionally loaded: not necessarily expecting to have, literally, a monkey climbing on your shoulders and snatching your sunglasses, or eyeglasses, or cell phones. But it happens. And I’m absolutely not saying that there is any form of aggressiveness in these interactions. The monkeys are extremely, fast extremely swift. Skilled. And sometimes the body contact – the physical contact between the monkey and the human – is just, you know, a couple of seconds. And we reported no injuries on the human side. That’s not aggressiveness, but still the tourists are certainly emotionally disturbed by the interaction. So the temple staff are the accustomed human barterers. And there are many members of the staff all around the temple grounds. And they have all the array of food rewards that the monkeys may want at this point. So that’s the interesting part of the story: it’s like the monkeys now can apparently make a decision based on both the type of object they have in hand and the type of food reward – or maybe the number of food rewards – that they might have in mind.

[ Back to topics ]

Token economies

https://www.parsingscience.org/wp-content/uploads/2021/03/ParsingScience096-JB_Leca.mp3#t=09:34,12:03

Watkins: Sort of like symbolic tools, tokens are intrinsically non-valuable objects that only acquire an instrumental or functional value through the arbitrary associations made with goods or services. But prior research into human-animal token bartering has predominantly been done in laboratories, far removed from the naturalistic setting in which he and his team carried out their study. So, to this end, JB’s field research aimed to address the desire for a more ecologically valid primate model of human trading systems.

Leca: The token exchange paradigm, so far, has mainly been conducted in the controlled settings of laboratories in captive non-human primates: in great apes and monkeys. The idea is to test whether non-human animals have a sense of value for objects that are not edible. If they can engage in some form of economic behavior with a human experimenter assigning a value to a particular token through training, of course … and then buying – literally buying – a number of food items based on these tokens, these symbolic objects that most of the time have no direct, iconic reference with the food reward that’s going to be offered. Many, many studies have shown that several non-human primate species – monkeys and great apes – have many of the cognitive abilities … both symbolic and computational abilities … that are necessary to think economically and to produce economic behaviors in this context. And these behaviors somehow look like what we have in humans, even though to a lesser extent, of course. So, economic behavior in its most basic form relies on accurate calculation of subjective and objective value. [And] accurate memory of past events that pertain to present choices. Properly calculated anticipation of future outcomes, or some kind of prospection. And some level of inhibitory control. That’s also something we will be exploring in the next step of the study: whether our monkeys – to some extent – are able to delay gratification, and decline a number of food offers, possibly anticipating a next one that could be better.

[ Back to topics ]

Data cleaning

https://www.parsingscience.org/wp-content/uploads/2021/03/ParsingScience096-JB_Leca.mp3#t=12:02,16:00

Leigh: In Episode 72 of Parsing Science, we talked with Akchou Rasphone, who used timed camera traps to document the declining population of tigers in Laos, resulting in 1000s of photos taken over a two-month period. And since an endangered species is, by definition, rare, a lot of those photos were pretty much nothing, and all of which she had to manually screen out. So Ryan and I figured that if JB and his team had to transcribe hours of video-recorded robbing/bartering interactions with over 300 individual macaques, then it must have been involved quite a lot of data cleaning prior to being ready for analysis.

Leca: It’s quite a long process, for sure. We have several hundreds of hours of video, but these could be considered “rushes.” Because there is quite a lot to discard in some of these videos. So we did a first cleaning process, discarding the cases in which both the robbing and the bartering interaction were not clearly filmed, or we could not see everything. We we discarded a number of these videos, and we were left with, still, a relatively large number of hours of videos from which we started the scoring process. And the scoring processes is still ongoing; we’re far from being done.

So, we are going to dissect further the first component of the behavioral practice: the robbing interaction. Because we are interested in investigating … examining whether the monkeys have social strategies in stealing objects from humans. What appears from the observation is that, generally speaking, the monkeys will come from behind. That speaks to, somehow, perspective taking. And skilled monkeys probably have a sense that, “If I come from behind, you won’t see me. And my chance to grab an object from you by surprise is higher than if I come, you know, up front … and you see me approaching. And you move around and so on and so forth.” This will speak to social strategizing and perspective taking, and maybe some form of theory of mind: what do the monkeys perceive about humans perceiving them? And how they can increase their chance of successful robbing, if they approached this way or that way, and so on and so forth. So that’s the robbing part. And then, with regards to the bartering part, we scored the time lag between the end of the robbing event and the end of the consumption of the food reward. And pretty much everything in between, including the number of food rewards that were proposed, the behaviors of the monkeys when confronted with these food offers. We also have the behaviors directed by the monkeys to the token. We are going to investigate that. Because we have some indication that to help them wait for – sometimes, several minutes – some of these monkeys might have developed some individual strategies. And some monkeys start to bite a little bit at the token, sometimes to the point that they damage the token. And others don’t seem to need to bite the token, and they can hold it and just look around. So again, a lot of inter-individual variation, just like humans, right?

When I teach personality, one of the major tests is the Marshmallow Test. And the Marshmallow Test is quite different from what’s going on there. But it’s basically: “Okay, I give you this marshmallow – one piece of Marshmallow. You can eat it now. Or, if you wait several minutes until I come back, I will give you a second one and then you can have two marshmallows, right?” So this this speaks to the ability to delay gratification. We know that in humans, this ability has long term consequences. So, you know, is that the case for these monkeys? How did they learn to wait, that’s also something we are interested in.

[ Back to topics ]

Social learning theory

https://www.parsingscience.org/wp-content/uploads/2021/03/ParsingScience096-JB_Leca.mp3#t=15:58,17:51

Watkins: Social learning theory emphasizes the importance of observing, modeling and imitating the behaviors, attitudes, and emotional reactions of others in order to learn. The theory considers how both environmental and cognitive factors interact to influence learning and behavior. And because the area around the temple allows the monkeys to both interact with and observe the behaviors of humans and other monkeys, this leads to opportunities for learning directly, as well as indirectly.

Leca: So when we say “indirect social learning,” we’re talking about those examples in which the demonstrator is not present anymore. But the demonstrator, or the skilled individual, left in the environment – in the cultural landscape of these monkeys – physical traces of their behaviors. Of their skilled behaviors. So what you should picture around this temple grounds is that the surroundings are filled with discarded tokens. And young monkeys are going to start to learn from discarded tokens. They’re gonna pick up a token, and sometimes they’re gonna walk towards a temple staff, and apparently wait for some kind of bartering. But if it’s an old, discarded, damaged token, the monkey usually won’t get anything in exchange … or not much. And we think that this indirect social learning through the objects themselves that are part of the cultural landscape of these monkeys are also part of the learning process. So, that’s something I find fascinating: the idea that somehow there is an external memory of past behaviors – that is left as a physical trace in the environment – for other monkeys to learn, even though the monkey demonstrator is not around anymore.

[ Back to topics ]

Cross-sectional analysis

https://www.parsingscience.org/wp-content/uploads/2021/03/ParsingScience096-JB_Leca.mp3#t=17:49,20:35

Leigh: By conducting cross-sectional analyses based on age differences, JB and his team used the proportion of successful token-robbing and reward-bartering attempts as measures of the monkeys’ experiential learning. To calculate these percentages, they used each macaque as its own control, distinguishing for each attempt two possible outcomes – successful or unsuccessful – out of the total number of attempts by each monkey. We’ll here what they found out from this approach after this short break.

[ad: Figshare]

Leigh: Here again is JB Leca.

Leca: The first hypothesis was based on experiential learning. And we wanted to take – unfortunately, not a longitudinal approach to it – but a cross sectional approach. So we compared the individual from different age classes – juveniles, sub-adults, and adult monkeys – in their ability to steal objects efficiently, and then to barter for food … again, more or less efficiently. And our basic prediction was that there would be a more or less linear increase in these skills from juveniles to sub-adults and then adult monkeys. And this prediction was only partly supported: we were surprised to realize that sub-adult monkeys actually are quite good at stealing valuable objects. And engaging in, again, some kind of economic behavior during the bartering component of the behavioral practice. So we didn’t find any statistically significant differences between sub-adults and adults when it comes to robbing skills and bartering skills. The juveniles are not so good at that. It seems that it’s only through – probably personal experience – that adult and sub-adult monkeys have learned these skills. And also probably through observing other monkeys. But it takes several years in this species, probably about four years. That’s the transition between juvenescence and the sub-adult stage. So again, we initially predicted a gradual increase in robbing/bartering skill with age. But, in the end, we found that sub-adult monkeys between four and six years were almost as successful as adult monkeys beyond seven years of age.

[ Back to topics ]

Robbing higher value items and bartering for more food

https://www.parsingscience.org/wp-content/uploads/2021/03/ParsingScience096-JB_Leca.mp3#t=20:33,23:19

Watkins: JB was also interested in whether the macaques would seek to maximize their investment in their ill-gotten goods by trying to steal things of higher value from people, as well as whether they would use the token/reward bartering phase to get more food rewards … or a more preferred food when they stole those higher value items. If so, then that would provide evidence [that] they’re able to perform complex economic behaviors – like trading and cooperation – according to the same kind of criteria that humans apply to basic economic decision-making.

Leca: We wanted to test whether the monkey had learned to preferentially target objects that are more valuable … but of course valuable for humans. So for that, we combined both observational and experimental studies. And in the observational studies, we found that, again, sub-adult and adult monkeys tend to preferentially select objects that are more likely to be exchanged for food – so, electronics, sunglasses, or prescription glasses – over other objects that are less valuable for humans … and typically, not worth bartering, or bartering with less food or, or maybe lower quality food. And again, that’s probably something they learned through some form of operant conditioning. If they find themselves with, let’s say, an empty camera bag that they just stole, the probability for the tourists to say, “Forget about it, I’m not going to claim this object back; it’s not so valuable,” is quite high. And then the monkey is left with an object in his hand. And there’s no bartering, or the probability of bartering is lower. Or it might be delayed, you know? Because afterwards, the staff may try to retrieve some of these objects, but the tourist is gone. And then the value of the food proposed for an empty camera bag is generally lower [from] the temple staff end. And so somehow, it seems that the monkeys have learned something about this association between types of objects, and – not only the number of food reward that they could get in exchange for this object – but also the type of food reward. So that’s the token-selection part. And then there is the robbing/bartering payoff maximization hypothesis – which is suggestive of rudimentary economic decision-making by these monkeys – whereby the monkeys seem to have mapped their robbing and bartering behaviors on to our own scale of values when it comes to material stuff. So again, there is a lot of inter-individual variation.

[ Back to topics ]

Non-human animals’ engagement in economic behavior

https://www.parsingscience.org/wp-content/uploads/2021/03/ParsingScience096-JB_Leca.mp3#t=23:16,26:29

Leigh: JB and his team’s study produced three pretty amazing findings about economic decision-making and macaques. First, to be successful, it takes them about four years to learn these behaviors. Second, older macaques preferentially select tokens that are more valued by humans. And lastly, more skillful and selective monkeys appear to make economic decisions, as they clearly make behavioral associations between value-based tokens and the quantity or quality of the food rewards that they rejected or accepted. So Ryan and I were curious what JB sees as the most compelling of their findings.

Leca: What I find interesting – in this population, from an evolutionary perspective – is that, for the first time, we have some evidence that non-human animals can learn to engage in economic behavior. And somehow think economically, even if it’s more rudimentary than humans, of course. They have some kind of economic decision-making process. And these skills, these abilities, can be transmitted in a social context. Because in most cases – not all the studies – but in most studies conducted in captivity, we are dealing with individually-trained individuals. The example of a monkey learning from another monkey how to engage in a bartering interaction like that are not so frequent. So, for the first time, we are showing that free-ranging animals can learn at the population level to engage in what looks like the basis of economic behavior. To me, that’s the first important point, that somehow draws a parallel between the human economic world and what these monkeys do: again, in their cultural landscape. To me, that’s very important. So, yeah, ultimately, our goal is to explore the evolutionary origins of economic behaviors and cultural behaviors. And, of course – because behaviors don’t leave a lot of traces in the fossil record; it’s not like bones, of course – one way to probe into the ancestral past of our behaviors, including economic behaviors, is to compare the behaviors of different but closely-related species like monkeys, and humans, and then try to reconstruct evolutionary scenarios about step by step changes in behavior. So it’s a bit technical, but as I’m sure you know, one of the most powerful methodological tools in behavioral and [psychological] science is called the “comparative method.” So that’s what we are doing: we’re using the comparative method to reconstruct evolutionary scenarios about step by step changes in economic behaviors in our common primate ancestors. But, of course, that’s a long story. But the idea is: if at least we can find some evidence of socially learned economic abilities, then we are making, you know, one step further into comparing what could have occurred in [our] human ancestors … and then what’s going on with these monkeys right now.

[ Back to topics ]

Who’s training who?

https://www.parsingscience.org/wp-content/uploads/2021/03/ParsingScience096-JB_Leca.mp3#t=26:26,28:50

Watkins: As the Uluwatu macaques have been engaging in these kinds of behaviors for the past 30 or 40 years, Doug and I wondered: just who’s training who? Is it we humans that have taught the monkeys how to cooperate, or the monkeys who have trained us in their care and feeding?

Leca: There are many variables that we want to control through some kind of field experiments. But, clearly speaking, I remember filming an event of a monkey rejecting a banana. Rejecting … and clearly rejecting – when I say rejecting: grabbed the banana, threw it away. And then being offered a small bag of fruit pieces. And clearly – with an arm gesture from up to down – clearly rejecting the bag of food rewards. This sequence lasted several minutes. The monkey rejected food offer after food offer, still holding on to the glasses – the prescription glasses – until what was proposed to the monkey was a large piece of cucumber. And that’s when the monkey decided to return the glasses. Somehow the monkey is in charge of the deal. Not always – because again, you have to remember we’re talking about a situation in which the tourists may want to get their items so bad that sometimes, unfortunately, the temple staff members are going to threaten the monkeys to get the token back. So these are cases that we want to clean up in our data set. Unfortunately, there are several of these cases. And that’s what we would like to control in our field experiments: there won’t be any threats directed to the monkeys anymore. But only fair dealing.

[ Back to topics ]

The macaques’ motivation to rob and barter

https://www.parsingscience.org/wp-content/uploads/2021/03/ParsingScience096-JB_Leca.mp3#t=28:03,31:40

Leigh: There’s a sufficient amount of natural food sources surrounding the temple … and any way, the temple staff feed the monkeys daily, independent of the robbing and bartering. Add to that 1.5 million potential human victims that visit the temple each year, nd it’s an understatement to say that the monkeys are very well fed. So might the monkeys be doing what they’re doing for sport?

Leca: This would need to be tested through, you know, carefully collected and analyzed data. But if you ask me personally – based on a long time observing this behavior – they partly do it for the food reward. But I’m not sure they are terribly, terribly hungry. So they are provisioned with food, in addition to, you know, the wild food occurring in the trees and the bushes around the temple grounds. We’re not talking lush forests, right? The ecological conditions are not extremely harsh, but it’s not so easy for these monkeys to find natural food around. And there’s seasonal variability, and so on and so forth. But they are called crab-eating macaques, right? So they can also go down the cliff and look for crabs and shellfish and this kind of thing. So, they do have a lot of natural food around. And they spend a lot of time naturally engaging in foraging behavior. But when they come back to the temple grounds … if you ask me personally, I think many of our cases might be the result of … getting a kick out of it. It’s like, there might be a thrill – particularly adult males – there might be a thrill in engaging in a successful bartering behavior. And then, you know, knowing what’s probably gonna happen next: they wait, they get a food reward. And because we have the data on the food consumption, we could actually test the level of hunger of these individuals across the day. So, one thing we want to do next time we go there – when when the field site reopens – is that we’re gonna follow the same monkey throughout the day. Because that’s not what we’ve been doing so far. So far, we’ve been opportunistic in our data collection. But if we can follow the same monkey – probably not throughout the day, because again, the landscape might not allow it – but at least for some period of time, at different time periods throughout the day, then we might have a sense of anecdotal observation. But we got the feeling that some individuals might go on some kind of “shopping trip” throughout the day. And it seems that some individuals may start the day robbing some items, and then only accepting some kind of salty food rewards, like crackers and these kinds of things. Then a little bit later in the day, they accept fruits, and sometimes vegetables or a little bit sweeter food. And towards the end of the day – again, it’s only anecdotal needs to be tested, of course – but some monkeys at the end of the day, they will reject food offer after food offer. They will accept a plastic bottle of water, and of course, they will remove the cap and they will drink. I mean, the idea is, “What’s going on there?” Do they have a sense that they can pretty much fill their day with a range of food items, and they’re gonna go through the exchange from, you know, a type of food rewards, and then they’re going to end their day thirsty, and that’s when they accept water and they actually drink.

[ Back to topics ]

The conservation of long-tailed macaques

https://www.parsingscience.org/wp-content/uploads/2021/03/ParsingScience096-JB_Leca.mp3#t=31:37,33:32

Watkins: While the macaques easily adjust to human settlement. They’re often considered to be passed by farmers and villagers. So – before wrapping up our conversation – JB wanted to share his concerns for the future of long-tailed macaques.

Leca: Many non-human animal populations are collapsing. And with regards to this specific species, long-tailed macaques are not doing good in terms of population status and conservation-wise. So, what I hope is that some of the research we do in my lab – but [also] you know, other kinds of behavioral research – is going to bring attention to, you know, stakeholders and decision makers about the need to maintain the environment for these non-human animals. Because they do carry a lot of very interesting knowledge. And we can learn a lot from observing the behaviors of these monkeys. When you study cultural behaviors – particularly cultural behaviors – that occur in close proximity to humans, the probability for human expansion to wipe out many of these cultural-traditional behaviors is extremely high. And in many places throughout Southeast Asia, the population of monkeys – macaques, particularly, but other non-human primate species – have completely disappeared. So, yeah, I always try to make people aware of the need to maintain the habitat as intact as possible. Of course, that’s not the case anymore in many, many places, but some some kind of environmental awareness, let’s say, and conservation awareness. So, again, it’s all about involving people living in these places where non-human primates live, and make sure that they are still there for generations to come.

[ Back to topics ]

Links to manuscript, bonus audio and other materials

Leigh: That was JB Leca, discussing his open-access article, “Acquisition of object grabbing and object food bartering behaviors: a culturally maintained token economy in free ranging long-tailed macaques,” co authored with Noëlle Gunst, Matthew Gardiner and I. Nengah Wandia, and published on January 11th 2021, in Philosophical Transactions of the Royal Society B: Biological Sciences. You’ll find a link to their paper at parsingscience.org/e96, along with transcripts, bonus audio clips, and other materials that we discussed during the episode.

Watkins: If you want to catch up on recent episodes of Parsing Science that you may have missed – like Episode 95, in which Shiri Melumad discussed with us how information gets increasingly opinionated and negative when repeatedly retold … or the one before that in which Lindy McBride and Zhilei Zhao joined us to discuss their research on how mosquitoes are able to track us down so quickly while ignoring other warm blooded animals – then head over to parsingscience.org to check them out.

[ Back to topics ]

Preview of episode 97

Leigh: Next time, in Episode 97 of Parsing Science, we’ll talk with Alexander Puzrin from ETH Zurich about his research into a 62-year old mystery over the deaths of nine hikers in the freezing Russian wilderness: a tragedy that’s been attributed to everything from a Yeti, to military weapons testing, and an avalanche.

Puzrin: This was not like they were all found on the slab of snow. They were found all in different places, about 1.5 kilometer(s) to 600 meters away from the tent. And with very – some of them – with very strange injuries. So they had to open [a] criminal investigation. But then they closed it.

Leigh: We hope that you’ll join us again.

[ Back to topics ]