The ability to plan for the future beyond immediate needs would be adaptive to many animal species, but is widely thought to be uniquely human. Although studies in captivity have shown that great apes are capable of planning for future needs, it is unknown whether and how they use this ability in the wild. Flanged male Sumatran orangutans (Pongo abelii) emit long calls, which females use to maintain earshot associations with them. We tested whether long calls serve to communicate a male's ever-changing predominant travel direction to facilitate maintaining these associations. We found that the direction in which a flanged male emits his long calls predicts his subsequent travel direction for many hours, and that a new call indicates a change in his main travel direction. Long calls given at or near the night nest indicate travel direction better than random until late afternoon on the next day. These results show that male orangutans make their travel plans well in advance and announce them to conspecifics. We suggest that such a planning ability is likely to be adaptive for great apes, as well as in other taxa.

Introduction

It seems obvious that animals, just like humans, would benefit from being able to plan, i.e. to follow a self-determined goal over a longer time period (i.e. beyond its current motivational state [1], [2]). This way, short-term motivational states such as hunger would perhaps briefly divert but not dissuade the animals from pursuing this superordinate goal. Optimal exploitation of a landscape's resources often requires movements between different areas in specific time periods. Similarly, finding a mate may require moving to distant locations where the presence of the opposite sex can be anticipated. Planning for the future beyond immediate ecological or social needs should therefore generally be favored by natural selection. However, despite these potential benefits, the evolution of this ability may be prevented due to the caloric costs of the increased brain size needed to implement the requisite cognitive capacity [3]. Moreover, most animals probably manage well without the planning ability. First, reliance on a set of innate rules to respond to a variety of environmental cues may often be sufficient to achieve near-optimal scheduling of maintenance and social activities and space use, even on time scales well outside current motivational states, such as long-distance migration [4]. Second, cognitively simpler mechanisms, such as associative learning, may be equally effective [5].

Planning for the future is therefore expected only in a subset of animals. First, the ability should be selectively favored where the target location changes frequently and unpredictably and travel costs are high. A high efficiency in range use is especially at a premium because animals have only a limited scope for increasing their foraging effort without reducing their net daily return [6]. Second, animals must be able to bear the energetic costs of the brainpower needed for such a high-level cognitive ability. Thus, species that are already relatively large-brained may have a head start in evolving the ability to plan ahead. Indeed, the ability to plan for future needs has long been considered uniquely human [7]. This ability critically relies on two cognitive abilities: self-control and mental time travel [8]. Self-control is the ability to suppress immediate responses and thus delay reaping a reward, if necessary for long periods of time. Mental time travel, the capacity to construct mental experiences of potential events, critically involves the presence of episodic memory, the recollection of specific events [9]. This episodic system is active when the brain is at so-called wakeful rest in humans [10] and probably chimpanzees [11].

The question is whether this ability to plan for the future is indeed limited to humans. Recent experiments have provided evidence for the presence of both self-control [8], [12] and episodic memory, the two key ingredients of planning for the future, in a range of animal species [13]–[16]. These results suggest that planning for the future may also be found in nonhuman species, especially wide-ranging and highly encephalized ones. Indeed, several recent experiments and observations have shown that some animals, including great apes, possess the ability to plan for future needs [1], [2], [8], [17], [18]. A zoo chimpanzee was observed to cache stones and pieces of concrete before the zoo opened for later use as ammunition on visitors [18], and to even hide them from view to avoid detection [19]. In experiments, chimpanzees and orangutans chose the correct tool to get a reward one hour later [8]. Bonobos and orangutans selected, transported, and stored appropriate tools to use them up to 14 hours later [1]. Western scrub-jays stored food in a place where they had learned they would have access to it the next morning, and when provided with two different food types would store more of one type in a room where they knew only the other food type would be available the next day [2]. They and Eurasian jays also anticipated their own future needs independent of their current motivational state by caching those food items they know they would need by the time they would be able to recover them, rather than the ones they preferred at the time of caching [17], [20]. These results have cast doubt on the hypothesis that the ability to plan for future needs is uniquely human. This ability may be present at least in some (relatively large-brained) corvids and great apes, but perhaps also in rodents [21].

Although experimental evidence is both easier to obtain and meets more stringent standards in captivity, it is important to examine the occurrence of planning for the future in wild animals. It is possible that the presence of such a peculiar mental ability is a mere byproduct of selection on other domain-specific or domain-general cognitive abilities [22]. Thus, to understand the evolution of cognition, we must examine the use, and if possible adaptive significance, of cognitive skills in natural conditions. Although planning in the context of range use is certainly not the only adaptive higher cognitive skill in animals, it may be much more widespread than others such as tool use or intentional communication. Therefore, evidence for long-term planning in one wild-living species would be important as a proof-of-principle that high-level cognitive abilities can indeed be adaptive, rather than being mere spandrels, and stimulate future research in other species.

To investigate whether there is evidence for long-term planning in wild animals, we best start looking in those species that have already demonstrated planning abilities in experimental conditions. Non-human great apes are large and partly or wholly arboreal, range semi-nomadically in large home ranges, and often live in dispersed social systems. Therefore, they face high travel costs to approach their highly versatile targets, and thus meet the kinds of conditions where planning for future social and subsistence needs may be adaptive. However, despite decades of intense fieldwork on great apes, it remains unknown whether they actually use this ability to plan for future needs in natural conditions. A major obstacle is that it is very difficult to rule out the influence of environmental cues without cumbersome field experiments. For instance, chimpanzees carry scarce tools to nut-cracking sites that are out of view [23]. This looks like planning for future needs, but because they do so when traveling toward the nut trees, the animals may already be motivated to crack nuts, allowing for the possibility that associative learning (tools and nuts go together) established the habit. This problem may largely disappear when animals communicate their ever-changing travel patterns to others hours before they execute them.

The present study capitalized on this situation by examining the extent to which the direction of long calls emitted by male Sumatran orangutans (Pongo abelii) indicated the direction of their future travel. These orangutans live in dense tropical forests and are semi-solitary, and thus often out of visual contact from others in their population [24]. Sexually mature males may, after highly variable periods of time [25], grow cheek flanges (wide cartilaginous pads at the sides of their face [26]). The average flanged male at our study site (Suaq Balimbing) travels about 1000 m per day in a home range of at least 2,000 ha [27]. Flanged males emit loud vocalizations, audible for over 1 km distance, known as long calls, on average about four times per day [28]. Previous work has shown that long calls reveal the caller's identity [28], [29] and that spontaneously produced long calls differ acoustically from long calls elicited by the presence or activities of other flanged males [29]. Long calls repel lower-ranking males, while attracting higher-ranking males [30], [31]. However, they also attract sexually active females and are often used by females with dependent offspring to remain within earshot of the dominant male, probably because this reduces harassment by other males [31], [32].

Calling males face in one particular direction during the entire long call. Because only fully flanged males emit long calls, the function of the flanges may be to concentrate the sound energy in the direction of the call. As a result, the perceived distance of the calling male to a listening female will be affected by the angular difference between the direction of the call and the line connecting caller and listener (Fig. 1). Thus, a long call perceived as faint by a listening female (say, female A in Fig. 1) could mean that the male is far away, or alternatively, less far but calling in a direction away from the female. In both cases, a female must approach the call to reach him or remain within earshot with the least travel effort, but only if he consistently travels in the direction he calls. Given the practice of earshot association [31] and because sexually active females sometimes go to calling males [32], we therefore expected that flanged males, if they have a target location or area in mind, give spontaneous long calls in the direction they are later travelling.

PPT PowerPoint slide

PowerPoint slide PNG larger image

larger image TIFF original image Download: Figure 1. Orangutan males give long calls to attract females and repel rival males. To reach the calling male or remain within earshot of him, female orangutans should move in the direction of the flanged male if they hear his long call only faintly. Due to the large cheek flanges that probably act as a megaphone, to female A the call is not as loud as to female B, even if they are at the same distance from the calling male, and A should move in the direction of the male whereas B need not. However, this system works best if the male actually travels in the same direction as he is calling. https://doi.org/10.1371/journal.pone.0074896.g001

Such consistent travel in one direction is unlikely to be an artifact of the direction of movement during calling. Sumatran orangutan males are strictly arboreal, and usually give their long calls when stationary, e.g. during feeding, sitting, or resting on a nest. They face consistently in one direction during the entire call, which lasts about 80 seconds on average, but may last up to 4 minutes [28]. It could be argued that the male initially travels simply in the direction he was facing. However, during a single feeding bout in a tree, the male faces in many different directions. Moreover, when he calls during rest, it would be extremely unlikely that he can maintain travel in the direction he faced during the call for more than a few minutes because he cannot travel in a straight line from tree to tree. Instead, the male must look for branches that can hold his weight, creating a meandering route, albeit one with an overall direction. Hence, if males actually continue to travel in the direction of the long call for a long period of time despite interruptions, such as overnight rest, and do so despite the fact that their target location or direction changes from day to day, this would indicate that they plan their travel routes in advance [1].

Here, we first test whether the direction in which flanged male Sumatran orangutans give spontaneous long calls generally predicts the subsequent travel direction. Second, we investigate whether a new spontaneous long call indicates the subsequent travel direction better than the old one would have, if no new call had been given. Third, we test the extent to which long calls given in the evening at or near the night nest still indicate travel direction during the next day, thus indicating future planning independent of the current motivational state.

In addition, we examine whether the audience actually uses the information contained in the calling direction by adjusting their travel routes. Because animals are not active during the night, and thus cannot be distracted by other cues or social activity, we selected the response to long calls heard the evening before. If during the following morning, they adjusted their travel direction, and did so, before a new long call could affect it, this would strengthen the adaptive interpretation of the announcement of travel plans by the flanged males.