Research at the Köhler Center (Pongoland) is strictly observational; there is no invasive research of any kind. Apes are observed as they interact with their environments, humans or groupmates, and also as they attempt to solve problems that are presented to them.
Apes are never deprived of their regular food or water in problem-solving studies. Instead, they work for special foods that they normally receive only rarely.
We take a broad approach to social cognition by investigating a number of socio-cognitive abilities. We have focused our work on four main areas: theory of mind, gestural communication, cooperation and prosociality and social learning.
Theory of mind
Over the past years, our work has now established - with a variety of experiments using a variety of methodologies - that great apes posses key aspects of a "theory of mind". Most of this work has concentrated on apes' understanding of the psychological states of perception and attention, and we have produced several publications demonstrating apes' considerable skills in this domain. Our latest work has focused on apes' understanding of knowledge in others. Using a novel methodology, we have found the strongest evidence to date that apes understand that others know “things” (i.e. have seen them in the immediate past). But importantly, in this exact same experimental paradigm the apes have shown no signs of understanding beliefs - even though this particular research paradigm involved “competition” with conspecifics, which we have some reason to believe to be the paradigm that brings out apes' best performance. Overall, our group has now established experimentally that great apes operate with the basics of a perception-goal psychology in their social interactions with others. But there is no evidence - even though we have looked with similar methods - that they operate with a more human-like belief-desire psychology. The latter may be connected to humans' more sophisticated skills of shared intentionality.
We have now completed a long series of observational studies on the gestural communication of apes, culminating in the publication of an edited volume in which we gathered together and synthesized our work in this area from the last two decades. This volume links aspects of the above-mentioned work on theory of mind with work on the evolution of intentional communication and language. In fact, one of the main motivations for this volume was to call attention to the pivotal role that gestural communication has played in the evolution of language. We have also combined the data from observational studies with experiments. Thus, we have investigated the ability of apes to enlist the help of human experimenters to retrieve a hidden reward (for the ape) or a hidden tool required to retrieve the reward. Orangutans and bonobos can easily do this. However, they are less forthcoming when the hidden item is not something that they desire, but something that the experimenter desires. In this sense, apes can use pointing imperatively but not simply to inform others. We have also shown experimentally that apes can place themselves in a position where a human experimenter can see them in order to beg for food. Chimpanzees even leave food behind in order to seek visual contact with the human to gesture to her face.
Cooperation and prosociality
We have argued that cooperation and prosociality are two key aspects of human nature. It is hard to understand human life without cooperation between large numbers of unrelated individuals (often strangers). The comparison with our closest relatives is fundamental to gain insights into the evolution of cooperative behavior in humans. Several studies have shown that chimpanzees can cooperate with each other as well as with humans in order to solve technical problems. More interestingly, chimpanzees can select the best cooperators and can also help third parties even if doing so does not produce a direct benefit to them. However, chimpanzees appear less inclined to help when it comes to food distribution. Thus, they will not incur a negligible cost (e.g., pull a rope) to give food to another individual if they do not benefit as well. Interestingly, they will not act spitefully either when given the opportunity. These results are confirmed by our recent administration of the so-called ultimatum game to pairs of chimpanzees (in which a proposer can make an offer of some percentage of an endowment to another individual, who can in turn accept it or reject it so that no one gets anything). Unlike humans, who reject offers of less than 30% of the endowment, chimpanzees accept any non-zero offers. This is precisely what traditional economic theory predicted – maximizing gains while reducing costs. Unlike chimpanzees, humans are instead willing to pay a cost to punish unfair offers. This is not to say that chimpanzees are insensitive to perceived losses. Another of our studies also established that chimpanzees will retaliate against individuals that have stolen their food.
We have completed several studies on social learning abilities in great apes. These studies have confirmed the hypothesis that great apes are not inclined to copy the precise body movements of others if they are not specially trained or raised to do so. One of our latest research lines involves testing the conditions that foster the emergence and maintenance of certain tool-using behaviors observed in wild populations such as ant-dipping in chimpanzees and nettle processing in gorillas. The reasoning we apply here is that if the behaviour occurs spontaneously and without the necessity for actually seeing the behaviour performed by others then it will be a natural result of the interaction between the individual and its environment. This evidence questions the long-held belief that social learning plays a pivotal role in the development of the precise technique used by individuals to exploit certain resources. For example, we have found that a complex way of eating nettles develops spontaneously in gorillas, without the need for so-called program-level imitation, as has been claimed before.
Our broad approach to social cognition is matched with a corresponding broad approach to physical cognition. Thus we study various aspects of physical cognition that can inform us about the cognitive processes that the great apes use to solve problems. We have focused our attention on three main areas: spatial cognition, planning, and causal knowledge.
Spatial cognition and memory
We have investigated spatial frames of reference from a phylogenetic, ontogenetic and cross-cultural perspective. We have found that nonhuman apes and 1-year-old children focus their attention on spatial information (e.g., the container located on the right side) whereas 3-year-old children pay more attention to the container features (e.g., the round and brown container). This has led us to hypothesize that at least in the great ape family, spatial coding is more ancient than featural coding. Further studies suggest that culture, and language in particular, plays an important role in the development of this in human ontogeny. We have continued to work on how apes encode and retrieve spatial information from memory. Previously we had established that orangutans can remember what is where. However, this has never been systematically quantified and it is unknown how much exposure is necessary to acquire such knowledge. We investigated the speed of acquisition and the long-term retention of food locations in chimpanzees by hiding rewards in their indoor area and measuring whether they revisited those locations once they had discovered the reward on a previous day. We found that chimpanzees were capable of remembering food locations after a three-month period.
If memory can allow subjects to store useful knowledge, planning is an essential component to put that knowledge to good use, particularly in novel situations. We have found that orangutans and chimpanzees can spit water inside a tube to make an out-of-reach peanut located at the bottom of the tube float into reach. The solution to this task had all the hallmarks of insightful problem solving and it represents the first case in apes of using water as a tool. We also discovered orangutans and gorillas selecting and transporting tools to various locations to solve particular problems, and systematically selecting the appropriate tool based on a mental template of the features that make a tool suitable. We have also investigated whether apes can perform an action in the present not for its present consequences, but for the consequences that it will have in the future. We found that bonobos and orangutans can select, transport and save a tool to use it the next day. This means that these species have some ability to plan for future needs. We would now like to know if apes will make a tool that they not currently need and save it for a future occasion (something that has been hypothesized to play a key role in human evolution). We are also investigating whether subjects can keep track of the time that has elapsed since the occurrence of certain events – an aspect that is related to mental time travel and future planning. Our current results strongly suggest that chimpanzees can keep track of how long it has been since a reward was hidden under one of three cups.
Previously we established that subjects understand many of the relations between objects and their effects in the environment. We have now turned our attention to causal understanding in tool-using tasks. Preliminary results suggest the hypothesis that apes treat object configurations that enable causal connections differently from objects configurations that only enable arbitrary connections. In other words, apes seem to have evolved cognitive mechanisms that help them to take advantage of the richness of meaningful object-object relations in their environment, and not just associations. Our work on several versions of the famous trap-tube task has shown that apes are more proficient in this task than previously thought. Thus, subjects’ performance dramatically improves if subjects are allowed to 1) rake out the reward (as opposed to push it out), 2) use their finger (as opposed to a tool) to move the reward, or 3) insert the tool (rather than having to choose between two pre-positioned tools). This suggests that certain task features may have masked the causal knowledge that subjects may possess about the task.