Great apes and humans perceive events in a similar way
This is an agents’ world. Through a transspecies study, researchers from the University of Neuchâtel and NCCR Evolving Language have discovered that great apes could have a similar perception of events as humans. Just like us, our close cousins tend to focus more on the agent of an action (i.e. who is performing the action), especially if it is animate. This hints at an ancestral origin of syntax features that are universal in human languages, according to Sarah Brocard and her team.
Chimpanzees working with the touch-screen device. © Zoo Basel
“The cat eats the mouse” or “The mouse is eaten by the cat”. We humans seem to have a (almost) universal preference for agents in unfolding events, as shown by our sentence structure. This means we tend to assume the first being mentioned will be the one performing the upcoming action. And when we are proven wrong by the rest of the sentence, our brain provides extra efforts to reevaluate the information. Is this feature unique to humans, or can we also find it in our closest living relatives, the great apes? According to researchers, this agent preference may have appeared before our species branched out.
More interest in the agent
Researchers from the University of Neuchâtel and NCCR Evolving Language have discovered that great apes could have a similar perception of events as humans. A transspecies experiment conducted at the Basel Zoo showed that our close cousins tend to focus more on the agent of an action (i.e. who is performing the action), especially if it is animate.
In the experiment, the scientists showed short videos of agents acting on patients to all species. Different conditions were tested, from animate agents acting on inanimate patients (i.e. a gorilla drumming a bucket), to inanimate agents acting on animate patients (i.e. a ball hitting the face of a human). At the end of each video, the participants could touch the screen wherever they liked. As a result, a significant part of them touched the agent of the event, rather than the patient.
However, and to the surprise of the researchers, the agent preference was not as prominent when it was animals acting on animals. In this condition, the preference was also influenced by the type of interactions. “This finding suggests that the agent preference, which is a really strong cognitive bias in humans, can still be influenced and modulated,” comments Sarah Brocard, first author of the study and PhD candidate at the University of Neuchâtel. “And in terms of syntax evolution, I think it’s really interesting,” she adds.
A one-of-a-kind transspecies study
In this study, the researchers worked with many hominids (or great apes), including human children and adults, chimpanzees, gorillas, and orangutans. It was capital for them to ensure that all the groups were tested with similar settings, so that the results could be comparable between species. For this reason, they chose to work with a touchscreen device, which would play the videos and then allow participants to select their region of interest on the screen.
This approach came with challenges. “Initially, I thought that the training of the apes would be the most challenging part because, at the time, they were naive to video watching on the touchscreen and I was not entirely sure on how they would react,” Brocard remembers. “But this turned out to be the easiest part, as the apes were all fascinated by the videos.” Instead, humans were the most difficult to test she says: “I aimed to conduct my tests in the ape house of the Basel Zoo, right in front of the chimpanzees’ enclosure, while the apes were also participating in a similar experiment. However, it proved really difficult to recruit and test participants in this environment due to noise, the many distractions, and visitors simply wanted to continue their visit.”
In the end, the researchers could get results from 20 adults and 50 children, and the experiment provided them with a great way to share and explain their research to the public. “Despite these challenges, I’m glad I tried it as it offered a unique engagement with the public,” Sarah Brocard adds.
Event cognition and the evolution of human language
According to the event cognition theory, the ability to decompose events into their different parts (agent, patient, and action) serves as a foundation for syntax. “Broadly speaking, syntax connects the elements of an event in a systematic way, but before being able to do so, the brain needs to be able to discriminate and understand the relation linking these entities,” Sarah Brocard explains. So, a first step in having syntax in one’s communication system is the ability to understand the unfolding of events.
This study highlights a striking similarity between the way we and our close cousins process events and make decisions. In the majority of human languages, the first being mentioned in a sentence is usually assumed to be the agent of the action that will be performed. The researchers’ results hint that this way of thinking is not a consequence of how we speak, but rather the other way around: our ancestors could have had a preference for agents long before language existed. “Now that we discovered that great apes have a similar perception of events as humans do, we need to understand what prevents them from communicating in terms of “who does what to whom”,” Brocard comments.
The mystery still remains
The researchers are overjoyed with the potential impact their work could have on event cognition theory, especially from a communication perspective. But many questions remain : “Why do great apes and humans differ in how they communicate about these events? I find this question incredibly fascinating, and I really hope that our research will bring us closer to an answer,” Brocard says. The group also conducted a similar study with an eye-tracker, which gave quite different results. “We are now in the process of developing a new setup, combining eye-tracking and touchscreen, to help us understand how gaze and decision-making relate to one another,” Brocard explains. “It would also be nice to test if such a preference is present as well in other groups of non-primate animals,” she adds.
Reference
24. A universal preference for animate agents in hominids. iScience, Volume 27 Issue 6. https://doi.org/10.1016/j.isci.2024.109996
