Science suggests that chimpanzees understand death and experience grief like humans. The primates have been observed carrying their dead infants for days, making whimpering calls when an individual dies and refusing to make a nest where one of their group has passed away.
Do chimpanzees experience grief, do crows seek joy and do elephants show affection? Pythagoreans long ago believed that animals possess the same range of emotions as humans; but in more recent times, many thought animals lacked that capacity and essentially viewed them as mere creatures without any internal emotional states, justifying their use for human purposes without considering their potential suffering. So, how do today’s animal behaviorists feel about the feelings of animals? A new survey helps to answer that question.
Understanding nonhuman animal emotions has also been a long-running question at the forefront of animal welfare studies, and new scientific findings may hold the key to decoding the chatter. Integral to those discoveries were two baby chickens and a mirror.
A similar groundbreaking study takes a different tack: it eschews chickens and mirrors for artificial intelligence (AI). Machine learning has just deciphered emotions in seven ungulate species. Could this be the game changer for improved animal welfare? Can artificial intelligence truly help us understand what animals feel? A pioneering report suggests the answer is yes.
Crows are known to enjoy playing with each other, exhibiting behaviors such as chasing, mock fighting and even creating “toys” from random objects, demonstrating their high intelligence and social natures.
Scientific thought on emotions in animals
Since ancient times, philosophers—such as Pythagoras—have pondered the seemingly simple question of whether animals experience emotions. Aristotle believed that animals and humans share similar emotions, while Descartes argued that animals were more like machines, lacking the capacity for consciousness or emotions.
In the mid-1800s, famed naturalist Charles Darwin wrote that “the lower animals, like man, manifestly feel pleasure and pain, happiness and misery.” By the mid-20th century, however, leading behavioral theorists denigrated the idea of studying animal emotions since even if they did exist, the scientists were convinced that they couldn’t be scientifically measured and verified.
The late primatologist Frans de Waal, an Atlanta, Georgia, Emory University emeritus professor of psychology, helped change this dynamic through his innovative studies of animal cognition. From the time of de Waal’s 1982 book Chimpanzee Politics: Power and Sex Among Apes to his 2019 tome Mama’s Last Hug: Animal Emotions and What they Tell Us about Ourselves, attitudes about whether animals have thoughts worthy of scientific exploration changed dramatically. De Waal gave a new generation of scientists the permission to ask questions about the inner lives of the nonhuman beings among us. His popular, best-selling books also shaped the public’s perception of animal minds.
Social animals that form close bonds with each other, elephants are emotional, loving and sensitive. They can recognize each other even after long periods of separation. The pachyderms show affection through gentle touch, trunk twining and other behaviors.
As the field has grown, researchers at Emory University—and colleagues at New York’s Cornell University and North Carolina’s Duke University—wanted to quantify animal behaviorists’ perceptions of the taxonomic distribution of animal emotionality. They developed a survey of multiple-choice questions, free-form text fields and rating scales and sent it to those in prominent graduate school programs in animal behavior research. They also posted solicitations for the survey, aimed at researchers in the field, on social media.
The 100 survey respondents spanned a range of specialties, including behavioral ecologists, biological anthropologists, biological psychologists, cognitive psychologists, evolutionary biologists and neuroscientists. They comprised graduate students (45), faculty (28), postdoctoral fellows (20), retired faculty (2), other PhD researchers (3) and undergraduate students (2).
The most common taxa of animals studied among the respondents were birds (43%), nonhuman primates (32%) and other mammals, though each of the taxa that the survey asked respondents to assess were studied by at least some members of the sample.
The belief that nonhuman animals have no feelings was prevalent for a long time, stemming from a homocentric view that considered humans as the only beings capable of complex emotions. But in a recent survey, 78% of animal behaviorists ascribed emotions to birds.
The results, published in the journal Royal Society Open Science in November 2024, showed that a majority of the survey respondents ascribed emotions to “most” or “all or nearly all” nonhuman primates (98%), other mammals (89%), birds (78%), octopuses, squids and cuttlefish (72%) and fish (53%). And most of the respondents attributed emotions to at least some members of each taxonomic group of animals considered, including insects (67%) and other invertebrates (71%).
Regarding animal consciousness, the survey defined it in its most basic form, meaning that animals are aware of their own existence. Most of the respondents credited consciousness to a broad, taxonomic expanse of animals, although at slightly lower majorities as compared to emotions.
This is the first known assessment of how animal behaviorists across a range of disciplines think about consciousness and emotions in nonhuman animals. It provides a snapshot in time so that 20 years from now, we can revisit how scientific experts may have changed their views.
Octopuses can feel fear, pain and stress—and they try to avoid those emotions. Such findings have led to calls to recognize octopuses as sentient beings and to establish welfare regulations for them.
The survey also included questions about the risks in animal behavioral research of anthropomorphism (inaccurately projecting human characteristics onto animals) and anthropodenial (willful blindness to any humanlike characteristics of animals). A surprising 89% of the respondents thought that anthropodenial was problematic in animal behavioral research, compared to only 49% who thought anthropomorphism poses a risk. This represents a big shift, as anthropomorphism has long been a leading argument against those who attributed feelings to animals.
Near the end of the survey, respondents were asked to define the term emotion. A little more than half of their definitions referred to emotions as a response to either external or internal stimuli. A majority also referred to emotions as being subjective experiences or related to consciousness or mindedness. And 40% of the responses identified emotions as functioning to motivate behaviors. Only 81 out of the 100 survey respondents provided a definition, perhaps due to the challenge of verbalizing a working description.
Even in human studies, it is challenging to determine which biological markers to measure and how to adequately describe and quantify something as complex and variable as emotions. They may include everything from instinctual reactions of disgust or fear to deep feelings of affection and empathy for others.
There is mounting evidence that insects experience a range of feelings, including depression, fear and joy. For example, bumblebees feel positive emotions when they find unexpected rewards, like a sugar solution.
Animal studies are further complicated by the fact that researchers can’t ask an animal how it’s feeling. And while experiments with animals in labs can be tightly controlled, the results may be skewed since those animals are not interacting within their natural environments. Animal behavior experiments in the wild provide valid ecological and social contexts, but they are challenging to control and to design.
The researchers conclude that they’ve only scratched the surface of exploring what nonhuman animals are capable of experiencing. Knowledge gained about the interior lives of other beings might even help us better understand our own ancestry. In what ways are humans a unique species? Understanding the evolution of emotions is essential for discovering answers to that question.
Distressed chick calls on being alone
To learn more about emotions in nonhuman creatures, animal welfare researchers in the United Kingdom recently looked at the common domestic chick’s peeps and whistles. Their results were published in the journal Applied Animal Behaviour Science in August 2024.
A loud chick is a distressed chick. Researchers in the United Kingdom recently used the calls and vocalizations of the birds as a noninvasive way to measure stress states.
Wondering whether there was a noninvasive way to measure stress states in freely moving animals, the UK researchers decided to look at the calling and vocalizations of chickens. They used an experimental setup that involved complex acoustic equipment in a secluded room. In one box, they placed a chick; and in another, they put a chick with a mirror.
Farmers and researchers have long known that a loud chick is a distressed chick. In this experiment, the chick that could see its reflection believed it was not alone, and its calls sounded relatively calm, meaning that the chick wasn’t too stressed. The chick that was alone, however, began to voice loud, higher pitched calls, indicating an anxiety-like state. The acoustic data the researchers gathered measured how the quality of the chick’s vocalizations changed and what that meant for their stress levels.
In previous models, the way to measure an animal’s stress included capturing it, withdrawing blood and measuring levels of corticosterone, a stress hormone. Capture, restraint, collection and release are, in themselves, stressful. This study presented a more sophisticated, noninvasive, comparatively inexpensive way to study anxiety-like states in animals.
Using acoustics to monitor domestic animal welfare could be worthwhile. Being able to detect stress could open the doorway to more humane, responsive practices in the egg and meat industries.
Medications intended for humans often must go through chick, primate and rat studies before any testing on human subjects. But since chicks are often used as an early precursor to human studies, understanding their emotions could improve studies on anxiety and depression medications for humans.
But if researchers cannot prove that a chick first has anxiety symptoms, they cannot show that a medication improves that state. To claim that an antianxiety drug is working, a chick must be shown to have anxiety akin to what we experience, meaning that animals must have emotions.
The research, though, also points toward an answer to an older, more complex question: what rights do animals have? The long-held answer is that animals do not have the same emotional capabilities as humans and, therefore, have fewer needs and wants. Using acoustics to monitor animal welfare, however, shows the similarities between bird emotions and human anxiety and depressive states. That makes a very strong argument for nonhuman animals having negative emotions. And if that’s true, then ethically it follows that we are absolutely obligated to be worried about animal welfare—particularly in commercial production facilities in the egg and meat industries—and provide the best living conditions possible for the more-than-humans that live alongside us.
Denmark scientists successfully trained an AI model to distinguish between positive and negative emotions in various ungulate species, including wild boars. By analyzing the acoustic patterns of the animals’ vocalizations, the model achieved an impressive accuracy of 89.49%.
AI on animals’ emotional language
It’s an exciting time, say some researchers, as new methods are currently being developed that may help us better understand how an animal may be feeling. And AI is one of the newest strategies being employed.
Recently, researchers from the Department of Biology at Denmark’s University of Copenhagen successfully trained a machine-learning model to distinguish between positive and negative emotions in seven different ungulate species, including cows, pigs and wild boars. By analyzing the acoustic patterns of their vocalizations, the model achieved an impressive accuracy of 89.49%, marking the first cross-species study to detect emotions using AI.
By providing solid evidence that AI can decode emotions across multiple species based on vocal patterns, this breakthrough has the potential to revolutionize animal welfare, conservation and livestock management, allowing us to monitor animals’ emotions in real time.
The AI-powered classification model developed in Denmark could be used to develop automated tools for real-time monitoring of animal emotions, transforming the way we approach veterinary care.
The Denmark researchers analyzed thousands of vocalizations from ungulates in different emotional states and identified their key acoustic indicators. The most important predictors of whether an emotion was positive or negative included changes in amplitude modulation, duration, energy distribution and fundamental frequency. Remarkably, these patterns were somewhat consistent across species, suggesting that vocal expressions of emotions have an evolutionary component.
The study’s findings, published in the journal iScience in February 2025, have far-reaching implications. The AI-powered classification model could be used to develop automated tools for real-time monitoring of animal emotions, transforming the way we approach conservation efforts, livestock management and veterinary care. They also offer insights into the origins of human language and could reshape our understanding of animal emotions.
To support further studies and accelerate research into how AI can help us better understand animals and improve their welfare, the researchers have made their database of labeled emotional calls from the seven ungulate species publicly available.
Whales have cultures and emotions. They have complex societies with dialects and traditions that are passed down through generations. They display a range of feelings, including empathy, grief and joy.
Growing evidence on more-than-human sentience
Understanding how nonhuman animals express their emotions can help us improve their well-being. If we can detect discomfort or stress early, we can intervene before it escalates. Equally important, we could also promote positive emotions in them.
It seems hard to fathom now, but until recently, we thought we were the only creatures to have emotions. Thankfully, more and more studies are supplying evidence of altruism in nonhuman animals, beliefs in more-than-humans and sentience in the “others.”
Here’s to finding your true places and natural habitats,
Candy
