A large and growing body of research supports the conclusion that many animals are sentient, meaning they have the capacity to experience feelings like pain, pleasure, and emotional states. The evidence is strongest for mammals and birds, but in recent years researchers have built compelling cases for fish, octopuses, crabs, and even some insects. What makes the question scientifically interesting is not whether your dog feels something when you step on its tail, but how far sentience extends across the animal kingdom and how we can measure something as fundamentally private as another creature’s inner experience.
What Scientists Mean by Sentience
In everyday conversation, “sentient” often gets used loosely to mean smart or aware. In the scientific literature, it has a more specific meaning: the capacity to have feelings, particularly experiences with a positive or negative quality, like pain or pleasure. One influential framework distinguishes sentience from mere consciousness (being awake and responsive to stimuli) and from awareness (a higher-order state involving concepts of self and environment). Under this framework, an animal can be conscious without being sentient. A sea slug that recoils from a hot surface is conscious in the sense that it perceives and responds, but it may not have the internal experience of suffering.
1Journal of Consciousness Studies. Concepts and Interrelationships of Awareness, Consciousness, Sentience, and WelfareThis distinction matters because the real scientific debate is not about whether animals detect harmful stimuli. Almost every animal with a nervous system does that. The debate is about whether detecting those stimuli produces something that feels like something to the animal, an experience with emotional weight that influences future decisions. Researchers call this the difference between nociception (an automatic detection-and-response system for harmful stimuli) and pain (a subjective experience that motivates flexible, context-dependent behavior).
2Journal of Experimental Biology. From nociception in aneural animals to human suffering: toward a comparative biology of painYou Do Not Need a Big Brain to Feel
For decades, many scientists assumed that sentience required a large cerebral cortex, the wrinkled outer layer of the mammalian brain responsible for complex thought. Some even argued that pain perception was impossible without it. That view has eroded substantially. Research now shows that basic emotions are generated in deeper, evolutionarily older brain structures found across a wide range of species, not just in the cortex. Non-mammalian animals that lack a cortex entirely, like birds and fish, have been shown to possess complex cognitive abilities and emotional responses.
3PubMed Central. Animal Sentience: Where are We and Where are We Heading?This makes evolutionary sense. The basic neural machinery supporting consciousness in vertebrates is ancient, appearing at the earliest points of vertebrate brain evolution. Studies using general anesthesia across species reveal that the neurophysiological mechanisms that switch consciousness on and off are highly conserved, meaning they are built from the same fundamental components in fish, amphibians, reptiles, birds, and mammals.
4PubMed Central. Evolution of consciousness: phylogeny, ontogeny, and emergence from general anesthesiaThe practical implication is that drawing a line at the cortex, or at brain size more generally, no longer holds up. A crow with a brain the size of a walnut can solve multi-step puzzles, recognize individual human faces, and apparently hold grudges. A zebrafish with a brain barely visible to the naked eye can form memories of painful and pleasant events and use environmental cues to retrieve those memories later.
5Applied Animal Behaviour Science. Use of conditioned place preference/avoidance tests to assess affective states in fishHow Researchers Test for Feelings They Cannot See
The central challenge of sentience research is that you cannot ask a fish or a crab how it feels. Scientists have developed several behavioral tests that get at the question indirectly, each designed to distinguish genuine feeling from automatic reflexes.
Conditioned Place Preference and Avoidance
One of the most powerful tools is the conditioned place preference (CPP) test. The logic is straightforward: if an animal remembers a location where something unpleasant happened and actively avoids returning to it, that avoidance reflects more than a simple reflex. It requires the animal to form a memory of the experience, associate it with environmental cues, and use that information to guide a later choice. Similarly, if an animal is in ongoing pain and receives a painkiller in a particular location, it will later prefer that location, but only if it was actually hurting. A painkiller given to a pain-free animal produces no preference.
These tests have been validated across fish, chickens, rodents, and octopuses. In fish, appetitive stimuli increased the time spent in the associated location, while aversive stimuli decreased it, with stress hormone levels rising in tandem, confirming the emotional dimension.
5Applied Animal Behaviour Science. Use of conditioned place preference/avoidance tests to assess affective states in fish In chickens, researchers have extended the paradigm to measure not just whether an emotional response exists, but its intensity, how long it lasts, and whether it generalizes to similar situations.
6Applied Animal Behaviour Science. Examining affective structure in chickens: valence, intensity, persistence and generalization measured using a Conditioned Place Preference TestCognitive Bias Tests
Another approach borrows from human psychology: the judgment bias test. Animals are trained to learn that one cue (say, a low-pitched tone) predicts a reward, while a different cue (a high-pitched tone) predicts something unpleasant or less rewarding. Then they are presented with an ambiguous, in-between cue. Animals housed in poor conditions or recently exposed to stressful events tend to treat the ambiguous cue as if it predicts the bad outcome, a “pessimistic” interpretation. Animals in enriched, comfortable environments are more likely to interpret it optimistically.
7Applied Animal Behaviour Science. Cognitive bias as an indicator of animal emotion and welfare: Emerging evidence and underlying mechanismsThis maps onto a well-known pattern in humans: depressed or anxious people tend to interpret ambiguous situations negatively, while people in good emotional states give them the benefit of the doubt. The fact that animals show the same pattern, with their bias shifting predictably based on welfare conditions, is strong circumstantial evidence for underlying emotional states. Cognitive bias testing has been applied across many species and contexts, and the results consistently link an animal’s bias to its welfare state.
8PubMed Central. Cognitive Bias in Zoo Animals: An Optimistic Outlook for Welfare AssessmentThe Octopus Problem
If you want a case study in how far sentience research has come, look at cephalopods. Octopuses, cuttlefish, and squid are mollusks, more closely related to snails than to any vertebrate. Their nervous systems evolved independently from ours for over 500 million years. Yet the evidence for their sentience is now considered strong.
A landmark study injected dilute acetic acid under the skin of an octopus arm, essentially giving it a localized chemical burn. Octopuses that received the injection clearly avoided the chamber where the pain had been delivered. More telling, when they were given a local anesthetic (lidocaine) in a different chamber, they developed a strong preference for that chamber, but only if they had been injected with the acid first. Octopuses that received only saline showed no preference for the lidocaine chamber, demonstrating that the painkiller was only rewarding in the presence of actual pain.
9PubMed Central. Behavioral and neurophysiological evidence suggests affective pain experience in octopusAn updated assessment evaluating the evidence across all cephalopod groups found strong evidence of sentience in octopuses and cuttlefish, with very high or high confidence in meeting six of eight criteria used to evaluate sentience. Squid met five of eight criteria with similar confidence. Even nautiluses, the most primitive living cephalopods, showed some suggestive evidence.
10PubMed. Sentience in cephalopod molluscs: an updated assessmentCrabs, Lobsters, and the Trade-off Test
Decapod crustaceans, the group that includes crabs, lobsters, crayfish, and shrimp, have been the subject of increasingly sophisticated pain research. A key approach is the trade-off test: if an animal is responding purely by reflex, its response to a harmful stimulus should be the same regardless of what else is going on. But if an animal is making a decision, it should weigh competing needs against each other.
Hermit crabs provide a clean example. When given small electric shocks inside their shells, they sometimes evacuate. But the likelihood of leaving depends on context. Crabs are less willing to leave a good-quality shell than a poor one, and less willing to leave when the odor of a predator is present, even when the shock intensity is the same. They are trading off pain avoidance against predator avoidance, weighing competing motivations in a way that reflexes do not.
11PubMed. Trade-offs between predator avoidance and electric shock avoidance in hermit crabs demonstrate a non-reflexive response to noxious stimuli consistent with prediction of painThe researchers behind this work are careful to note that these results are “consistent with predictions of pain” rather than definitive proof of it. That caution is warranted, but the cumulative picture across multiple crustacean studies has been persuasive enough to change policy. The United Kingdom now officially recognizes decapod crustaceans as sentient beings under the Animal Welfare (Sentience) Act, with implications for how they can be slaughtered, kept in captivity, and used in research.
12Animal Sentience. Decapods as food, companions and research animals: Legal impact of ascribing sentienceCan Insects Feel Pain?
This is where the evidence gets thinner and the debate gets hotter. Insects display defensive behaviors in response to harmful stimuli, and those behaviors can be modulated by different stimuli and contexts, which goes beyond what you would expect from a simple reflex arc.
13Advances in Insect Physiology. Can insects feel pain? A review of the neural and behavioural evidenceRecent work has built a case that insects have descending controls for nociception, meaning their brains can dial pain responses up or down depending on circumstances. Neuroanatomical and molecular evidence shows that the insect brain can facilitate or suppress defensive behavior, and that this modulation involves the central nervous system rather than happening entirely at the periphery.
14PubMed Central. Descending control of nociception in insects?Honeybees are a focal point in this debate. Research on their cognitive, emotional, and sensory complexity has grown rapidly, and some scientists now advocate for including them in animal welfare legislation. The European Union currently protects mammals, birds, and cephalopods as sentient beings, but honeybees remain excluded despite what some researchers call robust evidence of their capacities.
15PubMed Central. Honeybee Sentience: Scientific Evidence and Implications for EU Animal Welfare PolicyThe honest assessment is that while insects show several behavioral hallmarks that in vertebrates are considered indicative of pain, the gap between those behavioral signals and a confident attribution of subjective experience is larger here than it is for mammals or even for octopuses. The nervous system of a bee has roughly a million neurons; an octopus has about 500 million. Whether that quantitative gap maps onto a qualitative difference in experience is an open question.
Not Just Pain: Evidence for Positive Feelings
Sentience research has historically focused on pain and suffering, partly because those states are easier to study and partly because they are most relevant to welfare policy. But the capacity for positive emotions matters too. If animals are sentient, they should be able to experience pleasure, excitement, and something resembling joy.
Researchers have found behavioral markers of positive emotional states across a surprising number of species. Play behavior is one indicator: it is energetically costly, appears to have no immediate survival benefit, and occurs primarily when animals feel safe and well-fed. Laughter-like vocalizations during play have been documented in rats and great apes. In mini-pigs, play was associated with increased tail movement, proposed as a feasible indicator of positive emotions that could be used to assess welfare on farms.
16Applied Animal Behaviour Science. Selection of putative indicators of positive emotions triggered by object and social play in mini-pigsThese positive markers matter for welfare policy because they flip the question from “how do we minimize suffering?” to “how do we enable animals to have lives worth living?” Intensive farming systems that prevent animals from playing, exploring, socializing, and satisfying naturally motivated behaviors are increasingly seen as problematic not just because they cause pain, but because they deny opportunities for positive experiences.
17Animal Research and One Health. Animal sentience: The science and its implications, with particular reference to farmed animalsSelf-Awareness and Knowing What You Do Not Know
Some researchers have pushed beyond basic sentience to ask whether animals have more sophisticated forms of inner life, like self-awareness and metacognition (the ability to monitor your own mental states). The classic test for self-awareness is the mirror self-recognition (MSR) test. Beyond chimpanzees and orangutans, which have long passed this test consistently, reproducible evidence of MSR has been confirmed in bonobos, bottlenose dolphins, and the cleaner wrasse, a small tropical fish. Intriguingly, only social species have reliably demonstrated self-recognition; solitary species studied so far have not.
18PubMed Central. Sociality and self-awareness in animalsMetacognition studies have produced similarly striking results. In one paradigm, rats were trained to classify brief sounds as “short” or “long.” When given the option to skip a trial (declining meant a smaller guaranteed reward instead of the larger reward for a correct answer), rats were more likely to skip the hardest trials, those near the ambiguous middle of the range, exactly as you would expect from an animal that knows when it does not know the answer. When forced to answer, accuracy on those hard trials was lower than when the rats chose to answer voluntarily, confirming they were selectively avoiding their weakest trials.
19Current Biology. Metacognition in the RatGrowing evidence that animals share functional parallels with human conscious metacognition has been found across dolphins, pigeons, monkeys, and apes, though whether these parallels reflect the same kind of inner experience remains an open question.
20Trends in Cognitive Sciences. The study of animal metacognitionEvery Animal Lives in Its Own Sensory World
One of the biggest obstacles to understanding animal sentience is that we tend to assume other animals experience the world the way we do. The concept of the Umwelt, originally from early twentieth-century biology, captures the idea that every species inhabits its own subjective sensory world. A nocturnal mouse navigates a world dominated by odors, sounds, and textures, with vision playing a supporting role. A primate’s world is primarily visual, built from colors and shapes. A bat’s world is constructed from echoes.
21PubMed Central. The Visual Umwelt of primates and Hippocampal Representations of SpaceThis has direct consequences for research. If you test a bird’s cognition using stimuli designed for human perception, you may get misleading results because birds perceive ultraviolet wavelengths that are invisible to us. If you test an animal’s emotional response using conditions that seem neutral to a human experimenter but are stressful to the animal (bright lighting for a nocturnal species, for example), you will draw wrong conclusions.
22Frontiers in Psychology. How to Apply the Concept of Umwelt in the Evolutionary Study of CognitionThe methodological lesson is that sentience research needs to be species-appropriate. Researchers must enter the animal’s world rather than forcing the animal into ours. Some of the most productive recent work has adopted what is called “theromorphism,” an animal-centered approach that attempts to discover how the animal actually perceives and is motivated, rather than projecting human assumptions onto it.
23Frontiers in Psychology. Anti-anthropomorphism and Its LimitsWhat About Plants?
Every few years, a wave of media coverage suggests that plants might be sentient too, citing their ability to respond to touch, communicate chemically with neighbors, or “learn” from experience. The scientific consensus is firmly skeptical. Plants respond to stimuli reactively but have not been shown to perform the proactive, anticipatory behaviors associated with consciousness. Their electrical signaling serves immediate physiological functions rather than the kind of integrative information processing seen in animal nervous systems.
24PubMed Central. Debunking a myth: plant consciousnessEven the most widely cited claim, that plants can undergo a form of Pavlovian conditioning, has been challenged on the grounds that this type of learning does not require consciousness even in animals. A broader philosophical critique holds that the behavioral and physiological similarities between plants and sentient animals that get pointed to in these discussions are not, in themselves, indicative of consciousness. Without a validated test for plant sentience, there is currently no evidence that plants feel anything.
25Biology & Philosophy. A critical review of plant sentience: moving beyond traditional approachesHow Sentience Research Is Changing Law and Policy
The practical stakes of the sentience question are enormous. If a species is recognized as sentient, that recognition can trigger legal obligations regarding how the animal is housed, transported, slaughtered, and used in research. The EU has recognized animal sentience in its treaties since 2009, and its welfare laws protect mammals, birds, and, more recently, cephalopods. The UK’s 2022 Animal Welfare (Sentience) Act explicitly extended recognition to decapod crustaceans and cephalopods, driven directly by the kind of behavioral evidence described above.
12Animal Sentience. Decapods as food, companions and research animals: Legal impact of ascribing sentienceFor farmed animals, the welfare implications are profound. A review of sentience in farmed land and aquatic animals concluded that intensive farming systems deprive animals of opportunities for positive emotions and prevent them from satisfying naturally motivated behaviors. The authors argued that regenerative, agroecological, or organic farming systems better protect and respect animal sentience, leading to less suffering and more opportunities for positive experiences.
17Animal Research and One Health. Animal sentience: The science and its implications, with particular reference to farmed animalsThe frontier of this debate lies with insects. If honeybees and other insects are eventually recognized as sentient under EU law, the implications for agriculture, pesticide use, and the beekeeping industry would be significant. Some researchers have already advocated for a gradual, evidence-based approach, guided by a permanent scientific observatory that would ensure legislation evolves alongside scientific understanding.
15PubMed Central. Honeybee Sentience: Scientific Evidence and Implications for EU Animal Welfare PolicyWhy Sentience May Have Evolved in the First Place
If sentience is so widespread, why did it arise? One hypothesis frames consciousness as an adaptive solution to a specific engineering problem. As animal bodies became more complex during the Cambrian explosion, roughly 540 million years ago, they gained more degrees of freedom in how they could move and respond. A jellyfish has limited options; a shrimp with jointed limbs, sensory antennae, and a segmented body has many. Managing all of those degrees of freedom purely through reflexes becomes computationally unworkable. Subjective experience, in this view, emerged as a way to compress and prioritize all that incoming information, allowing an animal to make fast, flexible decisions rather than relying on an impossibly large library of pre-programmed responses.
26Biological Theory. Complexity and the Evolution of ConsciousnessThis “pathological complexity” hypothesis connects sentience research to evolutionary biology and behavioral ecology in a way that makes predictions: you should find sentience in animals with complex bodies and lifestyles that demand flexible decision-making, and you should be less likely to find it in simpler organisms with fewer behavioral options. Whether this framework will hold up is still being tested, but it offers a concrete evolutionary reason why so many different animal lineages appear to have converged on something like feeling.