Sponges and placozoans are the animals most confidently placed outside the boundary of sentience, because they have no nervous systems at all. Beyond that short list, the scientific picture becomes genuinely murky. Researchers have spent decades trying to draw a line between creatures that merely react to their environment and those that have some inner experience of doing so, and that line keeps shifting as new behavioral and neurological evidence accumulates.
What Scientists Mean by Sentience
Sentience, in the scientific sense, is not intelligence. It is the capacity for subjective experience, the ability to feel something, whether that is pain, pleasure, distress, or comfort. An animal can be sentient without being smart, and an animal can perform complex behaviors without being sentient. The key question is whether there is “something it is like” to be that organism, or whether it is running on autopilot with no inner life at all.
Researchers who study animal pain have proposed concrete criteria for evaluating sentience. Responses to harmful stimuli should differ from responses to harmless ones in measurable ways across neurobiology, physiology, and behavior. Crucially, subsequent behavior should change: the animal should learn to avoid the source of harm, show protective responses, or alter its decision-making in a way that suggests it is not just reflexively recoiling but is motivated to avoid future pain.1Animal Behaviour. Defining and assessing animal pain These criteria help distinguish nociception, the basic detection of tissue damage that every animal possesses, from the subjective experience of pain.2Physiological Reviews. Comparative Physiology of Nociception and Pain Even bacteria sense and flee from harmful chemicals. That does not make them sentient.
The Clearest Cases: Animals Without Nervous Systems
If sentience requires some form of neural processing, then the animals without any neurons at all are the most straightforward non-sentient candidates. Sponges (phylum Porifera) are the textbook example. They are multicellular animals that filter water through their bodies to feed, but they have no nerve cells, no synapses, and no capacity to transmit electrical signals from one cell to another in the way that nervous systems do. They respond to their environment through chemical signaling and cellular contractions, but these responses are local and slow. There is no central coordination, let alone anything resembling an inner experience.
Placozoans, tiny flat organisms that look like amoebas but are true animals, are in a similar position. They have only a handful of cell types and no neurons. Interestingly, recent genetic evidence suggests that sponges and placozoans may have once had ancestors with neural cell types and subsequently lost them, rather than simply never evolving them in the first place.3PubMed Central. Where is my mind? How sponges and placozoans may have lost neural cell types This does not change their current status. Whatever their evolutionary history, they lack the hardware that every proposed framework for sentience treats as a baseline requirement.
Nerve Nets and the Gray Zone
Jellyfish, corals, and sea anemones belong to the phylum Cnidaria, and they represent the first real complication. These animals have nervous systems, but they are structured nothing like a brain. Instead, cnidarians use nerve nets: diffuse webs of neurons spread across the body, with no centralized processing organ. Jellyfish swim, hunt, and react to light and touch, but the neural architecture driving those behaviors is remarkably decentralized.
Computational models of jellyfish swimming show that their motor-nerve-net neurons connect to local muscle groups, and the resulting contractions interact with water dynamics to produce coordinated swimming strokes.4PubMed Central. From single neurons to behavior in the jellyfish Aurelia aurita In hydrozoan jellyfish, a condensed ring of electrically coupled neurons processes sensory input and controls motor output. The neurons synchronize as signals travel around this ring, triggering coordinated waves of muscle activation that propel the animal forward.5PubMed Central. Neuronal Synchronization and Bidirectional Activity Spread Explain Efficient Swimming in a Whole-Body Model of Hydrozoan Jellyfish
This is impressive for an animal with no brain, but it is not clear that it constitutes sentience. The nerve net coordinates behavior, but nothing in the cnidarian body plan resembles the kind of integrated, centralized processing that most neuroscientists consider a prerequisite for subjective experience. A jellyfish can pull away from something harmful, but it is an open question whether it “feels” the harm or is simply executing a hardwired withdrawal. Most researchers place cnidarians in a gray zone: probably not sentient, but not as confidently non-sentient as sponges.
Starfish and the Puzzle of Radial Wiring
Echinoderms, the group that includes starfish, sea urchins, and sea cucumbers, pose a different kind of question. They have nervous systems, but those systems are organized around the animal’s radial body plan rather than concentrated in a head. A starfish has radial nerve cords running down each arm, connected by a central ring nerve. This system coordinates the movement of multiple arms and hundreds of tube feet, and even manages the regeneration of lost limbs.6PubMed Central. Structure and proteomic analysis of the crown-of-thorns starfish (Acanthaster sp.) radial nerve cord
Starfish can learn simple associations and navigate mazes, which has led some researchers to wonder whether their distributed nervous systems achieve more than they appear to. But the absence of anything resembling a brain, or any region where sensory information converges into a unified representation, makes it difficult to argue for sentience under most frameworks. Echinoderms sit in roughly the same uncertain territory as cnidarians: they process sensory input and produce coordinated behavior, but the architecture does not obviously support subjective experience.
Bivalves, Worms, and the Surprisingly Hard Cases
Not all mollusks are created equal in the sentience debate. Octopuses and cuttlefish are now widely considered likely sentient, with complex brains, problem-solving abilities, and rich behavioral repertoires. But their cousins the bivalves, including oysters, mussels, and clams, are a different story. Bivalves are largely sessile. They have ganglia (clusters of neurons) rather than a centralized brain, and these ganglia control basic functions like shell opening, feeding, and withdrawing from threats. In the Pacific oyster, for instance, the nervous system develops as several paired or fused ganglion clusters that innervate specific body parts: the cerebral ganglia control the velum, the pedal ganglia control the foot, and the visceral ganglia manage the adductor muscles and internal organs.7PubMed Central. Nervous system development in the Pacific oyster, Crassostrea gigas
This is a real nervous system, but a minimal one, and bivalves show very little behavioral flexibility. They do not learn complex associations, do not play, and do not display the kinds of motivational trade-offs that researchers look for as indicators of sentience. Many scientists consider bivalves unlikely to be sentient, though few would say it is impossible. Bivalves are one of the cases where the precautionary principle, discussed below, becomes relevant.
Worms present a related puzzle. The nematode Caenorhabditis elegans has exactly 302 neurons, and its entire connectome has been mapped. Researchers found that a simple circuit of just two sensory neurons and three interneurons enables worms to respond in contextually appropriate ways to harmful odors depending on their hunger state. That sounds like it could be evidence of sentience, but the analysis concluded otherwise. The circuitry resembles what is found in the human spinal cord, retina, and primary visual cortex, three regions that are neither necessary nor sufficient for conscious experience in humans. The authors argue that motivational trade-offs produced by such simple circuits are not a reliable criterion for subjective experience.8Biology & Philosophy. What if worms were sentient? Insights into subjective experience from the Caenorhabditis elegans connectome
Earthworms have considerably more neural complexity than nematodes. They possess cerebral ganglia above the pharynx that function as a rudimentary brain, plus a ventral nerve cord with paired ganglia in each body segment. Removal of the cerebral ganglia results in uncontrolled movement, suggesting a genuine coordinating role.9PubMed Central. Cerebral ganglionic variations and movement behaviors of Lumbricus terrestris on exposure to neurotoxin Whether this level of centralization is enough for any form of inner experience remains an open question. The honest scientific answer is that nobody knows.
Insects as Surprise Contenders
If you had asked most biologists a few decades ago whether insects were sentient, the answer would have been a confident no. That confidence has eroded considerably. Insects have small brains, but those brains are densely packed with neurons and organized into specialized regions. And the behavioral evidence has become hard to dismiss.
Research on fruit flies and bees has revealed a range of behaviors that look strikingly like emotional states. Male fruit flies deprived of mating opportunities seek out alcohol. Fruit flies exposed to dead companions show altered sensory perception, changed brain physiology, and even reduced lifespans. Painful stimuli in flies can induce a lasting state of heightened vigilance and hypersensitivity to previously harmless stimuli, resembling chronic pain responses in humans. Bumblebees trained to avoid crab spiders on flowers sometimes reject safe flowers after inspecting them, as though they were experiencing false alarm “hallucinations” of a spider. And in crickets, repeated defeats in male contests produce a long-lasting depression-like behavioral state that can be reversed by drugs affecting the serotonin system.10PubMed Central. The exploration of consciousness in insects
None of this proves insects are conscious. It is possible to construct mechanistic explanations for each behavior that do not require subjective experience. But the accumulation of findings has pushed many researchers toward taking insect sentience seriously, or at least declining to rule it out. Insects are the clearest illustration of how the sentience boundary has shifted in recent years.
How Much Brain Does Sentience Require
One approach to the question is to work backward from what we know about human consciousness and ask: what is the minimum neural hardware needed? Research on human brain injuries and neurological conditions has identified several brain regions whose loss does not eliminate consciousness, including the cerebellum, the hippocampi, the amygdalae, large regions of the cortex, and the connections between the two hemispheres.11PubMed Central. Minimal neuroanatomy for a conscious brain: homing in on the networks constituting consciousness By process of elimination, the structures that remain essential appear to be relatively modest: certain thalamocortical circuits and, intriguingly, the often-overlooked olfactory system.
The relevance for animals is that if consciousness in humans can survive the loss of most of the brain, the bar for the minimum neural architecture may be lower than previously assumed. This does not mean that a jellyfish nerve net clears that bar, but it does mean that dismissing an animal as non-sentient simply because its brain is small may be premature. Size alone is a poor predictor. The question is whether the architecture supports the right kind of integrated processing.
The Problem Nobody Can Fully Solve
Underlying all of this research is a fundamental philosophical difficulty. The only feelings you can directly verify are your own. When it comes to other beings, you can only infer inner experience from outward behavior, or from what they tell you. This “other-minds problem” is manageable within our own species because we have language, but it becomes acute with animals that cannot describe their experience.12Animal Sentience. Animal sentience: The other-minds problem
Scientists can measure nociception, document behavioral changes, map neural circuits, and even identify brain regions associated with consciousness in humans. But none of these measurements directly detect the presence or absence of subjective experience. They offer increasingly strong indirect evidence, and that evidence has gotten much better over the past two decades. Researchers are no longer groping in the dark. But a definitive test for sentience in another organism does not exist, and it may never exist given the nature of the problem.
When Sentience Begins in an Individual Animal
The question of which species are sentient has a developmental counterpart: at what point does an individual animal become sentient? This matters for farming and veterinary practice, where fetal animals are affected by procedures performed on their mothers.
Research on farm animals suggests that the neural structures needed for sentience are absent for at least the first half of pregnancy. By the time of birth, the required brain architecture is in place. But even in late pregnancy, when the fetal brain may have the structural capacity for consciousness, the fetus appears to remain in a continuous sleep-like state, displaying electrical activity consistent with unconsciousness.13Applied Animal Behaviour Science. Onset of sentience: The potential for suffering in fetal and newborn farm animals The onset of sentience, then, appears to coincide roughly with birth or the period immediately surrounding it, at least in mammals. In species with less developed brains at birth, or in animals that hatch from eggs at very early developmental stages, the timing could be different.
Where Researchers Think Consciousness First Evolved
Competing evolutionary theories place the origin of consciousness at very different points on the tree of life. Some researchers favor “early evolution” views tracing consciousness back to the Cambrian explosion, roughly 540 million years ago, when complex nervous systems first appeared in many animal lineages.14Philosophy Compass. The Evolution of Animal Consciousness Under these views, most animals with centralized nervous systems could be sentient to some degree.
Others favor “latecomer” theories that restrict consciousness to mammals, or even just to humans. One influential proposal places the emergence of emotion, and consciousness more broadly, among the early amniotes, the evolutionary group that includes reptiles, birds, and mammals but excludes amphibians and fish. This proposal is based on differences in brain chemistry, sleep structure, and behaviors like play, taste aversion learning, and emotional responses to stress that appear to have emerged in amniotes but not in earlier vertebrate lineages.15PubMed. The emergence of consciousness in phylogeny
If the amniote hypothesis is correct, then fish and amphibians might lack sentience, and invertebrates almost certainly would. If the Cambrian hypothesis is correct, the circle of sentient life is vastly wider. The honest state of affairs is that researchers disagree substantially, and the evidence does not yet settle the debate. The animals we can most confidently call non-sentient (sponges, placozoans) are non-controversial precisely because they lack nervous systems entirely. Everything else involves weighing probabilities.
How Policy Handles the Uncertainty
Because the science cannot yet draw a clean line, policy has increasingly relied on a precautionary approach: when the evidence is ambiguous, give the animal the benefit of the doubt. This principle has been applied most prominently to cephalopods and decapod crustaceans, both of which were added to animal welfare legislation in the United Kingdom in 2022.16Animal Sentience. Support for the precautionary principle
Frameworks for implementing this approach use a combination of neurobiological and behavioral criteria. One prominent framework uses four neural indicators and four behavioral indicators to determine when precautionary welfare protections are warranted for invertebrate species.17PubMed Central. Defending and refining the Birch et al. (2021) precautionary framework for animal sentience Species that score well on multiple criteria, as decapods and cephalopods do, get brought under the regulatory umbrella. Species that score poorly, like sponges, do not.
The practical challenge is that extending welfare regulations to new groups requires species-specific knowledge that often does not exist yet. Even for fish and cephalopods, where the scientific case for sentience is relatively strong, detailed welfare guidance is lacking.18PubMed Central. Considerations for implementing regulation of decapods in science For the vast majority of invertebrate species, the data needed to evaluate sentience have simply never been collected. The list of animals confirmed as non-sentient is short not because most animals are sentient, but because the research needed to confidently place any given species on one side or the other is expensive, time-consuming, and only beginning to be funded at scale.