Definition
Sentience: Sentience is the capacity to have experiences that feel like something and that are good or bad for the subject, such as pain, pleasure, comfort or distress. Philosopher Jonathan Birch defines it as the capacity for valenced experience. It is widely treated as the main basis of moral status and of animal welfare law.
Also known as: sentient, capacity for valenced experience, phenomenal sentience
Sentience is the capacity to have experiences that feel good or bad to the one having them, such as pain, pleasure, comfort or distress. Philosopher Jonathan Birch defines it as the capacity for valenced experience. Sentience is widely treated as the basis of moral concern and animal welfare law, and it is central to debates about brain organoids.
Key takeaways
- Sentience means valenced experience: not just processing information, but having states that feel positive or negative.
- It is narrower than consciousness and distinct from intelligence; a system can learn without feeling anything.
- Scientists cannot observe sentience directly, so they assess it through indicators such as pain-related behavior and neural structure.
- UK law recognizes vertebrates, cephalopod molluscs and decapod crustaceans as sentient for policy purposes in the Animal Welfare (Sentience) Act 2022.
- There is no evidence that current brain organoids are sentient, and no law treats them as such, but the question is expected to be revisited as models grow more complex.
What does sentience mean?
The word comes from the Latin sentire, to feel. In current philosophy and animal welfare science, sentience usually refers to the capacity for experiences with valence: experiences that are pleasant or unpleasant, good or bad for the subject. In The Edge of Sentience (2024), Birch uses exactly this definition, the capacity for valenced experience, and builds a framework for decisions about beings whose sentience is uncertain.
Valence is the key idea. A thermostat responds to temperature, and a bacterium moves toward nutrients, but on mainstream views neither has anything that feels good or bad to it. Sentience requires that responses come with felt states.
Sentience compared with related terms
| Term | Core meaning | Example question |
|---|---|---|
| Consciousness | Any subjective experience; there is something it is like to be the system | Does it experience anything at all? |
| Sentience | Experiences with valence, such as pain or pleasure | Can things go well or badly for it, from its own point of view? |
| Intelligence | Ability to learn, solve problems and adapt behavior | Can it improve at a task? |
| Sapience | Higher cognition, such as reasoning or self-reflection | Can it think about its own thoughts? |
| Moral status | Mattering morally for one’s own sake | Do its interests count in their own right? |
The relationship between these is explored in depth in can brain organoids become conscious? and do biological computers have moral status?
How do scientists assess sentience?
Because sentience is private, researchers rely on indirect evidence. Common types of indicator include:
- Neural structures that integrate sensory signals, such as those linked to pain processing.
- Nociceptors, sensory receptors that detect potentially damaging stimuli.
- Flexible responses to injury, such as wound tending or guarding, beyond simple reflexes.
- Motivational trade-offs, for example accepting a cost to avoid a noxious stimulus.
- Responses to analgesics or anxiety-reducing drugs that change behavior in expected ways.
- Learning from harmful events, such as avoiding places associated with them.
No single indicator is decisive. In Trends in Cognitive Sciences (2020), Birch, Schnell and Clayton argued that consciousness varies along several dimensions, including perceptual richness and evaluative richness, rather than on one scale. Evaluative richness, how much an animal’s experiences feel good or bad, is the dimension most directly connected to sentience and welfare.
Most of these indicators are behavioral or anatomical, which creates a problem for brain organoids: they have no body, no behavior in the usual sense and no sensory organs.
What does the law say about sentience?
The UK Animal Welfare (Sentience) Act 2022
The Animal Welfare (Sentience) Act 2022 received Royal Assent on 28 April 2022 and extends across the UK. It does three main things:
- Creates an Animal Sentience Committee, appointed by the Secretary of State.
- Lets the Committee report on whether, or to what extent, government policy has had all due regard to ways it might adversely affect the welfare of animals as sentient beings. The Secretary of State must lay a response to each report before Parliament within three months (excluding certain periods when Parliament is not sitting).
- Defines “animal” as any vertebrate other than humans, any cephalopod mollusc and any decapod crustacean, with a power to add other invertebrates by regulations approved by both Houses of Parliament.
The inclusion of octopuses, crabs and lobsters reflects a legislature acting on credible but incomplete evidence of sentience, an approach connected to Birch’s work on the precautionary principle for animal sentience.
Limits for organoids
The Act’s scope is specific animals and government policy. It does not create welfare duties for laboratories, and it does not cover cell cultures, tissues or organoids. Similarly, EU Directive 2010/63/EU on animals used for scientific purposes protects live non-human vertebrates and cephalopods, not cultured human cells. As far as we are aware, no jurisdiction currently treats brain organoids as sentient in law. This entry is not legal advice.
Can brain organoids be sentient?
What the evidence says
There is no evidence that any brain organoid made so far is sentient. The US National Academies’ 2021 report on human neural organoids judged that awareness, emotion or pain in organoids was extremely unlikely in the foreseeable future, while recommending reassessment as the science advances. Current organoids are small, lack blood vessels and pain pathways, and receive no natural sensory input.
The DishBrain wording debate
Confusion has also come from language. Kagan et al. reported in Neuron (2022) that cultured neurons on a high-density microelectrode array improved at a simplified Pong game when given structured feedback, a study known as DishBrain. Their title used “sentience” in a technical sense tied to the free energy principle: responsiveness to sensory information. Balci et al. replied in the same journal (2023) that the term risked misleading readers, because adaptation is not evidence of felt experience. Learning through synaptic plasticity and closed-loop stimulation does not by itself show valence.
Why the question stays open
The field is moving toward larger, longer-lived and more connected tissue, including assembloids and vascularized organoids, and toward richer closed-loop interaction. Birch’s framework asks when a system becomes a sentience candidate: when there is a realistic, evidence-based possibility of sentience that should not be ignored. His book discusses neural organoids among its cases. Views differ on what would make an organoid a candidate. Some emphasize integrated, sustained activity and sleep-like cycles; others hold that without a body and sensory pathways, valenced experience is very unlikely whatever the activity. Our essays on switching off biological computers and the ethics of organoid intelligence cover what this means in practice.
Why does sentience matter for organoid intelligence?
Organoid intelligence research aims to compute with living neural tissue, so it needs a clear vocabulary. Careful use of “sentience” helps in three ways:
- Honest communication. Claims that a culture “learns” or “responds” should not be presented as claims that it feels.
- Proportionate oversight. Knowing which developments would raise sentience questions lets institutions plan review in advance.
- Measurable reporting. Size, age, cell composition and activity patterns can be documented so that others can judge risk.
At Organoid Programming we avoid describing organoids as sentient in our own voice, and we are building experiment records that keep culture details and activity data together, as described in our approach to ethics. Sentience remains the concept to watch as the technology develops.
Frequently asked questions
What is sentience in simple terms?
Sentience is the ability to feel. A sentient being has experiences that are pleasant or unpleasant for it, such as pain, pleasure, fear or comfort. It is narrower than consciousness in general, which can include neutral experiences, and different from intelligence, which is about solving problems. Philosopher Jonathan Birch defines sentience as the capacity for valenced experience.
What is the difference between sentience and consciousness?
Consciousness, in the philosophical sense, means having any subjective experience at all: there is something it is like to be the system. Sentience usually means having experiences with valence, that feel good or bad. Most researchers treat sentience as a form of consciousness. The distinction matters ethically because valenced experience, especially suffering, is what welfare protections aim to address.
Which animals does UK law recognize as sentient?
The Animal Welfare (Sentience) Act 2022 defines animal, for its purposes, as any vertebrate other than humans, any cephalopod mollusc such as an octopus, and any decapod crustacean such as a crab or lobster. It lets the Secretary of State extend the definition to other invertebrates through regulations approved by both Houses of Parliament. It does not cover cell cultures or organoids.
Are brain organoids sentient?
There is no evidence that any brain organoid made so far is sentient. A 2021 National Academies report judged that awareness or pain in current organoids is extremely unlikely in the foreseeable future, while recommending the question be revisited as models become more complex. Organoids lack sensory organs, pain pathways, blood vessels and the large-scale structure linked to experience in animals.
Did the DishBrain paper prove neurons in a dish are sentient?
No. Kagan et al. (Neuron, 2022) used sentience in a technical sense, meaning responsiveness to sensory input framed by the free energy principle, while showing that cultured neurons adapted their activity during a simplified Pong game. Balci et al. (2023) argued in the same journal that this wording risked misleading readers. The study showed closed-loop adaptation, not felt experience.
What is a sentience candidate?
A sentience candidate, in Jonathan Birch’s framework, is a system for which there is a realistic, evidence-based possibility of sentience that cannot responsibly be ignored, even though sentience is not proven. Being a candidate does not mean a system is sentient. It means its welfare deserves consideration and that precautions proportionate to the risk should be discussed.
Sources
- Birch, J. (2024). The Edge of Sentience: Risk and Precaution in Humans, Other Animals, AI, and Beyond. Oxford University Press (open access).
- Birch, J., Schnell, A. K. & Clayton, N. S. (2020). Dimensions of animal consciousness. Trends in Cognitive Sciences.
- Birch, J. (2017). Animal sentience and the precautionary principle. Animal Sentience.
- Animal Welfare (Sentience) Act 2022 (UK). legislation.gov.uk.
- Directive 2010/63/EU on the protection of animals used for scientific purposes. EUR-Lex.
- Kagan et al. (2022). In vitro neurons learn and exhibit sentience when embodied in a simulated game-world. Neuron.
- Balci et al. (2023). A response to claims of emergent intelligence and sentience in a dish. Neuron.
- National Academies of Sciences, Engineering, and Medicine (2021). The Emerging Field of Human Neural Organoids, Transplants, and Chimeras.
