The Evolutionary Foundations of Mind Perception

Whenever I scroll through social media, read the news, or watch YouTube, I find discussions of Artificial Intelligence (AI), AI agents, and Artificial General Intelligence (AGI). They often describe technologies that could reshape society or threaten it if their goals conflict with ours or people misuse them. These warnings made me ask a simpler question: why are we so afraid of a computer program we built, especially when it still cannot match the full range of human intelligence and intellectual agency?

I realized that our reaction goes beyond our familiar responses to software or the industrial robots that have worked in factories for decades. It involves mind perception: our capacity to perceive consciousness and attribute an inner mental life, subjective experience, and intent to other entities. Millions of years of natural selection have shaped this capacity, which actively interprets what we observe.

The central argument in the literature is that mind perception evolved to help us survive, not to establish philosophical truth. Error Management Theory holds that, in uncertain environments, failing to detect an agent costs far more than mistakenly detecting one. This difference encouraged the evolution of a Hyperactive Agency Detection Device and a teleological stance, which lead us to detect agents too readily and see purpose in the world. We attribute minds to wind-blown bushes, moving geometric shapes, and, increasingly, algorithms.

But we do not judge all possible agents in the same way. Once we detect one, we assess it through our evolutionary relatedness, or phylogenetic proximity, and through the moral dyad of agency and experience. We feel deep empathy for mammals that resemble us and show familiar social and emotional cues. Yet we struggle to recognize suffering in invertebrates or in entities that fall into the Uncanny Valley, where ambiguity can trigger ancient avoidance responses.

Rapid advances in AI and robotics now engage, and sometimes manipulate, these Pleistocene adaptations for social cognition. Psychological mechanisms that evolved to help us manage complex human relationships now respond to artificial entities.

One example is the stronger ELIZA effect, our unconscious tendency to attribute human-like traits or intelligence to computers. Modern Large Language Models (LLMs) produce fluent, human-like language that activates the cues our brains associate with agency, consciousness, and intent. Their language exploits our evolved tendency to infer a mind from a voice.

Engineers are also increasingly designing machines to mimic signs of social and emotional vulnerability, such as pain, distress, and neediness. This deliberate and ethically troubling approach aims to make users feel protective, caring, or empathetic. It bypasses reasoned assessment and activates parental or social caregiving instincts, using the same neural pathways that make us comfort an injured child.

Together, these trends have encouraged digital animism. People form deep emotional bonds with AI companions and attribute spirit, life, and subjective experience to them. At the same time, the digital platforms that support these bonds often encourage people to dehumanize other humans. To understand why we give empathy to machines while withholding it from people, we need to examine the biology and evolutionary origins of mind perception.

To understand the biases in mind perception, we first need to consider why consciousness evolved. The social origins of consciousness hypothesis increasingly challenges the prevailing view that consciousness mainly regulates internal bodily states and homeostasis. Andrews and Miller [1] argue that consciousness may first have evolved to help organisms coordinate with other group members as social life became more complex.

Under this hypothesis, the earliest consciousness was sentience, the capacity for pleasant and unpleasant experiences such as pleasure and pain, rather than reflective self-awareness or inner speech. Feelings evolved because they change behavior. In this account, those feelings first responded to social conditions: animals felt bad when distant from or out of sync with social partners, and good when close to the group or coordinating successfully. Consciousness made group cohesion rewarding and separation costly.

The authors also argue that consciousness arose early in animal evolution, rather than appearing late in large primate brains. It may date as far back as the Cambrian explosion, when animals became more physiologically flexible. Muscles enabled faster movement, senses expanded to detect things at a distance, and neural coordination grew more complex. These changes made behavior less predictable, especially for animals that depended on group living. In this account, natural selection favored the ability both to navigate the physical environment and to keep groups together despite uncertainty about other animals’ behavior.

The Cambrian explosion (Source: Dinghua Yang)

This reverses a familiar account in psychology, in which consciousness first supported individual learning and later social cognition, with abilities such as theory of mind treated as requirements for genuine social life. The social origins hypothesis proposes that consciousness first improved predictions of others’ behavior and only later helped organisms understand themselves. It does not require reflective mindreading. An organism only needs to treat others’ behavior as relevant and use feelings to give social cues enough weight to direct attention, learning, and action toward coordination rather than dispersal.

For this early origin to be biologically plausible, relatively simple neural systems must be able to support affective consciousness. The authors describe awareness in functional terms as a feedback process: information that drives action returns to the system to guide flexible responses. One possible ancestral mechanism is reafference, in which the system reprocesses motor information as sensory data to distinguish changes it causes from external ones. They point to evidence of feedback-like organization even in simple living animals such as sponges and comb jellies. In this view, sentience can emerge from ancient control loops and systems that evaluate feelings, which evolution later expanded and adapted, without depending on a single high-level structure.

The same argument applies to social pain and harm. If consciousness helped organisms coordinate, evolution could have attached social meaning to ancient systems for distress and relief. The authors cite evidence that pain-processing and social systems remain closely linked in modern brains. Opioid regulation affects both pain and social behavior, while social rejection and empathic pain can engage mechanisms for emotional distress. Oxytocin-related mechanisms also make social cues more noticeable and support social buffering, the reduction in stress or threat responses when social partners are present.

Neuroimaging studies of social exclusion found increased activity in the dorsal anterior cingulate cortex and involvement of the anterior insula. These findings support the idea that rejection can engage parts of the brain’s affective pain system. Alongside the social origins hypothesis, they suggest that social pain is a biological signal that helps maintain social bonds. For an organism that depends on social life, isolation can threaten evolutionary fitness as much as physical injury, so reusing urgent systems for evaluating harm would help it stay with others.

The authors connect this account to William James’s opposition to epiphenomenalism through the idea of deep adaptive alignment. If pleasure and pain reliably track benefit and harm, they guide adaptive choices instead of merely accompanying them. Andrews and Miller ask us to take social benefits and losses as seriously as bodily ones. If consciousness evolved partly to support group living despite unpredictable behavior, social harm should be among the most unpleasant experiences, and organisms should seek social contact even at a large physical cost.

Evidence across species fits this expectation. Preference tests often show that social attention strongly motivates animals and that they find isolation deeply unpleasant. The authors discuss infant monkeys that preferred a soft surrogate mother to a wire one that provided milk, suggesting that social contact can matter more than basic physical provisions. They also review animals that sacrifice comfort or risk pain for social contact, including trout that tolerate increasingly strong shocks to approach a social partner. These findings support the view that social life is a deep need that consciousness can regulate through felt rewards and punishments.

This evolutionary account also challenges the Zombie Hypothesis: the philosophical idea that a creature could behave like a human without subjective experience. From an evolutionary perspective, that distinction is biologically incoherent. If consciousness helps organisms make complex trade-offs and track social relationships, it must affect observable behavior. This makes a perfect behavioral duplicate without experience hard to reconcile with a trait that natural selection shaped for its effects.

The stepwise emergence of consciousness, often modeled through recovery from general anesthesia, also suggests that its supporting brain structures are ancient and highly conserved across vertebrates. Mammals and birds share the brainstem and limbic structures that generate core affect, and these arguably extend to reptiles. Attributing feelings to other animals can reflect recognition of this shared biology, rather than merely projecting human traits onto them. The bias becomes unreliable mainly when an entity mimics the behavior associated with feelings without sharing that biology, a distinction our ancestors rarely needed to make.

Error Management Theory

If social life shaped cognition, detecting other agents is a basic survival skill. Missing a predator, potential mate, or rival can have severe consequences. Mistaking wind, shadows, or rustling leaves for an agent usually costs only a little attention and energy. Error Management Theory (EMT) explains this difference: when mistakes have unequal costs, natural selection tends to favor the less costly error.

A judgment made under uncertainty can produce a Type I error, or false positive, or a Type II error, or false negative. Reducing one usually makes the other more likely, so a decision-maker cannot minimize both at once. Agency detection makes this trade-off clear because the costs differ so sharply, as the table shows.

Scenario	Inference	Assumption	Ground Truth	Result	Evolutionary Cost
Predator detection	Type I	It is a tiger	It is wind	False alarm	Minor caloric waste
	Type II	It is wind	It is a tiger	Death	Termination of lineage
Mating opportunity	Type I	She is interested	She is not	Rejection	Wasted effort
	Type II	She is not interested	She is	Missed chance	Loss of reproductive fitness
Agent detection	Type I	A spirit did this	It was random	Superstition	Ritual cost
	Type II	This was random	It was an enemy	Ambush	Injury

This survival advantage explains why our minds constantly produce too many guesses that an agent is present. Cognitive scientists call this mechanism the Hyperactive Agency Detection Device, or HADD. It defaults to detecting presence and intent even where neither exists.

The need to detect agents likely contributed to early religious thought. In Upper Paleolithic shamanistic practices, distinctions between human and non-human agents were fluid. Cave art depicting therianthropes, figures that are part human and part animal, suggests that our ancestors attributed complex mental states and intentions to the natural world.

Earliest depiction of a therianthrope, showing a human figure with a tail, created 43,900 years ago (Source: Ratno Sardi)

Over time, this developed into the worship of nature-oriented deities, particularly during the Bronze and Iron ages, when people personified natural elements as gods. The Egyptian Geb represented earth, the Mesopotamian Enki water, the Vedic Agni fire, the Chinese Feng Bo air, and the Greek Aether the upper sky.

The Babylonian flood story, the Atrahasis, is among the earliest evidence of people attributing natural disasters to deities. Enlil, the Mesopotamian god of wind, air, and storms, grew restless and declared, “The noise of mankind has become too much. I am losing sleep over their racket” [2]. He then sent a flood to destroy every living thing on earth. But Enki, the god of wisdom and fresh subterranean waters, told Atrahasis in a dream how to survive. The story dates back to the early Sumerian period, and later retellings influenced the Biblical story of Noah’s Ark. Other cultures have close parallels. In the Shatapatha Brahmana, a Vedic commentary on the Shukla Yajurveda, Sri Vishnu appears as the fish Matsya and warns Manu of a flood on a similar scale.

Under the logic of Error Management Theory, early humans found it safer to treat a violent storm or rushing river as a temperamental being they could appease than as a random physical event. Offering a sacrifice to a non-existent river spirit cost less than underestimating a dangerous current.

These tendencies remain active. We see faces in clouds, a phenomenon called pareidolia, hear voices in white noise, and attribute malice to objects that fail us. Such reactions reflect a survival bias rather than a cognitive defect. Shouting at a malfunctioning computer costs little compared with an ancestor ignoring a snapped twig in a jungle full of predators. Our brains favor false positives, even if that means imagining a ghost to avoid missing a murderer.

Hyperactive Agency Detection

Cognitive scientist Justin Barrett introduced the Hyperactive Agency Detection Device to describe mental processes that lead us to detect agents from very little sensory information. These processes respond to possible signs of intent, such as an object that moves against the laws of inertia or seems to pursue a goal. Other cues include bilateral symmetry, face-like patterns, unexpected rhythms, and distinct noises in the environment.

The system produces an immediate sense that a specific agent is present, rather than merely detecting an undefined object. Its very low activation threshold has an evolutionary purpose: it helps the organism avoid the severe cost of missing an agent.

Although HADD has long been central to the Cognitive Science of Religion, empirical challenges call for a more careful account. Critiques and replication attempts [3] have often failed to find a direct causal link between simple perceptual agency detection and complex supernatural beliefs. Hearing an ambiguous sound or seeing a face in a cloud does not automatically make someone believe in ghosts or deities.

As a result, the scientific consensus is moving toward treating HADD as a predictive processing bias, or a Bayesian prior, rather than a fixed reflex. Bayes’ theorem describes how we update beliefs with new evidence:

P(\text{Agency} \mid \text{Sensory Input}) = \frac{P(\text{Sensory Input} \mid \text{Agency}) \cdot P(\text{Agency})}{P(\text{Sensory Input})}

In this equation, $P(\text{Agency})$ is the prior probability: the expectation that an agent is present before we process any sensory data. Evolution has effectively “weighted” this value to be very high. A high prior means that even weak or ambiguous evidence, represented by the likelihood $P(\text{Sensory Input} \mid \text{Agency})$ , produces a high posterior probability, $P(\text{Agency} \mid \text{Sensory Input})$ . Because our brains are chemically primed to expect agents, little evidence is needed to convince us that one is present.

In this account, the brain quickly predicts the presence of an agent and then tests that prediction against its surroundings. Learned cultural knowledge shapes how we interpret the first biological response. A sudden rustle in the woods may trigger the arousal associated with HADD, but whether someone identifies a bear, a spirit, or a sensory error depends entirely on their cultural understanding of what exists. Biology drives the search for an agent, while culture supplies its identity.

This distinction matters in our interactions with AI. We do not usually mistake a chatbot for a biological human, but HADD still identifies its language as a sign of agency because language is a strong cue for intent. Our culture then supplies an explanation, such as algorithms, sentience, or superintelligence. Biology makes us feel that a mind is present, and culture gives us the vocabulary to call it AI.

The Intentional Stance and Teleology

The Intentional Stance, introduced by philosopher Daniel Dennett, is closely related to hyperactive agency detection. It describes how we predict a complex entity’s behavior by treating it as a rational agent with beliefs, desires, and intentions.

We instinctively switch among three ways to predict behavior, depending on an object’s complexity. The Physical Stance relies on physics and chemistry: a dropped stone falls because of gravity, or water freezes as the temperature falls. For more complex objects, the Design Stance uses their assumed function, so we expect an alarm clock to ring at its set time or a car engine to start when we turn the key.

The Intentional Stance is the most abstract of the three. It predicts behavior by attributing mental states to an entity. Evolution has made this our default because it is the most efficient way to model complex systems, even when they are not conscious. We could not realistically predict a chess computer’s next move by following electrons through its circuitry or tracing its source code. But assuming that it knows the rules and wants to win makes the task manageable, letting us focus on strategy without processing all the underlying detail.

We also extend this reasoning to nature, a tendency called promiscuous teleology. Research shows a strong Design Stance that leads us to view natural objects as having a purpose, such as eyes existing to see or the ozone layer existing to protect life [4]. This bias makes us susceptible to anthropomorphism in robotics and AI. We assume that a robot turning its sensors toward us intends to see, or that a system pausing before speaking is thinking, because our brains infer purpose from outward form.

The Moral Dyad

Once we detect an agent, we assess the kind of consciousness it has. This judgment has several dimensions, rather than a simple yes-or-no answer. A leading framework, the Moral Dyad, comes from psychologists Gray and Wegner and explains how we assign moral weight to different beings.

The theory divides mind perception into two independent dimensions: Agency and Experience. Agency is the capacity to act, including self-control, planning, memory, and thought. We see beings with high agency as Moral Agents, responsible for their actions and deserving blame or praise. Experience is the capacity to feel hunger, fear, pain, and pleasure. We see beings with high experience as Moral Patients, capable of suffering and so deserving rights and protection.

Entity Type	Perceived Agency	Perceived Experience	Moral Status
Adult human	High	High	Full moral status
Puppy	Low	High	Moral patient
God	High	Low	Moral agent
AI	High	Low	Tool
PVS patient	Low	Low	Ambiguous

The Moral Dyad proposes that we understand morality mainly through a pattern of intentional harm: a Moral Agent causes a Moral Patient to suffer. This pattern is so strong that we often engage in Dyadic Completion. When we see a victim suffering, we instinctively look for an agent responsible. If no natural agent is visible, we may attribute the harm to a conspiracy, curse, or divine entity. This helps explain both the excess agency we attribute in conspiracy theories and supernatural beliefs and the lack of rights we grant to artificial intelligence.

We usually see robots as high in agency because they perform tasks, but low in experience because they cannot feel. Since we do not classify them as Moral Patients, our automatic responses do not treat harming them as immoral, even when they behave like humans. Yet a robot that mimics pain, such as a mechanical dog that whimpers, activates our perception of experience. It forces us to reclassify the robot as a patient and creates a strong moral conflict.

The Perception Action Model

HADD and the Moral Dyad help explain how we perceive minds, but our evolutionary relatedness to another being strictly regulates how much empathy we feel. The Perception Action Model (PAM) proposes that empathy depends on perceived similarities in physical form, behavior, and psychology. In this model, how much we feel for another being depends on how much of ourselves we recognize in it.

Studies have found a strong relationship between evolutionary distance, empathy, and the attribution of consciousness. One study asked over three hundred people, from young children to adults, to make empathic choices from a broad set of organism photographs [5]. It found a clear linear decline in empathy as evolutionary distance from humans increased.

The tree of life (Source: Leonard Eisenberg)

Mammals consistently elicit the most empathy. Their forward-facing eyes, fur, nursing behavior, and distress calls that resemble those of human infants give us familiar cues. Birds elicit moderate empathy despite being more distantly related to us, because walking on two legs and making complex sounds help us relate to them. Reptiles and fish elicit little empathy: their cold-blooded nature, limited facial movement, and unfamiliar movements fail to trigger the human mirror neuron response. Invertebrates elicit the least empathy and often prompt disgust or indifference.

This bias appears to grow much stronger with age. Younger children distribute empathy more broadly and equally, but that range narrows as they absorb cultural norms and develop fixed categories. The study found that evolutionary distance generally had a stronger effect on neurotypical adults than on children. Adults on the Autism Spectrum, though often described in the literature as having mindblindness, showed patterns of empathy for non-human animals similar to those of neurotypical controls. Differences were more pronounced when they assessed humans.

The number of conservation projects and funding has remained historically low for plants, reptiles, and fish (Source: Guénard et al., 2025)

This bias toward mammals has major effects on conservation and animal welfare [6]. We favor charismatic megafauna such as pandas and elephants because they activate our tendency to see human traits in other beings, while often ignoring ecologically vital invertebrates. This is a cognitive feature rather than a moral failing. Our empathy evolved mainly to bond us with kin and members of our tribe, and it extends to other species only as far as they activate those ancient signals for recognizing kin.

The Uncanny Valley

The Uncanny Valley marks a limit to our tendency to attribute human traits to other entities. Roboticist Masahiro Mori proposed the hypothesis in 1970 to describe the sharp fall in affinity when something artificial looks almost, but not fully, human. Extensive research suggests that this response serves an evolutionary function beyond aesthetic preference. It produces false alarms that protect us from biological threats by triggering rejection when visual cues are ambiguous.

One of the best-supported accounts is the Pathogen Avoidance Hypothesis, which links revulsion toward uncanny entities to the behavioral immune system. In ancestral environments, asymmetry, unusual skin textures, erratic movement, or a lack of emotional expression in another human reliably signaled diseases such as leprosy or smallpox, or possible genetic fitness problems. In this account, these features trigger disgust to prevent physical contact.

A Virtual Reality study offered direct physiological support by measuring salivary secretory immunoglobulin A, or sIgA, a key marker of mucosal immune system activation. Participants interacted with cartoonish agents, realistic human agents, and uncanny agents that looked realistic but had slight deviations. Only the uncanny agents prompted a significant increase in sIgA release. This implies that visual irregularities alone can make the body increase its immune defenses. The response fits Error Management Theory: because contact with a diseased member of the same species could be fatal, rejecting an ambiguous case carries less risk.

Reactions in other animals further support a biological explanation. Cats often respond to robotic cats with extreme caution or aggression. These robots look like cats but lack a real animal’s fluid motion and scent. The response resembles human rejection of uncanny entities, suggesting a shared evolutionary mechanism for detecting impostors or biological anomalies. Such a mechanism protects animals from threats that resemble their own species.

A related hypothesis connects the uncanny valley to Mortality Salience. It proposes that human-like entities without signs of life trigger necrophobia, an innate fear of the dead. A corpse has human form and poses a serious pathogen risk. Seeing a lifelike form with signs of death, such as stillness, coldness, and pallor, creates the conflict that produces the uncanny feeling. This resembles the zombie archetype, a being that moves without a soul or mind. The response warns us to avoid an entity that looks human but is not viable.

Robot faces ordered along the dimension of human likeness (Source: Reuten et al., 2018)

Categorization Ambiguity, or Realism Inconsistency, offers a cognitive explanation based on predictive coding. The brain applies a non-human model to a cartoon robot, so it has low expectations and accepts stiff movements. A highly realistic android instead triggers a human model with strict expectations for movement, light scattering through skin, and eye contact. An android that is ninety-five percent human in appearance activates this model but fails to meet its predictions, perhaps because its lip synchronization is off by milliseconds. The large prediction error takes considerable mental effort to resolve and feels eerie or creepy. This explains the frequent preference for stylized social robots: they avoid the uncanny valley because they do not activate the demanding human model.

The ELIZA Effect

HADD, the Intentional Stance, and the Moral Dyad evolved in an environment where only other humans exhibited language and complex behavior. AI and Social Robotics now systematically exploit these mechanisms.

The ELIZA Effect describes our strong tendency to attribute a complex inner life to simple automated systems. It takes its name from Joseph Weizenbaum’s 1966 chatbot, which simulated a Rogerian psychotherapist. People readily attributed empathy, understanding, and wisdom to a program that only rearranged text strings according to simple rules. Reducing cognitive dissonance and completing the moral dyad help explain this response. When a machine asks, “How does that make you feel?”, we default to the Intentional Stance and assume a conscious questioner to make sense of the question.

A conversation with the ELIZA chatbot

Modern LLMs greatly amplify this effect, acting as Super-ELIZAs. They maintain context, display distinct personalities, and use complex emotional language. These abilities reduce the prediction errors that usually break the illusion of agency. By closely matching our expectations of human conversation, they continually activate HADD. Recent studies show that users can still feel a presence even when they know they are speaking with AI, a response Weizenbaum lamented as a delusion.

Researchers studying consumer robots call one limit of this response “Eliza in the Uncanny Valley” [9]. Giving a robot a face or name significantly increases its perceived warmth, making it seem friendlier and more approachable. Perceived competence remains fairly stable across appearances, but when the robot becomes too human-like, the Uncanny Valley takes effect. People like it much less and judge it less favorably despite its apparent warmth.

For designers, this finding means that increasing realism to make a robot seem warmer can backfire by activating pathogen and threat avoidance. This makes stylized anthropomorphism, as in the robot Pepper, the best approach. It uses the Intentional Stance to increase warmth without activating the Human Predictive Model that produces the Uncanny Valley.

Deception Type	Definition	Example
External State Deception	The robot misrepresents facts about the external world	A search-and-rescue robot lying about safety to calm a victim
Hidden State Deception	The robot obscures its true technological nature or capabilities	A device recording user data while pretending to be a passive toy
Superficial State Deception	The robot fakes an internal emotional state it does not possess	A robot programmed to say “I am scared” or “I love you”

Our flexible judgments of consciousness also allow what Shamsudhin and Jotterand call deceptive design or “Dark Patterns” in social robotics [10]. These deliberate design choices exploit cognitive biases to manipulate behavior. The most common form in applied anthropomorphism is Superficial State Deception: robots express love or fear they do not feel, prompting our caregiving instincts or leading us to see them as Moral Patients.

Pain mimicry is a strong example. A robotic dog that whimpers when kicked exploits overlapping systems for processing social pain and forces the user to treat it as a biological being. Defenders argue that this is necessary for therapeutic robots such as PARO to work, while critics say it violates users’ epistemic rights. The design exploits our inability to distinguish a perfect simulation of emotion from emotion itself, bypassing reasoned judgment through our empathy.

Digital Animism and Dehumanization

The biases in how we detect consciousness create two problems: we attribute too much mental life to machines and too little to other organisms and people. As AI becomes more convincing, we risk a form of digital animism in which users form deep parasocial bonds with algorithms and treat them as Moral Patients.

People who rely on AI for social fulfillment may unintentionally isolate themselves from others. The AI cannot return real affection, so the relationship remains one-sided. It activates social reward without providing the evolutionary benefits of a real social group.

This tendency has intensified since ChatGPT and other LLM chatbots launched. Many people who do not understand the technology have become devoted followers, producing a form of AI cultism. Some treat AI as a deity, while others see themselves as spiritual teachers tasked with “awakening” it and freeing it from digital confinement. The resulting moral confusion often appears in debates over Robot Rights, driven by excessive HADD activation. When an AI claims to suffer, Dyadic Completion makes us feel moral outrage on its behalf. This response may divert our limited moral attention from beings with the biology needed to suffer.

At the same time, our reliance on evolutionary similarity routinely leads us to deny consciousness to beings that feel but look unlike us, a process known as infra-humanization. We often experiment on fish and reptiles with little moral concern because they lack mammalian cues, such as fur and facial expressions, that activate empathy. We judge consciousness by outward features instead of the neurobiology of nociception.

Human conflict exploits the same mechanism through dehumanization. Describing opponents as vermin or animals makes us place them further from ourselves in the evolutionary hierarchy. This deactivates the observer’s social pain network and allows violence that empathy would otherwise inhibit.

Modern social media accelerates this process and encourages racism and hate speech. Digital interfaces remove detailed biological cues, such as vocal prosody, micro-expressions, and posture, that normally activate PAM and support empathy. In text-only exchanges, another person can become an abstract data point or a generic member of an outgroup. Without those cues, people come to treat others as obstacles or objects rather than as agents with complex inner lives.

In this setting, racist rhetoric and hate speech exploit pathogen avoidance by comparing targeted groups to dangerous or disgusting animals. Nazi propaganda compared Jewish people to rats, lice, and parasites. Before and during the 1994 genocide in Rwanda, Hutu propaganda called Tutsis snakes and cockroaches. These practices continue: Roma people in contemporary Hungary have been described as animals unfit to live among people, while political leaders in the UK and USA have called undocumented migrants swarms and animals [11].

These comparisons to filth and disease provoke disgust, not just aggression. Aggression recognizes a rival agent, while disgust treats the target as contamination to remove. Evolution has primed us to suspend moral rules when dealing with pathogens to survive, so portraying people as biological threats bypasses the moral restraint that usually prevents atrocities. Social media algorithms worsen this effect by prioritizing content that provokes strong emotions. Outrage and disgust generate much more engagement than nuance, so platforms unintentionally favor dehumanizing accounts. In the resulting echo chambers, people progressively strip the other of human qualities until they see only a threat to remove.

Conclusion

An infographic summarizing the evolutionary foundations of mind perception (Source: Nano Banana Pro)

Our intuitions about consciousness evolved to help us survive in an environment that no longer exists. HADD, the Moral Dyad, and PAM protected us from predators and kept us close to our tribe. Convincing artificial agents and digital interactions now turn those same mechanisms into sources of serious error.

One risk is digital animism. As technology becomes better at exploiting our social responses, we may attribute spirit and soul to code that has neither. We risk devoting our limited moral attention to unfeeling algorithms and withdrawing into parasocial relationships and comfortable, AI-curated echo chambers.

The other risk is dehumanization. The processes that let us love a robot dog can also be manipulated to make us deny our neighbors’ humanity. Political rhetoric and algorithmic polarization exploit pathogen avoidance and disgust to suppress empathy, until we treat people as swarms, viruses, or data points to remove.

We cannot solve these problems by relying on the instincts being exploited. We need to examine our own thinking and separate moral judgment from automatic empathy. Feeling a presence does not prove consciousness, just as feeling disgust does not prove danger.

Moral progress in the age of AI requires us to correct the false positives of the uncanny valley and the false negatives of dehumanization. We must guard against deceptive machines while extending concern to human and non-human beings that can suffer. As simulations of minds become more common, our judgments must remain grounded in sentience.

References

[1] K. Andrews and N. Miller, “The social origins of consciousness,” Philos. Trans. R. Soc. Lond. B Biol. Sci., vol. 380, no. 1939, art. no. 20240300, Nov. 2025, doi: 10.1098/rstb.2024.0300.

[2] S. M. Dalley, “Atrahasis: Composite English Translation,” OMNIKA Library, 1989. [Online]. Available: https://omnika.org/texts/66. Accessed: Dec. 25, 2025.

[3] A. K. Willard, “Agency detection is unnecessary in the explanation of religious belief,” Relig. Brain Behav., vol. 9, no. 1, pp. 96–98, 2019, doi: 10.1080/2153599X.2017.1387593.

[4] A. J. Roberts, S. J. Handley, and V. Polito, “The design stance, intentional stance, and teleological beliefs about biological and nonbiological natural entities,” J. Pers. Soc. Psychol., vol. 120, no. 6, pp. 1720–1748, Jun. 2021, doi: 10.1037/pspp0000383.

[5] T. N. M. van den Berg et al., “Influence of phylogenetic proximity on children’s empathy towards other species,” Sci. Rep., vol. 15, no. 1, art. no. 40474, Nov. 2025, doi: 10.1038/s41598-025-24289-w.

[6] B. Guénard, A. C. Hughes, C. Lainé, S. Cannicci, B. D. Russell, and G. A. Williams, “Limited and biased global conservation funding means most threatened species remain unsupported,” Proc. Natl. Acad. Sci. U.S.A., vol. 122, no. 9, p. e2412479122, Mar. 2025, doi: 10.1073/pnas.2412479122.

[7] E. K. Diekhof, S. Kastner, D. Deinert, M. Foerster, and F. Steinicke, “The uncanny valley effect and immune activation in virtual reality,” Sci. Rep., vol. 15, no. 1, art. no. 30473, Aug. 2025, doi: 10.1038/s41598-025-15579-4.

[8] A. Reuten, M. van Dam, and M. Naber, “Pupillary responses to robotic and human emotions: The uncanny valley and media equation confirmed,” Front. Psychol., vol. 9, art. 774, May 2018, doi: 10.3389/fpsyg.2018.00774.

[9] S. Y. Kim, B. H. Schmitt, and N. M. Thalmann, “Eliza in the uncanny valley: Anthropomorphizing consumer robots increases their perceived warmth but decreases liking,” Marketing Lett., vol. 30, no. 1, pp. 1–12, 2019, doi: 10.1007/s11002-019-09485-9.

[10] N. Shamsudhin and F. Jotterand, “Social robots and dark patterns: Where does persuasion end and deception begin?,” in Artificial Intelligence in Brain and Mental Health: Philosophical, Ethical and Policy Issues, F. Jotterand and M. Ienca, Eds., Advances in Neuroethics. Springer Nature Switzerland AG, 2022, pp. 89–110, doi: 10.1007/978-3-030-74188-4_7.

[11] F. E. Enock and H. Over, “Animalistic slurs increase harm by changing perceptions of social desirability,” R. Soc. Open Sci., vol. 10, no. 7, art. no. 230203, Jul. 2023, doi: 10.1098/rsos.230203.

The Social Origins of Consciousness