Calling a robot vacuum Greg can be a joke, a convenient label, or the beginning of a small social story. The name alone does not tell you which. The more revealing step comes when Greg is described as confused, stubborn, helpful, or offended: human motives are being used to explain a machine.
Psychologists call that attribution anthropomorphism. The video uses printers, cars, and household devices to explore it. A useful companion question is how a harmless nickname becomes an assumption about understanding or competence, especially when the object moves or talks back.
Naming and attributing a mind are separate steps
Names identify things. A boat, a computer server, or a vacuum can have a name for practical reasons without anyone believing it has feelings. Saying “use Greg upstairs” might be no more psychological than saying “use vacuum number two.”
Compare that with “Greg refuses to clean this room because he hates the rug.” Now a mechanical failure has been translated into a preference and intention. The sentence may still be playful. But it uses the same vocabulary that would explain a person’s behavior, replacing a question about sensors or traction with a social interpretation.
Epley, Waytz, and Cacioppo’s 2007 theory defines anthropomorphism around attributing human characteristics, motivations, intentions, or emotions to nonhuman agents. It is broader than naming and narrower than simply noticing that something moves. An object can have a label without a perceived mind, or a perceived personality without a name.
That distinction prevents a casual habit from being treated as a diagnosis. Someone who names a car may be organizing a household, repeating a family joke, or enjoying a story. Even seeing a face in the grille need not imply a belief that the car has a mind. Neither the face nor the nickname establishes loneliness or a particular personality trait.
Three reasons a human explanation becomes available
The 2007 paper proposes three factors: the accessibility of knowledge about people, the motivation to understand an agent’s behavior, and the desire for social connection. This is a theory about variation in anthropomorphism, not a formula that determines why every individual names an object.
Human behavior supplies a familiar explanatory model. If you do not know why a device pauses, “hesitation” is an available description. If you want to predict its next action, assigning a goal can make the behavior feel more intelligible. If you want connection, an object with a personality can seem more socially responsive than an anonymous mechanism.
Those possibilities can overlap. In an illustrative household example, a vacuum repeatedly circles a chair leg. Its owner calls it indecisive and gives it a name. The name may help everyone recognize the device, while the personality story makes the repeated movement easier to talk about. Neither establishes what caused the circling.
The theory suggests questions worth asking, rather than conclusions you can draw from the outside. Which cues invite a human comparison? What uncertainty is the person trying to resolve? Is the interaction serving a social function, a practical one, or both?
What experiments do and do not show
A 2008 follow-up study tested predictions about social connection and the motivation for control. In one study, 166 Harvard undergraduates who were current or former pet owners completed a loneliness measure and rated traits of their pets. Loneliness was associated with anthropomorphic traits relevant to social connection.
Another study examined responses to videos of dogs differing in apparent predictability and participants’ desire for control. The findings supported the authors’ account that both the observer’s motivation and the target’s behavior can matter.
These were studies of particular participants, measures, and nonhuman animals. The pet study did not randomly assign people to a lifetime of loneliness, and neither study established that naming a printer cures isolation. The results support parts of a psychological explanation; they do not license an inference about every owner of a named appliance.
A social cue can change trust
There is a practical consequence beyond amusement. Waytz, Heafner, and Epley’s 2014 experiment compared experiences in a driving simulator, including an autonomous vehicle with added humanizing features: a name, gender, and voice. Measures of participants’ responses indicated greater trust as the vehicle acquired these features.
That result concerns perception and behavior in a simulator. It does not show that naming an autonomous vehicle makes it safer on a road. The manipulation also bundled several features, so it should not be reported as a demonstration that a name alone caused the entire effect.
The distinction gives a useful way to evaluate another speaking device. If a pleasant voice makes a response feel competent, ask what evidence of competence would remain with the voice removed. Did the device complete the task? Can the answer be checked? Did it communicate its limitations clearly? These are practical questions prompted by the research, not claims that the experiment tested every modern assistant.
Keep the story while checking the mechanism
You do not need to stop calling a vacuum Greg to investigate why it gets stuck. Keep the nickname, then translate the complaint into something observable: it stops at the same rug edge, loses its route near a chair, or returns before the room is finished. An observable pattern gives you something to troubleshoot.
Do the same when the story is flattering. “It understands me” can mean the device reliably follows certain commands, remembers a preference, or simply responds in a friendly way. Those are different observations. Naming the behavior precisely prevents warmth from doing the work of evidence.
An object can become part of a household’s history without acquiring a mind. The human relationship to it is real as a relationship of memory, habit, or affection; the object’s attributed intentions still need scrutiny. That leaves room for the joke and for the practical question of what the machine can actually do.
