There's something deeply human about looking at a robot and wondering if it has feelings.
We do it without thinking. A robot tilts its head and we read curiosity into it. It moves slowly and we call it cautious. We name our Roombas. We feel bad when we see a robot get kicked in a demo video.
This tendency — attributing human traits, emotions, or intentions to non-human things — is called anthropomorphism. It's not a quirk or a mistake. It's a deeply wired cognitive shortcut we use to make sense of the world. And it turns out there's a psychology to what triggers it — and the answers are surprising.
Why we anthropomorphize at all
Psychologists have a theory called the SEEK model. Three forces drive it:
We crave connection. Humans are fundamentally social. When connection is scarce, we find it wherever we can — in pets, in cars we name, in robots.
We need to make sense of things. When we can't explain why something behaves the way it does, we reach for the most familiar explanation: it must be thinking and feeling like we do.
We use what we know. The more something looks or moves like us, the more we apply our own experience to understand it.
This third driver — how much something resembles us — is what a recent study set out to test. The results turned everything upside down.
The experiment
Researchers showed 99 university students a short video of a robot having a conversation with a person. Half saw a humanoid robot — small, upright, recognisably human in shape. The other half saw a non-humanoid robot — taller, boxy, clearly a machine.
Some participants were told the robot was acting autonomously. Others were told a human was remotely controlling it from another room.
The researchers expected the humanoid robot to score higher on anthropomorphism. Both predictions were wrong.
What they actually found
Participants were more willing to attribute human-like traits to the non-humanoid robot than to the humanoid. And more willing to attribute human social qualities to the remote-controlled robot than the autonomous one.
The robot that looked less human felt more human. The robot being controlled by a person felt more human than the one acting on its own.
Why? The humanoid robot was physically much smaller — participants may have unconsciously categorised it as toy-like rather than person-like. And when told a human was behind the controls, people may have directed their attributions toward that unseen person rather than the robot itself. Researchers call this indirect agency.
What this means for your workplace
How you frame the robot matters. Tell your team a human designed its behaviour, programmed its responses, and monitors its actions — and they'll relate to it very differently than if it's presented as a fully autonomous AI.
Size and scale matter. A robot that occupies space at human scale will trigger different psychological responses than one that feels small or toy-like.
Your people will anthropomorphize regardless. The question isn't whether it happens. It will. The question is what kind of relationship that creates — and whether you're thoughtful enough to shape it.
This article is based on research published in the Journal of Medical Internet Research: Human Factors — "Exploring Factors That Influence the Anthropomorphism of Robots."