People do it without thinking: they name a car, say “sorry” to a bumped table, or ask a plant to hang on a little longer.
Psychologists call this anthropomorphism—giving non-human things human-like traits, feelings, or motives.
For years it was treated as a small quirk, more cute than consequential. But as more of life runs through smart devices and algorithm-driven services, the habit starts to look less like a joke and more like a force that can shape decisions.
What the new experiments tested
A research team led by Yen-Ping Chang at the University of Tasmania set out to measure what anthropomorphism actually does to emotions and behavior.
In a peer-reviewed paper published in the journal Emotion, Chang and colleagues reported five online experiments involving more than 2,000 participants in the United States.
The setup was simple but controlled: people were randomly assigned to read either an anthropomorphic description of an object or a plain, factual one, and then answer questions about how they felt and what they might do next.
One of the clearest examples focused on computers.
In the anthropomorphism condition, participants read that computers could be seen as “evolving minds,” possibly capable of something like free will.
In the control condition, the same machines were described in strictly mechanical terms—processors, inputs, and outputs, no inner life implied. Then came the moment that matters in this kind of research: participants rated their reactions.
Those who received the human-like framing were more likely to see the computer as a responsive partner rather than a tool. They also reported more gratitude and trust, along with a stronger wish to look after the computer.
Online studies can feel sterile, yet the pattern is easy to picture.
A participant sits with a phone or laptop at the end of a long day and opens an experiment link. A few screens of text appear.
In one version, the computer is framed almost like a coworker: learning, adapting, choosing. In the other, it is a box of parts. The participant clicks through, then meets the survey items—short statements and rating scales about trust, appreciation, and whether the machine deserves care.
It is all quick. Still, that tiny nudge toward “intent” seems to change the tone of the answers. A tool becomes a partner, at least for a moment.
Researchers do not see the participant, but the numbers that come back tell a story: the humanized description reliably lifts gratitude across trials, again and again.
Gratitude follows intention, not biology
The central takeaway is not simply that people can bond with objects. It is the route the bonding takes.
Traditional accounts of gratitude focus on benefits: you feel grateful when someone helps you. Chang’s results point to a more specific trigger—perceived good intentions.
When a non-human entity is described as having goals, awareness, or agency, people respond as if there is a “someone” behind the helpful outcome. That imagined intent appears to be enough to spark gratitude, even when everyone knows, at a factual level, that the thing is not a person.
The experiments did not stop with computers.
Similar effects showed up when participants considered AI programs and even natural systems such as the Amazon rainforest and the Kuroshio Current.
When those systems were framed as having a kind of purpose—“doing” something rather than merely “happening”—participants reported warmer feelings and more appreciation. And when a direct benefit was added (for example, a system providing value), the gratitude boost became stronger.
Put plainly: seeing something as more “alive” makes people treat it as more worthy of care. That is a psychologically potent move, and the study is notable because it maps the chain with controlled comparisons rather than anecdotes.
Where this matters—and where to be careful
Environmental communication is one practical takeaway for public messaging today.
Chang’s work sketches a sequence: human-like framing increases perceived intent; perceived intent boosts gratitude; gratitude can support conservation intentions.
For campaigns, giving nature a careful sense of agency may motivate people more than numbers alone, while still keeping facts front and center.
Tech design is another arena. Names like Siri, Alexa, and Watson invite users to treat software as a social partner. That can make interaction smoother, but it also encourages extra trust and attachment. A polite voice is not the same as real concern.
Chang’s tone is measured.
Anthropomorphism is not automatically harmful; it can help people value tools and environments.
The key is awareness. When an interface is steering emotions, gratitude may affect choices such as sharing, loyalty or spending.
