Robotic Dog

Robotic Dog

The daily companionship of a pet can mean a lot to an elderly person, especially if they have only limited contact with friends and family. But not all elderly people are able to care for a pet, and perhaps their nursing home or retirement community doesn't even allow them. That's why researchers are trying to see if an interactive robot can act as a substitute. In this Science Update, you'll hear about their investigation.


Bonding with a plug-in pet. I'm Bob Hirshon and this is Science Update.

(sounds of AIBO waking up)

This is the sound of the Sony AIBO. It's not a computer or a video game—it's a robot shaped like a small dog. The AIBO can chase a ball, understand commands, and even walk to a docking station and recharge its own batteries.

The aibo has a devoted following among high-tech hobbyists. But it may also serve a practical purpose. Alan Beck is director of Purdue University's Center for the Human-Animal Bond. He and his colleagues are studying whether the AIBO can act as a companion for elderly people.


Animals are a source of attention, and a source of humor, and a source of touch and comfort. Well, for people who can't even have animals, will the robotic dog serve at least some of those same roles, and therefore some of the same health benefits?

The researchers work with residents at a local assisted living facility, giving them an AIBO for six weeks.


It was amazing how soon older people get caught up into appreciating the animal as a target of care, of something worth nurturing, of something worth talking to just like it was a puppy.

The study will continue through next year, and Beck says they hope to learn whether robots like AIBO can, in fact, improve an older person's well-being. For the American Association for the Advancement of Science, I'm Bob Hirshon.

Making Sense of the Research

You probably wouldn't expect that people would bond with a robot dog. Dr. Beck didn't think so either. He assumed that since the AIBO looks more like a robot than a dog (it's made of shiny metal and has glowing lights instead of eyes), people wouldn't relate to it on an emotional level. And in fact, many of the elderly people he studied thought the same thing—but after owning the dog for a few weeks, they changed their attitudes without even realizing it.

Being surprised by the results of your own research is fairly common in science. There's nothing wrong with starting an experiment with some expectation of how it might turn out. In fact, it's encouraged (it's called a hypothesis). But once you form your hypothesis, you have to be equally willing to see the evidence supported or overturned.

One of the challenges in this study was how to take an open-ended, subjective question ("Can a robot dog be a good companion?") and turn it into something that could be objectively studied. The researchers did this in several ways. Among them were videotaping people's interactions with the robot dog, and keeping track of the number and type of interactions they had (for instance, did they try to play with the dog, or just recharge its battery?). The researchers also analyzed their elderly subjects' facial expressions as their relationship with the dog developed.

In addition, the researchers asked their subjects a number of questions about AIBO at the start of the experiment, and asked the same questions again after three and then six weeks of ownership. For example, one question was "What does AIBO need?" People who didn't yet have the robot usually gave technological answers like "a place to plug in the charger," but after a few weeks, the responses were more nuturing, like "he needs to be held and talked to."

Beck and his colleagues are also studying the AIBO in relation to young children. They're asking children questions about the robot's physical properties (like "Does AIBO have a brain?"), mental states ("Can AIBO feel scared?"), potential for companionship ("Can AIBO be your friend?") and morality ("If AIBO knocks over a glass of water, should AIBO be blamed?"). And they're comparing their interactions with AIBO to separate interactions with real pets. The goal is to figure out exactly how different robots are from the real thing in the minds of the children—if they are very different at all.

This research is supported by the National Science Foundation, and involved many scientitsts at the University of Washington, including Drs. Peter Kahn and Batya Friedman, Dr. Nancy Edwards of the School of Nursing, and Dr. Gail Melson from the Department of Child Development and Family Studies. Although the AIBO is the robot dog used in the study, the researchers are not affiliated with the makers of the product.

Now try and answer these questions:

  1. What hypothesis did Beck form about his study with elderly people and AIBO? What was the result?
  2. How did he examine the bond between the elderly subjects and the robot dog?
  3. Can you think of other ways that you might scientifically study this issue? In what ways can you measure the quality of a relationship?
  4. In the interviews conducted for this study, almost everyone said they preferred a real dog to a robot dog, all things being equal. But when asked if they would like to live on a planet where only real dogs existed, many elderly people said no, because they couldn't have one. Taken together, what do both of these answers suggest?
  5. What aspects of the bond between a human and a robot pet (or a real one) cannot be measured scientifically? Are there any?

For Educators

Value Sensitive Design Projects is the University of Washington's site devoted to these projects. Follow links to related studies on robot dogs and children.

The Center for the Human-Animal Bond is Dr. Beck's home base, and one of the sponsors of this research.

AIBO Worldwide is Sony's official site for the AIBO robot dog.

Related Resources

Solar Power from Moon
6-12 | Audio
Nonstick Chewing Gum
6-12 | Audio
Electricity from Sewage
6-12 | Audio

Did you find this resource helpful?

Science Update Details

Grades Themes Project 2061 Benchmarks