Scientists Test Artificial Therapy Dogs Against The Real Thing With Shocking Results

Written by: Tori Holmes

March 10, 2016

In recent years, there has been a huge increase in the number of robotic animals that are used for therapy with seniors. Some of the biggest names, like Hasbro, have even begun making their own line of robotic companion animals targeted to seniors. But how do these robotic pets fare when compared to real life therapy dogs?


Animal visitation programs are now standard in many retirement and assisted living communities. Unfortunately, despite their popularity, we don’t actually know that much about whether or not they are making a positive change.

That’s where Dr. Karen Thodberg and her team at the University of Aarhus in Denmark come in. They wanted to get to the bottom of whether or not animal assisted therapy worked, and more importantly, whether robotic animals or the real thing were best.


In the first ever robot therapy animal/real therapy dog showdown, 100 individuals living in four nursing homes in Denmark were randomly assigned to one of three groups. The first was a therapy dog group, where participants would meet with a specific therapy dog each week. The second was a robot group, where participants spent time with PARO, a robotic baby seal that responds to touch and sound by moving and making funny noises (If you’ve seen Aziz Anzari’s “Masters of None” you’ll know exactly who PARO is). The third group was a toy group, where participants would spend time with a soft animal cat named Tom. Although cute, unlike PARO, Tom doesn’t move.

Picture 4-640

By dividing the participants into these three groups, the researchers hoped to answer two specific questions: “How nursing home residents interact with a real dog as compared to an interactive robotic pet or a cuddly stuffed animal” and “Do regular interactions with real and fake animals actually improve the psychological well-being and cognitive abilities of nursing home residents?”

Before the study began, participants underwent psychiatric evaluations to assess their cognitive status, dementia, ability to handle activities of daily living, levels of depression, confusion, and delirium. In addition to these factors, the participants’ quality of sleep was also measured, as it can play a significant role in cognitive abilities.


Once the evaluations were completed, the assigned “animal” and its handler visited the participants twice a week for six weeks. While the participant spent time with their animal visitor, the handler would monitor their behavior and take notes.

After the six weeks of visits had been completed, the same psychiatric assessment was again completed. The two assessments were then compared to examine the long-term impact of the visits on the participants.


While a detailed publication is in the works, Dr. Thodberg was excited to give a brief summary of what they found. Interestingly, at the start of the six weeks the differences in how participants responded to the dog and the robot were minimal. Participants touched, talked, and looked at both the robot and the real dog more when compared to the stuffed animal. These results didn’t last, though. Over time, the probability and duration of talking to and about the animals, and the likelihood that the participant would look at the animals decreased for both the robot and the toy cat.


Though, what is exciting is that the probability and duration of talking remained constant for the dog. The robotic PARO actually showed better results than the real therapy dogs at the beginning of the study, but slowly decreased, whereas the dogs stayed consistently high.

Interestingly, the level of interaction with the animal depended significantly on the participant’s level of impairment. Participants who were more cognitively impaired interacted more with both the real and fake animals, whereas participants with low cognitive impairment preferred to talk to the humans in the room.


These results are pretty amazing, but don’t get your hopes up too much. We have some bad news…

When Dr. Thodberg’s team compared the psychiatric assessments before and after the six weeks of animal therapy, they found no measurable benefits of either real or fake animals. Even man’s best friend had no impact on the cognitive abilities, levels of depression, psychiatric symptoms, or abilities to function in everyday life of the participants.


This doesn’t mean that therapy animals, both real and fake, should be completely foregone, though; they each have their place in retirement and long-term care facilities. For residents with severe cognitive impairments, the robotic animals especially helped them come out of their shells during the visits. For those with a higher level of cognitive functioning, spending time with a real dog provides companionship and a four-legged friend to talk to.


Overall, it seems that therapy animals do provide a temporary pick-me-up for residents, but the results are not long lasting. These might not be the results we were hoping for, but it’s important to note that these robotic and real life therapy animals did brighten the days of the participants who spent time with the animals, and that is still important.

H/t to Psychology Today

Featured image via @mahryacorgi / Instagram and Passages Hospice / YouTube

Print Friendly, PDF & Email
Written by: Tori Holmes

March 10, 2016

Nutritionist-crafted food for your dog's breed or mix.

Recipes designed for dogs' individuality



A themed collection of BARK-designed toys, treats, and chews.


A themed collection of BARK-designed toys, treats, and chews.