AI in Play

Robotic pets are a relatively new entrant into the senior care and memory care space. Joy For All launched their first Companion Pets with Hasbro in 2015, and the company has since worked on several research projects and studies on how the Pets combat the epidemic of loneliness among seniors. Ted Fischer, president and CEO gives HomeCare an update on the company below.

How is the AIRES Project going?
We were originally awarded that grant—back in our Hasbro days—with Brown University and their Humanities Center Of Robotics Initiative. The grant is probably two-plus years in. We’ve done quite a bit on the research side of that project, which is really trying to figure out if we are to add artificial intelligence (AI) to the platform, what would that be and to what end? Why would we add it?

One of the things We’ve been enormously protective of is some of the magic that occurs between our product and the end user is the simplicity. You don’t really need instructions for the product. You take it out of the box, name it Fluffy and the magic starts. We don’t want to ruin that anything about that relationship, and we think that’s the most important piece. 

We went into this grant process with Brown with the idea that we aren’t adding AI for the sake of AI. We want to make sure it fits in with our product and what the product was intended to do and how we thought about what brings joy and happiness and makes meaningful connections. 

We’ve just now started the prototyping process as the next phase. We’ve got a number of prototypes that will contain a chip that that will allow us to do more research to really define why we would make this a connected device. The guiding principles are 1) to enhance the user experience; and 2) to enhance the feature set to make it more lifelike, more fun, more playful. I hate to use the word monitoring, because there’s a lot of devices out there to monitor older adults. But I think there’s a passive monitoring. There’s a lot of knowledge to be gained from analyzing the play pattern. How often does mom or dad interact with the pet, with what frequency and for how long? I think understanding those things in the next phase, I think that will help us end up at the right place if we do develop a connected product to enhance what we’ve already got.
    
What are some additional goals that have occurred from the partnership? 
The whole intent of AIRES was to figure out some simple daily weekly tasks or things that are bothersome in some way that out product in a connected form might help alleviate or eliminate. A lot of people are saying, “Your cat should give pill reminders!” But, would a real cat do that? Probably not, but our cat might. I know that pill reminding is a great and wonderful thing and I don’t want to undermine that. We want to maintain the sanctity of the relationship with the person and I don’t want to be aggravating the person by adding something that will take away from that relationship. So what we want to know is does mom or dad wake up every day and by 10 interact with the pet. Is there a cadence to that? How many times during the day? So you start thinking about the meaningful implications of that, to a family member, that could be very valuable to know. 

How do you address concerns about potential vulnerabilities in robotic companions?
We had the ability to add a chip when we launched. There were other products out there at Hasbro and other toy manufacturers that were intended for children with these capabilities. And to be honest, it scared the heck out of me for this demographic. I don’t think at that point you could say it wasn’t hackable. So part of this grant is thinking—depending on the chip or the AI compatibility—how do we ensure safety? If the chip is not part of operation of the product, and just monitoring touch or movement, I think you can separate those things in a safe way where the hackability wouldn’t get in the way of gaining data or other information and it wouldn’t impact usability. This won’t be the mic that’s listening or the camera that’s watching. Those are active monitoring. I go back to why we launched this brand and what its about and it’s about fun and joy and play and enhancing human interaction. 

There will always be a nonconnected version of our product and if we launch a connected version, it will be up to the caregiver to make that decision that that’s the right Pet for their loved one. We won’t say you get the connected device or nothing. We’ll be relying on caregivers to make that decision. 

Where do you see artificial intelligence (AI) heading in the seniorcare/homecare market in the future? 
My focus group was my 97-year-old grandmother, and I knew I didn’t want a camera on her bathroom door in her memory care. But others will make different decisions for their own family. There’s all levels of monitoring, from very aggressive with cameras to less severe with sensors. And you keep going down the rung depending on what situation or condition your loved one is in. For the last five years, I’ve believed that family members and caregivers are better decision makers than people like us. It’s up to them to decide what’s best for their family.

Do you think robotics/technology eventually replace trained health care providers?
At a 50,000 foot level, I don’t see it. One of the things we love about this space is this is one of the most authentic marketplaces I’ve ever been part of. I find that compassionate people drive the innovation in this space. I can’t see AI or robotics of any kind replacing that compassion of the incredible people in this space with our older adults. Can it enhance, can it help can it improve? Absolutely. We get excited that our products give caregivers a little more time to do things for their loved ones because when they’re engaged with the Pets, they don’t require the full hands-on attention. So they get 30 minutes to go something for their loved one rather than being tethered due to agitation or memory issues. There will be continued innovation to enhances the interactions or enhance the important work that needs to be done. But I never see it being replaced.

How might AI/robotics help people stay home longer?
If there’s one positive out of the pandemic is that it’s shown a spotlight on the epidemic of loneliness and isolation in the senior population. And all these safety measures put in place to protect older adults are further isolating them. 

We’re working with several state agencies who were fixing this. We’ve been focusing on the epidemic of lineless way before the pandemic. We’ve seen as a problem that’s sort of front and center and one of the things we’re committed to is helping end that in some way. So as the pandemic struck, we leaned into our relationship with several states. They were telling us they had a problem that went from A to Z overnight because we cut off the peer interaction, we cut off the family interaction, and in some cases we cut off the caregiver interaction. We’re seeing that our products are being used and passed out by offices of aging to those in need. We’re part of a program in Florida where you can get a Pet or you can get a tablet. The tablet can help a ton if people with video calling. For others, it’s not going to be 24/7 you’re on this device, so you have the companion pet. In this use case, I think we’re helping people who would otherwise be really lonely or isolated with at least some joy and happiness.

What are some of the ethical questions surrounding artificial intelligence in senior care?
I think at an ethical standpoint [when we started], a lot of the focus was on social robotics and the role it takes on replacing human interaction. We never looked at it like that. Our goal is enhance family connections. We’re not trying to replace human interaction, we’re trying to enhance it. 

I’ve longed stopped defending trying to fool people or convince them it’s a real pet. That’s not the role we play. We’re not trying to fool anybody; it’s a robotic pet. But, I’m a person-centered solution person. If there’s a person with really aggressive dementia and they happen to believe the cat is real, and it helps bring joy and happiness and improve the interactions, I think that’s fantastic. Let’s meet them where they are, and not try to impose our paradigms on them. We’re not living in their shoes and don’t know what the dementia is doing to them. We’re not trying to fool people, we’re trying to provide solution because not everyone can have a cat or dog anymore. Companionship that folks were used to is not necessarily available to them because of their health or living situation. 

What are some of the features on the Pets that make them lifelike, and what went into the development of that?
We were at a big company focused on a very different demographic that didn’t know how to interact with older adults. We went around to nursing homes and independent and assisted living and asked if this was a viable product and two what should it do, what should it have? And what we kept hearing was realism. I had meetings at Hasbro about paw pads and tapered whiskers. But the designers were adamant that if you wanted realism, you had to have tapered whiskers, because a cat’s whiskers aren’t straight. That requires a different machine to create that part of the product. And things like VibraPurr, that’s something the older adult can feel. One of the things they loved about the cat the most, what that when it sat on the lap you could feel it purr. What we heard was there is a tactile sensation that’s calming and soothing and that people loved and when they heard that pur they could feel the actual vibration in their cat. The cat play pattern is completely random, because cats never do what you want them to do.

What’s the difference in a Companion Pet and other robotic pets on the market? 
We leaned into the insight from the furReal friends; when it was first launched the product was a little different than today, we saw 20% were moms not buying it for the 4-8 girl category it was designed for, but for their aging loved one. We were searching for how to really be impactful in the health and wellness space. So that was based on some of the early furReal Friends.

We think play is a basic human need and you don’t grow out of a need to play. In memory care there’s a lot of baby dolls, but even though it’s helping the person, someone walks in and sees their 93-year-old dad with a baby doll in his arms, there’s a visceral reaction to that. And what we heard is that as much as baby dolls serve the purpose in the memory care space, family members and caregivers who didn’t like the infantilization that was happening with their older loved one. And we found there was no reaction to cat or dog or kitten, because people have had pets forever. I don’t judge; I think it’s person-centered so if a baby doll helps you that’s great. But from an acceptance standpoint it’s a point of contention with activity directors and those in memory care who were hesitant to have family members visit with the baby doll present.

Another key difference is affordability. Many of the robotic pets on the market today are thousands of dollars. One of the first entrants into this market was Paro the seal—it’s $6000. It’s a great product and they chose a one to many model. We try to keep the product as close to $100 as we can because we think a one to one model is best.