Hacker News new | past | comments | ask | show | jobs | submit login
A visit from a social robot improves hospitalized children’s outlook (ucla.edu)
47 points by geox on Oct 8, 2021 | hide | past | favorite | 46 comments



It’s sad this is needed in an analogous way it is sad that foster parents are needed.

Yes there is an ideal solution of having lots of volunteers to speak with the children, but ask yourself: how many people reading HN today do you think have volunteered in a hospital in the last year?

The world can be a harsh place, and we should consider whether we are resisting improvement under the given ground truth for the sake of the ideal.


When my FiL had an accident and eventually died in the burn center of a big hospital, we were shocked to find that many kids in the unit had no visitors.

Even their parents struggled with the nature of their injury. We visited daily over 3 months until he passed away and were able to redirect a local charity to support those kids.

The world can be cruel, a little kindness goes a long way.


As long as the robot is human operated I don’t see an issue.

There is a volunteer organization called Radio Lollipop that operates at certain children’s hospitals. They run a radio program for the kid and visit their rooms bringing arts & craft supplies (or sometimes just to chat). If you live near a participating hospital I highly recommend you join as a volunteer. I had an amazing time doing it.

However many of them had to be paused during COVID. A robot like this could help volunteers or paid staff visit kids safely until they can return. It is also great to help them reach rooms or wards they normally couldn’t


I can't be alone in finding it convenient that the department publishing the study receives its funding from a toy company


I believe it. Parents can't be around all the time. Doctors and nurses are busy. Having someone stop by that just checks in is comforting, maybe a social worker or psychologist. I get that the robot seems dystopian, but it's sterile and can be moved out of the way in an emergency. It's probably more bias neutral. For example, I think I talk to people I find attractive differently than people I don't. Not on purpose, but I'd be even worse at managing that bias if I'm stuck in the hospital.

Being stuck in a hospital bed for a week or more sucks. Can't really focus on a book or reflex video game. There's only so much TV that's entertaining. Family is freaked out so you can't have real conversations.


As a computer scientist who makes a living from developing new technology, I hate that many of my peers seem to think the solution to every problem is a piece of technology or some app.

Seriously, this is sad, and feels so cold. As others have mentioned we have service dogs in hospitals to visit kids. And also as others have mentioned, the hospital that piloted this is funded partially by the toy company Mattel, so I question the bias in this study.


It's probably because the social robot is unusual/different/fun for the children. In the future, if everything is done by robot, then you'll see a study which claims "a visit from a social human improves hospitalized children's outlook"


Yeah; I definitely think for kids it has a certain novelty that makes the hospital stay more interesting and enjoyable. I don't think this could scale beyond a fun novelty though; but that is more of just a personal opinion on my part.


I am glad your peers are not burdened by your cynicism and are willing to try new things. If these things do not work, then do not use them. Your cynicism should be reserved for the deployment/productization of these technologies in the face of research that says they do not work. The research itself should be encouraged!

Do you believe that this problem is already solved and not worth the researcher's time? If so, that reflects negatively only on you and I am glad others are striving to do even better.

Of course such studies should be reviewed and critiqued, but it is intellectually lazy to point only to the source of funding without further consideration.


Darn, your reply reads as super harsh.

Also a reply on hacker news is not the place for me to go into a complete peer review of this research. I am merely pointing out that they are promoting essentially a toy and the hospital is funded mainly by a toy company.

Also, no, I don't think that this research is necessarily a waste, but I would caution people from thinking this solves anything meaningful, and also would caution people from using this technology in lieu of real human interaction.

It is a novelty, a fun toy. My cynicism comes from seeing people trying to automate human interaction in seemingly every facet of life. My point is not everything has a technical solution, and beyond a small research project I could see larger scale versions of this being detrimental to patients since it will be used in lieu of other non-technical solutions.


My reply was a deliberately harsh reply to what I perceived as an (undeliberately) harsh/cheap critique of the work presented here.

These researchers spent seven months interacting child patients, some of whom were likely terminally ill, in an effort to make one of the hardest experiences of their lives a little less difficult. They would spend an hour interacting with and playing with the children via the robot. 90% of parents found the experience positive enough to request another visit.

There is plenty of room for valid critique here. Perhaps the kids would have preferred an Xbox. Perhaps the novelty will wear off quickly.

There is less room for critiques like "this feels sad and cold" or knee-jerk reactions about techies trying to solve everything with technology. If you had paused and asked yourself "Who even ran this study?" you would have noticed that the lead author is not even a techie but a doctor! Presumably they have some exposure to the problems these children face.

I was frustrated by your comment because you have allowed a misplaced cynical view to halt any curiosity you might have had about this study. Your cynicism allowed you to immediately dismiss it.

My comment was deliberately harsh. I think you should re-read your comment and imagine how Dr. Gabriel Oland or other members of the team might perceive it.


Alright; fair enough. If I am willing to give criticism I need to also be willing to be criticized.

I stand by my original comment though. I really don't think the team should take my feedback as a personal insult, it is simply my own personal view on it, not a formal critique.

I read the research because I was genuinely curious, and I still think it feels like it falls in line with the old adage "when you have a hammer, everything looks like a nail". The fact that the lead researcher was a doctor doesn't change that for me.

I dislike a lot of projects like this because I have had experiences in my industry where people want to optimize and automate everything. Perhaps this will work, but in my experience things like this fall flat; I am always opened to being surprised though, and I certainly will not be upset if this does end up helping people and my read on it was wrong.


This is interesting, but I'm guessing most of the effect is due to the novelty of the "robot". Most kids have probably seen a tablet before, and not this.


As someone who spent a very solid chunk of time as hospital patient as an older kid getting to witness a lot of younger kids in the same board, anything social-ish improves things for hospitalized kids. A robot is more like a pet or teddy bear than a tablet, so not that shocking a result unless the tablet was locked to some kind of fancy tamagotchi analog.


Rather than compare this robot to caring parents or human volunteers - maybe it should be compared to a teddy bear (or other "stuffed animal" toy), gentle pet tortoise, economy "robot pet" dog or cat, or similar. All of those seem likely to cheer kids up...but lack the "easily wiped down with hospital-grade sanitizer" advantage of the robot. (Or the high tech cred, or the price tag...)

An elderly neighbor of mine recently got a "robot pet cat". After her husband had to go to a nursing home, leaving her living alone. It was a gift from another elderly lady, similarly living alone. My neighbor knows darn well that it's a cheap toy (very limited behaviors, obvious gear noises, etc.). But she still finds it quite comforting to hold and "play with", when the stress (bills, uncertainty, dealing with endless doctors and social workers, etc.) is getting to her.


The experiment's control is a tablet. How about the control being a visit from a dog instead?

Oh, I know: because that's a volunteer effort where both parties benefit, and it costs nothing. And no toy companies are involved.


Because a tablet is something that most people have, and a dog is not, so a dog would be another intervention type.


Really? Where do you get that statement?

According to [1], 63.4 million or 53% of American households own dogs.

And why do you think it's a study within "intervention type = mechanical" ?

The studies I looked at subsume "tablet" within "computer, smartphone, or tablet" so I'm not sure what we're comparing here.

[1] https://spots.com/pet-ownership-statistics/


I am allowed to bring tablets (or phones) into the hospital, but not dogs. I meant "In the hospital".


There are groups that visit hospital patients with dogs. Unfortunately the requirements are pretty strict for a dog to qualify. My dog loves people and would do great but she is not very trainable so she won’t pass any test.

In general I wish we wouldn’t make our environments at work and also hospitals not as sterile and purely functional as they are. I know some people like it that way but from my experience having a few animals like dogs around makes a huge difference for peoples emotional well being.


Indeed. That's the very reason I don't bring my dog for visits (also afraid the kid will get too attached to him since he's so sweet).


I think this concept would work in the elderly population. Including patients with severe dementia. An interesting way to improve quality of life for this population.


I remember in 2013 learning about the robot Paro. It was this little plush seal robot intended to help the elderly with loneliness. It would respond to touch and interact with you. It was kind of like a super advanced Furby. Findings at the time were that it improved patient wellbeing. It was a really cute little guy!

https://www.youtube.com/watch?v=oJq5PQZHU-I


During a pandemic a robot is less likely to spread it


Ouch. As an adult, I’m finding solace in my iPhone because I didn’t succeed at social relationships (and the social fabric is shredded). Like in the movie Her, I write to my personal robots and it gives me just enough affection to get by.

I wish that to no child.

Even less if we spend time manufacturing those little spawns of the devil, instead of spending time with those who are in need.


> “Our team has demonstrated that a social companion robot can go beyond video chats on a tablet…”

How does it benchmark to a caring human?


I know that's a rhetorical question. The opposing rhetorical question is how does it benchmark to no visits from humans and no visits from robots.

The real question is how the hypothetical caring human fares against infrequent visits from a compassion-fatigued and sleep-deprived resident nurse, or a genuinely caring parent who's stressed and similarly sleep deprived from working long hours to stave off bankruptcy and homelessness due to their child's medical expenses. Can they work together?


How does it benchmark to experiencing some stress in a situation where stress is the normal response and learning to deal with it is antifragile.


Kids are also cheered up by puppets. I'm not sure what one would expect. It would work on adults who weren't sick, too.

Also, I feel strongly like I recognize that "robot" and that exactly this same thing was done in the 70s, down to emoticons on a monitor for a face.


Next step: outfit robots with a Primer and send them off to impoverished neighborhoods!


Compared to what? A parent, a grandparent? A medical professional? Robin Williams?


I don’t see this as any more dystopian than having a costumed human in character as Santa or Spider Man visit hospitalized children. For those who do, I think that’s the comparison you have to make an argument for.

Kids delight in plastic objects just because they are shaped like dinosaurs instead of sippy cups. That novelty is important to them. The robot is operated by a human, so the interaction is still rich.

My nephews like me. My nephews would also like me-as-a-robot even more as an occasional thing, whether they knew it was me driving it or not. Given that the intention here is “more visits to sick kids that would not otherwise happen,” and not “substituting existing human interaction with a human-operated robot,” I think it’s a good thing.

Don’t let the perfect be the enemy of the good.


There's a difference between a real person dressing up as a character from fiction, and a soulless robot coming to visit you.


The soulless robot is operated by a person, making it more like a puppet. The contrast in the article between satisfaction with robot visits and “tablet visits” indicates a clear preference for the robot over the tablet. It doesn’t give any detail about what a “tablet visit” looks like though.

The “visited by preference list” probably looks like this:

  Costumed Human > Robot > Tablet > Nothing
I do think an interesting question is whether “random stranger not-in-character” beats “random stranger driving shiny robot.”


Considering they trend 5-10 years behind the rest of the world, I sure will be glad when academia progresses past its 2010-2015 content-mill phase.


I worked on robotics, and I have to say this is sad.


This is akin to language learning. A child can take foreign language courses from kindergarten through high school (~20 years), but in the end they will only learn a stilted academic version of that language. They would be better off spending a single year fully immersed in that foreign country, learning the actual language and associated culture, with all its idiosyncrasies and subtleties.

Anything short of full immersion with actual human-to-human interaction is doomed to be a lesser than alternative.


Let's outsource social care/s


The blackest of mirrors.


Do people working on things like this have a shred of humanity?


They want to make sick kids happier and it appears to work. How is it lacking humanity to make sick kids happier?


Creating pseudo alternatives to human/animal interaction using low effort robots on a planet with 7 billion people is absolutely absurd as a solution to the problem. Children are already starting to live life staring at rectangles half the day. It's one step above hypermedication of children that's rampant in the US


But it may be the only practical solution.

- Those billions are busy doing other things. There is a time problem.

- Even families don't care enough to visit their kids and elderly parents, so there isn't a lot of willingness to take significant action. There is a what people are willing to pay (in terms of time, money, and/or resources) problem.

- The scale of the problem is enormous and humans don't scale. Are there even enough sociable people you could hire in the USA for this?


It is yet another solution to a problem no one posited


Of course they do!

Stuff like this does not replace human interaction, in both meanings.

1) No one "stops" visiting ill children because this robot exist, right? I know someone who is a "clinic clown", who keeps children a little bit happier by clowning around with a red round nose and funny make-up. Stuff like this will not "stop" because there is now a "better alternative".

2) This is in the same way a "solution" like a improvised splint is a solution to a broken leg. Not perfect in any way, not something forever, but it's better than nothing.

And "nothing" is the alternative for many people.

I worked in an elderly home after high school. Many of those people never had any visitors in a whole year, and the facility was severly understaffed, preventing us from "replacing" their families a bit. While I'm hoping that most childs in hospital get more visitors, I'm sure many are very lonely too.

I too would wish for it to be different. But we can't force people to interact with other people, can we?

If robots can emulate human emotions and affection close enough to at least ease the pain for those people, it's better than nothing, no? :-)


On the contrary I think people working on things like this might have a good amount of humanity. It's the pressures to cut humans out for labor costs that is without humanity. Even non-profit hospitals often have administrations that are incentivized to cut costs, which meant that even before the pandemic hospitals were cutting their workforce to the bone. There are currently very few, if any, laws that mandate safe healthcare-worker-to-patient ratios, which means hospital administration can arbitrarily decide how many ways a given medical personnel is expected to split their attention among the sick.

Make no mistake that if we are to blame anyone with this technology we should be blaming medical administration.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: