Robots Teach Autistic Kids Social Skills

posys writes "Wired Magazine has an article showing how the eternal patience of robots lends itself well to teaching new social skills to autistic children. 'The researchers hope that the end result is a human-like robot that can act as a "social mediator" for autistic children, a steppingstone to improved social interaction with other children and adults. "KASPAR provides autistic children with reliability and predictability. Since there are no surprises, they feel safe and secure," Robins said, adding that the purpose is not to replace human interaction and contact but to enhance it. Robins has already tested some imitation and turn-taking games with the children and his preliminary findings are positive.'"

Ummm

[Bombula]

Am I the only one who finds the idea of robots teaching autistic children to be social slightly ironic?

Re:Waste of effort

[provigilman]

Yeah and, you know, why waste time helping ALS sufferers? I mean, they can't really move around unassisted or even speak for that matter. Maybe we should just abandon them too? Of course that would mean no Stephen Hawkins, but hey, he's defective too...right? [/sarcasm]

Autism

[kernel_pat]

I have talked to many autistic children and this is only going to teach them how to mimic social skills rather than solve the problem.

Re:Bizarre.

[Spy der Mann]

Who would have thought that the blind leading the blind could actually work?

Technically, it's an incorrect analogy. It would be more like a guide dog leading the blind. The dog (in this case the robot) is trained to produce specific responses to specific situations, helping the user (the autistic child) to deal with those situations. Now the advantage is that autistic kids CAN learn... so they won't be dependent on the robots forever... I hope.

Re:Ummm

[the dark hero]

Ironic: Yes. Useful: Extreme Potential

Have you ever had to care for an autistic child? I was asked to watch over one for maybe half an hour to forty-five minutes. I wasn't prepared for what was to happen next, but i learned a lot. In fact, the kid taught me more than i could even try to teach him. The boy was about 9 years old, but had the mental capacity of a 5 year old. Sadly, that's probably never going to change. One thing to remember is that some autistic children don't speak. They do things largely based on habit. Kinda sounds less ironic if you have something to relate to in order to assist social interaction with other humans.

It's hard to judge mental capacity in the autistic

[benhocking]

I've known several autistic children, and you might be surprised at what some of them know. Now, granted, there are often other developmental disorders that accompany autism, but what you interpret as reduced mental capacity might be reduced motivation. I knew one kid that had almost no verbal skills and was thought to be very unaware of his surroundings, but when the right motivators were found it was revealed that he knew the names of everyone around him and much more. (Of course, he's still most likely developmentally delayed.)

That said, I agree that the robot could be extremely helpful. For many children with autism it might turn out to be the right motivator.

Re:Waste of effort

[arivanov]

Seconded.

Though you are actually not looking at the GP argument in sufficient depth. It is a popular tendency today to brand even the smallest deviation from the average as autism and try to "cure" it or "compensate" for it. Plenty of lousy, lame and lazy teachers use this as an excuse to avoid children that require individual attention and do not study well in a group.

Many great brains and problem solvers are wiped in the process. Einstein would have been put on Ritalin by the age of 6 nowdays and we would have never had the theory of relativity. Same for Mozart and his symphonies. And I am not going to even mention extreme cases like Tesla who had a seriously bad case.

Re:Ummm

[mooingyak]

As the parent of a mildly autistic child, this sounds brilliant.

It's all about PATIENCE. There are some behaviors that took enormous effort to drill into her. It required us providing consistent gentle reminders that some particular behavior was inappropriate. It does no good to get angry, it does no good to scream, these don't particularly register. It's very easy to get frustrated by the excessive repetitiveness.

Words of caution

[scaz]

While the use of robots with children with autism has generated intense interest from a variety of research groups (mine included), one thing that the media often leaves out when reporting these stories is that these studies are still experimental. There have been no published studies to date that show long-term behavioral change following interaction trials with robots, only one case study that has looked at long-term effects of these interactions, and no studies that have shown any transfer of skills from human-robot interaction to human-human interaction.

Every time an article like this is published, my office and my colleagues are barraged with requests from individuals, families, and educators looking for the robot that will "cure" autism. While everyone working in this area has great hopes for what is possible, we don't yet have any clinical or experimental data to support any claims of effectiveness.

Re:Autism

[Kohath]

And what's the difference between having social skills and being able to fake having social skills?

Re:Counterproductive?

[nelsonal]

It took a much more patient teacher to teach us that the derivative of x2 was x/2 and the integral of x was 1/2x^2+C than the lesson that the integral of sin(x) was cos(x) even though the latter is generally a more advanced concept.

Re:It's hard to judge mental capacity in the autis

[Analogy Man]

There was a very interesting piece on NPR a few months back about a convention for autistics. It may seem strange that they would hold such a convention, but it provided a forum for many incredible people to meet where someone wanting to wander off into a corner to be alone would be understood by others in attendance.

One speaker made a very compelling argument that this "disorder" should in many instances be considered just a different mode of operation rather than a disease. One benefit of the way his brain works is that he can concentrate at a high level on the same stream of thought for hours...something that is very rare in "normally" wired people. One persons compulsive disorder is anothers passion, persistence and dedication.

Sure of someone unable to engage the world around them is going to be an absolute nightmare to raise and this research may be a means of greatly enriching these people's lives. A "cure for autism" might not be welcomed by all those afflicted however?

Tutoring

[Das Auge]

In high school, I spend one period per for a semester tuturing autistic students. Most of them feel into one of two categories. They were almost either almost "normal" or barely functioning. it took a careful, thoughtful, hand to teach them. I can't see how a robot (with today's technology) can possibly do even come close to that.

I guess you could get a person to monitor the robotic tutoring, but you'd end up with a human watching a robot teach a human. Which would be a waste.

I don't care if the child is autistic or not, I'm not pleased with the thought of robots teaching children. Not only are you robbing them of learning social mores, but you're also telling them that they aren't important enough to be taught be a human. Just one more mechanical babysitter (along with computers and television) in their lives.

Re:Fixing Autism?

[Anonymous Coward]

Just because someone lacks a normal ability in one area doesn't mean they automatically get a strength in another area. Sometimes a disability in just a disability.

It's a very valid point

[benhocking]

OTOH, it should be noted that, more than likely, most people with autism at this convention were at the higher-functioning end of autism. For them, the autistic tendencies can confer real benefits. Most of those with autism are not necessarily so lucky.

One-on-one instruction

[benhocking]

I guess you could get a person to monitor the robotic tutoring, but you'd end up with a human watching a robot teach a human. Which would be a waste.

Or, you could have a human watching ten robots teach ten humans. Each of those humans get semi-individualized instruction with the benefit of a human instructor if one is needed. In my wife's school, most of the children with autism get a teacher all to themselves for most of the day. The exceptions to this are the more advanced children who work two to a teacher.

Also, many of these children work better with computers than they do with humans. That doesn't mean that humans should be taken out of the picture, but that the computer/robot can do what it's best suited for, and the human instructor can do what he or she is best suited for.

The Naked Sun

[kalirion]

I wonder if this is how the Solarians started....

Re:Counterproductive?

[Chris Burke]

Um, maybe because it's easier to learn basic social skills from an infinitely patient unflappable robot, then apply and refine those skills with finite-patience irritable humans, than it would be to try to learn social skills from the irritable human in the first place?

It's the same concept as learning to throw and catch with your dad just by tossing the ball back and forth just for fun, then practicing with a team, then playing in a little league game, rather than trying to learn to throw and catch by playing little league games where if you make an error you get yelled at by your team and coach.

It's all about the progession of learning. First teach what social responses are expected, then learn how to use them in the environment with people. By decreasing the amount of patience granted you increase the functional level of the autistic person -- from the infinite patience of a robot, to the nearly infinite patience of a parent, to the extensive patience of a teacher or friend, to the very limited patience of a human. That itself is learning, and it isn't counterproductive to lower the bar at the beginning, any more than it is counterproductive to teach kindegartners to only read books accompanied by pictures.

Re:Autism spectrum

[Nemetroid]

Yes. I know two persons with autism. One of them is in my IB class, studying Physics and Mathematics at Higher Level. The other one, while not a straight A student, is good at communicating socially. So, to label autistic children as mentally deficient is not right since it covers a huge spectra.

Some things

[RockoTDF]

I hate to say it, but the vast majority of slashdot posters have no clue what they are talking about when it comes to this stuff.

Sad as well since so many important people in science and tech who they hold in high esteem likely experience some form of autism or aspergers. I once read that "Computers were designed by people with aspergers for people with aspergers."

Re:Autism spectrum

[Evil Cretin]

I used to work with an autistic person, at a large technology company. Whilst socially awkward (prancing around wildly in corridors etc), he was one of the most intelligent people I've met, and although no-one ever saw him doing any work, he always finished his workload far ahead of schedule. Autism != mental deficiency. It's better described as a social deficiency.