Slashdot Mirror
Why Toddlers Don't Do What They're Told
Hugh Pickens writes "New cognitive research shows that 3-year-olds neither plan for the future nor live completely in the present, but instead call up the past as they need it. 'There is a lot of work in the field of cognitive development that focuses on how kids are basically little versions of adults trying to do the same things adults do, but they're just not as good at it yet. What we show here is they are doing something completely different,' says professor Yuko Munakata at the University of Colorado at Boulder. Munakata's team used a computer game and a setup that measures the diameter of the pupil of the eye to determine mental effort to study the cognitive abilities of 3-and-a-half-year-olds and 8-year-olds. The research concluded that while everything you tell toddlers seems to go in one ear and out the other, the study found that toddlers listen, but then store the information for later use. 'For example, let's say it's cold outside and you tell your 3-year-old to go get his jacket out of his bedroom and get ready to go outside,' says doctoral student Christopher Chatham. 'You might expect the child to plan for the future, think "OK it's cold outside so the jacket will keep me warm." But what we suggest is that this isn't what goes on in a 3-year-old's brain. Rather, they run outside, discover that it is cold, and then retrieve the memory of where their jacket is, and then they go get it.'"
Good stuff. I think a lot of parental frustration comes from completely forgetting what it was like to be a kid. The more we learn of measurable differences in functioning between children and adults, the better. Ingrained beliefs can only get you so far.
I think this research is meant to show a couple of things of import that you are seeming to gloss over in your criticism.
For one, the difference between a lazy teenager ignoring what their parents told them and a toddler doing the same thing is that a lazy teenager IS choosing to ignore their parents - there is nothing different going on in their brains, they just don't want to do what they are told.
A toddler, on the other hand, literally CAN'T do what they are told in certain instances, because they don't have the same thought process that adults have (which is what this research is trying to show). It's not that they are choosing to ignore their parents, they just don't have the reasoning capability at that age to comprehend complex conditional statements like "When I tell you it is cold outside get a jacket"
I think the point of the research is that many parents expect things from their very young children that are just not possible. They think their kid is being stubborn or misbehaving when it is just developmental. So many parents get frustrated and angry at their child when they should just realize that they just have to wait for the kid to grow up a bit.
it makes one ponder how one approach the development of AI's to.
sounds a bit like they are building up a bayesian database of conditions and actions, going more and more specific over time.
like say how cold at first will just be a generic sense of temperature thats uncomfortable (thanks to it driving the surface temperature of the outer skin below whats healthy for the cells that makes up the skin). then later one add specifics like snow on the ground, ice and other indicators. as more of these shows up, one get a stronger sense that its cold outside, and that again triggers conditioned reflexes like wearing thick clothing.
so, to turn this over to AI research, the approach may well be to start with a blank database and a collection of sensors and outputs. then one pile on a generic bayesian filter, and leave it running.
comment first, facts later. http://chem.tufts.edu/AnswersInScience/RelativityofWrong.htm
Maybe so, but weak empirical evidence. Even if you're completely accurate in describing what you see, and that assumption is often a stretch, your conditions are likely not controlled enough to isolate anything in particular--and it may conflict with what somebody ELSE sees, which opens a completely different can of worms.
It sounds like in your line of work, simply knowing that the blueprint you got handed won't work in the real world is enough. That's perfectly fine; everybody is concerned with different particulars depending on their own perspective. What you have is a conclusion: "No, you're an idiot. This is faulty." From your perspective that's important. From a scientist (or engineer's), it's a starting point: "Why is this faulty? What can we learn from it? How can we avoid the same mistake later?" Neither of you are wrong, neither of you are wasting your time, but at the same time if you two swapped positions everything would likely go to hell pretty quick.
Anecdote and casual observation are great things to direct us on what we need to study rigorously; they're not a study in themselves.
True, but that's not what this is. Hao Wu basically said (paraphrasing) "parents have known this for ages, if scientists could get any they would have known too!"
Aside from being a bit of a douchebag, his statement isn't particularly rigorous. Parents have known WHAT for ages? That children don't listen? That little kids have particular trouble listening? That's spectacular, and it's a good jumping-off point for exactly the kind of study that was done -- but it's not particularly meaningful in itself. I noticed the sky looks blue, too; that's meaningless as well. Somebody coming along and telling me about white light and wavelengths and giving me the reason WHY it's blue can be important. It chains a statement like "the sky is blue" into any number of potential discussions ranging from anatomy to physics to meteorology.
Knowing that little kids have trouble listening is interesting, and frankly even people without kids have observed that (making the little pot-shot comment about scientists not having kids distasteful,) but what's more interesting is to know WHY--the study seems to be pushing the idea that it's literally a functional difference in their brain. That's cool. Can we do anything about it? That might be useful. Why does it happen and what changes as they age that makes it stop? That might be useful too, in any number of applications and particularly for people who have any sort of learning disorders that we might find have similar physical causes and might respond to similar treatments. Is this just a lack of life experiences, or are we literally altering the way the brain works as we get older?
What your parent poster said was correct: Science is necessary to validate our observations because so many things we have "known" to be true have turned out to be false. I'm not big on name calling, and wouldn't have taken that tact myself, but saying that science wastes its time by studying things we "know" does seem illogical at best.
Yeah, science is all about gut feelings. Why bother researching anything when we already know what the answers will be. We already know God created the universe in 7 days, why the hell are we wasting billions of dollars on astronomy, biology and physics?
So children learn by DOING, I get it.
That's a nice summary, but can you describe the cognitive mechanisms by which they "learn by doing" and how that relates to brain development? I bet you can't without doing a study-- at least not in a way that provides anything but conjecture.
Personally, I think that is rubbish.
The pain response has been around for quite a while and is designed specifically to say to us... "That thing you just did... it was dangerous and damaging, DON'T do it again!!". I cannot believe that toddlers are somehow hardwired *not* to follow this piece of sensory advice. I am not a student of this subject, but it just makes good logical sense that there is a part of the brain (active at birth) that does the job of avoiding the repetition of actions that previously generated a painful response.
For this reason, I support so called 'corporal punishment' as a tool for parents to hijack this process to teach kids to avoid behaviors where the end result might otherwise not simply provide a quick 'sting' (like running out into the road), or behaviors that break more complicated rules (like stealing). You certainly cannot reason with children this young and expect them to understand, but you can hijack a basic evolutionary mechanism and use it to your own (and the child's) advantage.
Of course, there is a huge difference between a quick smack to the bottom to instill a sense of danger that is mentally linked with a given action and actually beating children in a way that causes lasting damage. The former is effective, proactive parenting, the latter should be punished to the full extent of the law.
No, they're not. They "merely" apply science to specific well-known problems.
I take offense to that. Although a lot of engineering involves solving well-known problems, there's a good deal of "never-before-done, at the conceptual stage, we're not sure if it's even possible" problems. Yes, we still use well known physics in the design process, but calling us "not scientists" is a little bit like saying modern mathematicians are not mathematicians because all they're doing is applying well-known math to solve their new problems.
Good engineers apply the scientific method in their design process. When creating something nobody has done before, they examine previous work, they construct a "hypothesis" of how to best solve the problem, they perform tests and simulations to make sure their assumptions are correct, and then they analyze the data, draw a conclusion (create a plan), and build the thing.
if we went by your standard of evidence it's doubtful that our collective scientific knowledge would actually have gotten far enough
I agree with you in principle. You can't take anything for granted, common sense is often wrong. And that applies in engineering a LOT.
If we went by your standard of evidence, we would consider there to be a mountain of evidence that the Sun goes around the Earth. Nowadays it's easy to see that it's the other way round
However, you picked the worst example ever to make your point, because you just used a "common-sense, everyone thinks this is right, but technically it's not" example. It's not necessarily "wrong" to say the Sun goes around the Earth. It's inconvenient for calculations because the center of mass of the earth-sun system lies inside the Sun. It doesn't mean that you can't come up with an elaborate mathematical model with the Earth as the reference center of the solar system (and it has been done), it just means that you'll be doing too much damn work.
There's no absolute reference points in the universe. Picking the Sun as the center of the solar system is the equivalent of using the cylindrical coordinate system instead of the cartesian one for problems that make sense. Things get a whole lot easier, and the math is way simpler and more elegant.