Chimpanzees Beat out Children in Reasoning Test

caffeinemessiah writes "The New York Times has a story on how chimpanzees seem to exhibit a better understanding of cause and effect than human children. While training chimps to perform a routine task with redundant steps, the chimps were able to figure out and eliminate the redundant steps, while the human children routinely performed them despite their evident uselessness. It says something about the way we learn compared to chimps and should be interesting to cognitive scientists and those interested in computational learning theory, at the least."

Comments

Experiment Proposal

[students]

I'd like to see another experiment done. Suppose, hypothetically, that a chimp showed a human child how to solve a puzzle, inserting unnecessary steps. Would the human skip steps more often if taught by a chimp than by another human? If so, it would show that what matters is if the species of the teacher and student are the same, not the what species the student belongs to.

Re:Experiment Proposal

[iocat]

It would probably eat the human child because chimps are vicious wild animals, not the cute, cuddly animals people think they are.

Also, the fact that humans are more likely to do unnecessary steps may indicate a greater willingness on the part of humans to experiment, which is why we have computers, and keep chimps in cages, and not the other way around.

Re:Experiment Proposal

[dajak]

In the nineties we changed from traditional rote learning to a revolutionary discovery-based teaching method in secondary schools in the Netherlands. My professor in Cognitive Psychology was extremely skeptical about this and had his students organize learning experiments with children and students to demonstrate how badly they performed at 'discovering' anything other than a simple correlation or inverse correlation. He was right: the method turned out to be a disaster for the generation of children exposed to it and we are now returning to rote learning.

What differentiates mankind from animals is not the kind of skills someone like Newton possessed, but our capacity to transfer acquired knowledge to others, and in particular future generations, in an efficient way. Language plays an important role, as well as the capacity of young humans to imitate seemingly pointless customs without a direct reward.

An important difference between the children and the chimps is that the children live in a magical universe with remote controls and mobile phones: they are used to learning how to operate devices they don't understand.

Propensity to experiment is highly overrated. Cats are for instance typical opportunists that 'experiment' all the time when solving mazes. They simply don't expect the rewards in a maze to be in the same place each time: they will keep checking random dead ends in the maze to make sure there isn't a reward there this time. This doesn't make them 'intelligent' from our point of view, since they fail miserably on the kind of experiments rats and mice excel in, but apparently it does work very well for catching small rodents. To appreciate the superior spatial intelligence of cats you have to see a hungry cat close in on a prey, positioning themselves to to intercept it on the only escape route left open.

Re:Experiment Proposal

[bleckywelcky]

I was thinking about something even better as I was reading the article. Have a chimp demonstrate the redundant process to another chimp. I was thinking that the researchers might be overlooking some cross-species assumptions they were making. Basically, if you saw another species carry out a process, I think humans are more likely to look at the goals of those actions and attempt to achieve those goals. But if you see another human carry out a process, humans are more likely to imitate since they think the other human might know what they are doing. Whose to say the chimps aren't doing the same thing?

A little bit biased, isn't it?

[ReformedExCon]

Human babies have a prolonged childhood. Whereas a chimpanzee may be considered an adult by age three, humans may not even reach (emotional) adulthood until well into their 30s. So it seems a little disingenuous to compare chimpanzees to human babies when the rates of growth and maturity are so different.

Re:A little bit biased, isn't it?

[ozmanjusri]

politicians seem to have no grasp on cause-and-effect regardless of age.

No, that's just ordinary sociopathic behaviour. Politicians are aware of cause-and-effect, but don't have emotional reactions to the consequences.

You may be right in that being the difference between the children and the chimps though - the child's goal may have been to please the experimenter, while the chimp's goal was to get the prize

I don't think this study shows just learning.

[ebob9]

I don't think this study shows learning processes as much as the poster says it does.

I think the real key here is communication and culture. The Chimps were 'shown' how to open the box to retrieve the food. The children were also 'shown', and told that they could do whatever they thought neccicary to retreive it.

I would think that upbringing and communication would have a big impact on what the kids will do. Lots of times, when an 'adult' shows a child how to do something, they will take that as the 'correct' way to do it, and not deviate from that - because if there was another way to do it, why would the 'adult' show them incorrectly? Kids that have been taught or had the experiance to question authority would be more likely IMO to skip unneeded steps.

However, a chimp most likely does not have this 'follow what the adult says' mentality, so it seems obvious that they would do whatever is the easiest to get the desired result.

Re:Human survival trait

[BewireNomali]

Right. Really good point.

I had a discussion with a friend of mine about religion. She was raised religious, and while an athiest now, she was happy to have been raised religiously. I asked why; she responded that the religious foundation answered questions she would have had (albeit falsely) about God, death, universe, etc. and thus eased her mind about them until she was mature enough to decide that it was mythology to her. In other words, she did exactly as you suggested, emulated a successful culture dynamic too complex for her to understand fully.

We all do it as humans. It's what religion is. Do this because I(tm) said so.

Good point.

Authority

[koreaman]

The children did the task exactly as it was described because the scientists were authority figures and their parents trained them that way. The chimps don't give a damn.

This view of authority is, however, a double-edged sword and could be dangerous.

Children get REWARDED for imitation?

[ClickOnThis]

[Disclaimer: I have no credentials in behavioural psychology, aside from what I have learned by reading and by experience as an amateur trainer and caregiver for several dogs, including two German Shepherds.]

Practically from birth, humans are conditioned to imitate each other, so perhaps it's no surprise that the children absorbed and retained the "ritual" portions of the tasks. Psychologists call it operant conditioning: when you reward a certain kind of behaviour, it tends to occur more often; if you don't, then it tends to extinguish. I wonder if chimps are more goal-oriented because their sense of reward is more focused on the final result rather than following a number of ritualized steps, at least initially. In short, perhaps young children are more conditioned to imitate, as well as being more capable of doing so.

Re:Wal*Mart Kids

[garcia]

See, this is what we get when our schools aren't allowed to whoop ass anymore.

Schools don't need to "whoop ass". Parents need to be able to "whoop ass" w/o their kids knowing (or even having) the ability to "call social services". Parents should be able to practice a little "tough love" as long as it stays as a red ass and doesn't cross into black eyes.

The problem these days is that "timeouts" are used instead of the *threat* of the belt (the sound of a belt coming off a pair of jeans *still* bothers me to this day and I think it was used on me less than 5x).

Oh and all stores need to follow the lead of the guy that put up a *well written* sign on his door that asks that parents ensure that their children behave. Those that were part of the "backlash" against it need to seriously rethink their parental abilities and theories as they may not be the perfect parents that their precious parenting books tell them they are.

Maria Montessori documented this 100 years ago

[michaelmalak]

Here's the briefest summary of Maria Montessori's four planes of development that I could find via Google. The first six years are known as the "absorbent mind". The "reasoning mind" doesn't start until the next six years (ages 6-12). The kids in TFA were ages 3-4. No big surprise they couldn't reason and abstract.

Now ask a chimp to have a vocabulary of 10,000 words.

Maria Montessori's major insight was that there are "sensitive periods" for various developments -- an age to walk, an age for toilet independence, an age to talk, an age to learn practical life skills, an age to acquire knowledge, an age to self-consciously play a role in human society, and an age to develop a profession. If a person does not learn and develop a skill during the sensitive period, that person will struggle with that skill until death.

Three and four year olds aren't ready to reason. Teach them to read, to sew, and to cook instead.

chimps & sign language

[weierstrass]

All i can say about this is that in Steven Pinker's book, 'The Language Instinct' he reckons:

some (behavioural) linguists said they got a bunch of chimps to communicate using sign language. the chimps were using sentences, combining words to build more abstract concepts etc.
they were doing this to try and disprove the ideas of Chomsky and Pinker and people that language is a builtin ability unique and essential to the human brain.

like what you seem to be suggesting above, that chimps lack the ability to make the requisite sounds for speech, but nothing else in the way of thought or language skills.

but:
Pinker and his cohorts reckoned the chimps were not really using language, they imitated some key words, but didn't originate their own, the researchers were very lax about what they accepted as a sign, etc.
they of course had their own agenda to push

but if anyone did do some proper communicating with chimps, i don't know about it.

Re:chimps & sign language

[Anonymous Coward]

There aren't many people who have ever originated their own words. I don't recall ever creating a new word. I only repeat words that have been told to me by other people in one form or another. The only thing I do is put them in a different order. How is this any different than what the chimps did but on a much larger scale?

Re:chimps & sign language

[aussie_a]

the researchers were very lax about what they accepted as a sign, etc.
they of course had their own agenda to push

While research bias (either for or against chimps communicating) is a problem that is difficult to overcome in such a strong issue (for many), I have read quite a bit on the successes. I was referring to an instance where chimpanzee's (or another primate) did create words. The example I remember is "bad+dream" for nightmare.

they imitated some key words, but didn't originate their own

Humans have the "inventing words gene," while I believe other primates don't. But that isn't a bad thing (IMO), as it allows us to continue to understand them. If they did invent new words, they would have to teach us, and their ability to teach humans (they are, after all, not equal to our intelligence) could be limited.

Having said that this article says that it's quite possible bonobo's (a type of chimpanzee) do create verbal sounds for specific things, which I presume they've invented. I don't know if it is true that they are verbal "words," but it does bear more research.

However I don't see their inability to create words as them being unable to learn language. This page (it was only a quick search, info may be a bit suspect, but it seems fairly valid and jibes with what I've read in the past) has info on both success and failures. Why I like it is because it outlines those against the results proving language's opinions, as well as those opinions who are for it. One man called Herb Terrace doesn't believe the results so far are indicative of language aquisition, but merely "aping." Some of his complaints are:
* That the apes were were performing rote memorization tasks similar to pigeons who are taught to peck at colors in specific orders.

This I take issue with, because the page earlier shows an ape taking a word in one context "more" and using it in others. It isn't a simple case of "sign X always follows action Y" but instead, reasoning what sign X actually means, and applying it in other situations.

* Primates only signed in order to please their trainers, not for the personal gratification of using the signs.

I take issue with this, as many sources I've read say apes do spontaneously speak with each other. Having said that, it appears Terrace's complaints were actually made a few decades ago, and that research since then has proven him wrong. More info here

* A primate might learn to connect a sign with food and reproduce the sign through simple conditioning, just as Pavlov's dogs were conditioned to salivate at the sound of a bell.

To be honest, is it possible to prove that human children don't speak for the same reasons? I don't think so. Think about it, when a baby is learning to speak, we heap attention and treats on them. The Pavlovian method of teaching requires this to begin with, which is then removed and the taught actions continue regardless. A problem with detractors of ape speech is that they often ask questions we can't answer when it comes to humans.

but if anyone did do some proper communicating with chimps, i don't know about it.

Unfortunately I to, do not know if anyone has. The article I linked to before, does suggest that researchers are doing their best to communicate properly with apes, but it's a hot issue for those involved. I believe current research is very indicative, but it can't silence critics yet. But I do believe it's enough (or at least enough to warrant a much more structured research program with a definitive goal of giving apes more rights) to say "y'know. Maybe we should reconsider how we treat them. Perhaps there is a better place in our society for them."

Re:chimps & sign language

[a_real_space_cadet]

Speaking from experience, you're wrong about "Babies don't usually get any tangible reward simply for saying a word or two". Babies are HEAPED with attention for their first nonsense syllables - and it's definitely not the same kind of attention they get by crying. It's eye-contact, smiles, laughs, high-pitched repetitive dialog. When the parent suspects that a sound is being used meaningfully, the attention is heaped even higher.

Re:chimps & sign language

[Sigg3.net]

Didn't that Amy gorilla that learned sign-language say "baby in my drink" when her toy doll fell in the food tray?

Of course, that's a long evolutionary step from: "Need another drink, baby"

Re:chimps & sign language

[Liam Slider]

Nobody has come up with a quantified definition of what human intelligence is, much less animal. Pretty much everything that is easy to test has been observed in animals, both in the wild and in the lab. For a long time it was tool usage, remember? And then it was creating tools, not using found objects, and now it's just "really complex tools that animals can't make".

That's because there's a bias among many humans, including a vast number of scientists, which is that humans and animals are somehow two different things. That we are somehow special, different, unique. We are just another animal. One of the smarter animals perhaps, but just another one of the many beasts on this world. We've got some neat, hyper-specialised abilities that evolution tossed our way in order to survive in this otherwise pathetic form....like not just tool-use or making (which many animals have), but tool-improving. We're also built for projectile weaponry, it's evolved into our eyesight, our reflexes, our strength level, and our complex brains which it takes to manage hunting via projectile weapon (be it spear, atlatl, bow, or firearm). And we're pack hunters....and complex hunting in a pack, using projectile weapons...you better damn sure know how to communicate with your packmates. It's nothing special about us, nothing secret....simple survival traits as applied to one animal.

The problem with the bias though, is that it causes people, including scientists, to make an assumption. That other animals cannot be almost as intelligent as we are, or think in ways that we do...or even think at all, or communicate on any meaningful level. They're just "apeing us" because they're "just dumb animals." Dolphins may be as intelligent or nearly so (or more so) as we are, but in an utterly alien way...yet you'll find few scientists with the guts to say so, even though there is a massive amount of evidence to back it up. Why? Because they are animals, of course!

Re:but children will become adults

[mattwarden]

So then instead of discussing intelligent design ad nauseum, we can argue about whether humans were really given domain over all animals. Great.

A slightly different perspective...

[MissingDividends]

I think this comes back to how we raise our children... (Well actually how you raise your children since I'm still one...)

I think that the whole "don't question, just accept" way of thinking is totally absurd.
Do you have any idea how annoying it is to hear "because I said so" or "What do you mean 'Why?' It just is!" from an adult?
What's worse are the "I'll tell you when you're older", "You wouldn't understand", and "You don't want to know that"...
My personal favorite is "It doesn't matter"...

My parents were great; they avoided these types of thought-quashing over-used "no-answers", but many of my teachers (even the good ones) at the elementary school level (USA) got so sick of me (it wasn't just me, but I did ask a lot of questions) asking questions that they decided that 'It doesn't matter' what I want to know, it only matters what the curriculum says to teach (another whole rant), that if the explanation wasn't simple, that they should make something up.

I'm not trying to brag, but even in 1st grade, I was really good at math. I have 3 math teaches in my family, and that helped a bit, but teachers often tell 'little lies' because it doesn't really matter... Do you have any idea how many recesses I sat trying to figure out why you couldn't divide by a fraction or decimal?! I certainly didn't know for sure what you would get, but it only makes sense that if you have 8 marbles and you put them into 2 piles you have 4 in each pile (8/2=4) and then you put them into 1 pile, you would have 8 in each pile (8/1=8) and if you were to keep cutting the number of piles in half, the number of marbles would double, which is true (8/(1/2)=16), but no first grade teacher is going to try to explain the fact that multiplication and division are the same thing when the teacher just spent days trying to teach the kids not to confuse them.
I'm going to go off on [another] tangent... the teachers don't call them little lies; they call them white lies. Does anybody who uses that term have any idea how incredibly racist that is?! It implies that black lies (big lies) are very bad; while white lies aren't so bad...I don't think I need to elaborate any further...

When I started writing this, I was going to give an example from each grade, but now that I'm done with that, and re-read it, I've decided not to bother as 99% of readers will have given up by now...

The other problem with the way kids are taught to reason is you* spend so much time telling kids to do it 'the right way'... in reality, kids are told to do it your way. They are told that their way is always wrong.

-You have to color in the lines. Why? Because I said so.
Is there any reason to make kids color in the lines? Can you think of one? Yes? Why do most parents/teachers/etc. refuse to explain it to their children? Is coloring in the lines a life-skill? No, but it does help to teach motor skills. There are other reasons, but that's the only one that makes any sense to me...
Isn't it more efficient to just scribble?
Isn't it actually stifling creativity to teach kids that you have to color each object one color?

-Walk in a straight line between classes. (Not sure if they do this most places, or if it's just a regional thing)
I can't count the number of times I questioned it in the first year or two, but after being yelled at because I asked so many times, I just kind of accepted it.
(Just for the record, 'It doesn't matter' isn't an acceptable answer when a kid asks a question multiple times. Obviously if they keep asking the question, it matters to them)

Again, I was going to ramble for a bit longer, but I think I've ranted on the school system and on conformity enough for one post...

It seems that you don't hae kids.

[hummassa]

Babies don't usually get any tangible reward simply for saying a word or two. They may get some attention, but they could get that far more effectively just by crying. You should never do that. When a baby is starting to speak, you should ignore it most of the time it cries, and give him reward in attention when he speaks; that way, it'll develop speech faster.

Not a coincidence ...

[hummassa]

If you don't do that, they will not start talking for many months. I know of a toddler that started talking at almost 2yo (as opposed to 8-18 months) because everytime the said "ah" and pointed to something, his parents gave it to him.