Friday Grey Matters: The Story of Nkisi

tenkss

Good post!!

Man. do I wish George knew 700 words and could use them in coherent sentences.

In the conversation about the car, N'Kisi sounds EXACTLY like a four-year-old kid!

Amazing.

Hey, but maybe parrots ARE psychic, which would make them even more amazing than they already are. ;)

So glad I found your blog!

Doing wordplay and forming opinions with a brain that size? Sorry, but there's nothing going on here that can't be explained by

1) repetition
Notice how both Aimee and N'Kisi love to talk about how 'cool' things are? N'Kisi says "Wow!" right after Aimee does. And how many times has Aimee said "Look!" or "Listen!" to N'Kisi before this conversation started?

2) the anthropomorphic fallacy
When you hear something that can repeat human words so convincingly, it's natural to assume that they understand you. There is no evidence in this text that it is so.

3) massive projection
Carers get close to animals, and tend to interpret and reinterpret the conversation to maximise 'hits'.

4) mental illness
People who do language work with birds in particular tend to go off the deep end. (No offense, Shelley.) While I can't be certain with Aimee, it sure seems to be the case with the owner of Victor the Talking Parrot, who was convinced that Victor's garbly speech was the fulfillment of prophecy (both Biblical and Nostradamus). Psychic indeed.

It's absolutely vital to exercise at least a portion of critical thinking when dealing with animal language experiments. People thought that Clever Hans could use language too. The illusions are so compelling that it's easy to trick yourself into thinking animals can use human language when in fact they cannot.

One should not dismiss the psychic abilities that are observed with some pets. When my wife leaves from 25 miles away to come home(and no, it's not always at the same time) monkey dog becomes agitated and asks to go outside to wait for her. We got scientific about it and watched it happen too many times to discount. We're not going to make it public due to people like fontor who wouldn't admit that anything exists outside of their own narrow view of this amazing world we live in.

There are only 2 words I'd like to hear Bush utter...

"I resign"...

Amazing story..

Jeffery, you're probably a good guy, but you know squat about the kind of person I am.

I'm sure you're convinced of your pet's abilities, and if you can demonstrate them in a statistically valid and non-anecdotal way, I'll do everything I can to help you get the word out. That, I think, would be a lot more constructive than name-calling.

What do you say?

Fonter - what kind of science do you have on your side?

Linguistics.

Science is a long narrow box into which all knowledge and experience gets funneled. Hypothesis: Storytelling is unique to humans because only humans have a fully developed neocortex. Because that is true, the parrot's ability to "remember" the car trip and create a story about going in a car, including the professed desire to renew the experience in the future, is a hoax. Just repeating and mimicking the trainer. End of story.

BTW, according to science, ALL psychic studies are flawed automatically because paranormal experiences don't exist. Assuming the flowed result before doing or evaluating the experiment is the only way to keep the box narrow enough. If a peer-reviewed publication isn't admissible, then obviously only the personal evaluation of the critic is ... and of course, since there's not enough time to review and analyze everything, it's OK to dismiss unsubstantiated "silliness."

When my wife leaves from 25 miles away to come home(and no, it's not always at the same time) monkey dog becomes agitated and asks to go outside to wait for her.

Most likely responding to (unintentional) cues from you. You've heard of the Clever Hans effect, yes?

In my experience, a lot of human interaction is little more than patterned responses to cues, verbal, physical, and chemical. I'm sure we've allinteracted with individuals who were demonstrably incapable of independent, original, or creative expression.

Rather than scoffing at examples of animal-human language-based interactions like this one, we might want to look more closely at the similarities between animal communication and human from this angle: seeking not to discern how animals are like humans, but rather how animal-like our communications are. And to ponder then what "real" differences between ourselves and the rest of the aimal kingdom might turn out to be. Ignoring point of view got us where we are today in large part, and only our egocentrism makes that a success story.

Oh, and btw, Shelley--great,thoughtful site!! I found my way here from a link on Daily Kos--am adding you to my favorite blogs forthwith! Thanks!

I figure I should weigh in on some comments here.

Fontor:

"Doing wordplay and forming opinions with a brain that size?"

One thing that often astounds me in neuroscience is "neural economy." Learning and memory have been observed in honeybees; a phenomenon called the serial position effect which is common in mammals, in addition to their abilities to relay complex positional information to hive mates. Their brain is the size of a pintip. A parrot, who's is the size of a walnut, can be capable of learning and complex behavior if an insect can.

"Sorry, but there's nothing going on here that can't be explained by

1) repetition

Indeed, repetition is important in any type of learning. Children learning language require this, and feedback. Parrots, including my own go through a babbling phase similar to human children. In addition, a "critical period" of word learning has been observed in parrots--wait too long to expose them to words and sounds they will never learn. This phenomonon has been observed in much "stupider" birds such as songbirds, and finches as well (in reference to learning their bird song, not words.)

2) the anthropomorphic fallacy

This works both ways. It is easy to dismiss animals' abilities by saying its absurd to attribute human characteristics to animals. The fallacy here is the belief than a certain characteristic is only possessed by humans. Before language learning, it was problem solving. When animals were observed solving problems, the next hurdle was tool usage. Woops! They can use tools. Now its language, or creativity, etc. These are artifical hurdles. While I'm certainly not suggesting that a bird has language abilities equivalent to any humans, we have all evolved from common ancestors in similar environments and it is certinaly not impossible that bird have also evolved complex communication and intelligence.

3) massive projection
This can be ruled out when the "test" is being delivered by a novel experimenter, or if the handler/bird are seperated by a wall (communicating by microphone). a lot of the work by Pepperberg with Alex has been done this way, similar to the way that the bonobo Kanzi was tested.

4) mental illness
People who do language work with birds in particular tend to go off the deep end. (No offense, Shelley.) While I can't be certain with Aimee, it sure seems to be the case with the owner of Victor the Talking Parrot, who was convinced that Victor's garbly speech was the fulfillment of prophecy (both Biblical and Nostradamus). Psychic indeed.

I also internally groan at the "psychic" claims! But I think it is unfair to make blanket statements regarding an entire field of people. No one who believes a parrot, or a person, is quoting prophecy should be taken seriously. But many "great minds" of the past have only had it half-right at best. The issue here is that this is uncharted waters, and birds deserve to be studied right along with any other species. The fact that they exhibit such flexible mimicry and communication is an added bonus to the research.

It's absolutely vital to exercise at least a portion of critical thinking when dealing with animal language experiments. People thought that Clever Hans could use language too. The illusions are so compelling that it's easy to trick yourself into thinking animals can use human language when in fact they cannot.

I completely agree that scientific rigor is the key. This is of course paramount in any scientific endeavor. While people thought Clever Hans could *understand* language (not use, lol), obviously this was quickly disproven. Pepperberg has been studying Alex for 30 years, and has stood the test of novel experimenters, new environments, etc which has won her grants, tenure, and a home at quite a few prestigious universities. The fascination here is obviously intoxicating. How can a species with such different neural architechture comprehend language? In humans, this is accomplished through Broca's and Wernicke's areas, which birds do not possess. But, they do possess areas which, if lesioned, destroy their ability to learn and/or reproduce bird song similar to aphasias. These structures are, by coincidence only, perhaps used in similar ways as our distinctly different human ones. However, what interests me most is the flexibility of neural structures to adapt to the envionment around them: birds aren't "using language" because the evolved that way; the structures which allow them to interpret and learn song are flexible enough in development to accomodate our mode of communication, perhaps. Just my perspective.

And to falcon, quote below:

"Science is a long narrow box into which all knowledge and experience gets funneled. Hypothesis: Storytelling is unique to humans because only humans have a fully developed neocortex. Because that is true, the parrot's ability to "remember" the car trip and create a story about going in a car, including the professed desire to renew the experience in the future, is a hoax. Just repeating and mimicking the trainer. End of story."

The antithesis of science is off-hand dismissals. As is accepting hypotheses as "true" before allowing the evidence and experience to be "funneled" though it. Also, as mentioned in my above posting, insects such as homeybees have demonstrated learning, memory, and the ability to relay complext positional detail. A fully developed neocortex has nothing to do with it, unless you are human. Lose youre neocortex, well you're screwed. :) But birds have developed very different neural structures than we have, and are not limited by the same constraints. At one time scientists thought only humans used tools, and then they observed a tiny wasp using a stick to pry a worm from a hole. Nature surprises us, and most often when we dismiss something as nonsense.

I have an ex-girlfriend who had an African Grey parrot and the parrot would mess with her dog, Maggie. The parrot would sit in her cage and yell "Maggie....come here Maggie". The dog would come running into the room and the parrot would say, "Bad dog! Go lay down!" And Maggie would just glare at her with a look that said, "Man, if she didn't love you, you'd be dinner."

Science is a long narrow box into which all knowledge and experience gets funneled.

Wow. Settle down, there, chief. Nobody's trying to funnel "Ulysses" or "Woman Descending a Staircase" or the Spear of Destiny into a long, narrow box. (Although you *could* fit the Spear of Destiny into a long, narrow box.)

Science is a tool, like a screwdriver or a block-and-tackle. It isn't some kind of fascist ideology that's threatening to take your healing crystals away. It's just a really good "scope" with which to peer at certain things, for help with certain kinds of questions.

Hypothesis: Storytelling is unique to humans because only humans have a fully developed neocortex. Because that is true, the parrot's ability to "remember" the car trip and create a story about going in a car, including the professed desire to renew the experience in the future, is a hoax. Just repeating and mimicking the trainer. End of story.

It must be really easy to have a negative opinion of science when you characterize science and scientists and science-minded people in such a negative light. It isn't honest, but it's easy.

BTW, according to science, ALL psychic studies are flawed automatically because paranormal experiences don't exist.

Nonsense. There is a parapsychology community out there that continuously experiments in hopes of producing a repeatable result. So far, they have failed. But there's ample funding and ample interest.

Just because, after centuries of keen interest and earnest experiment, there is ZERO evidence of psychokinesis and telepathy, etc., doesn't mean science is a mean old man who ruins your fun by being a party-pooper and telling you your favorite supernatural fancies "just aren't so". Maybe you just have to learn not to be attached to your favorite potential superpowers. Just because it would be really neat to be able to pass through solid matter or heal cancer with a tuning fork doesn't mean you can.

Assuming the flowed result before doing or evaluating the experiment is the only way to keep the box narrow enough.

Who makes that assumption? Why do you insist on this cartoonish characterization? And what makes scientists inherently more closed-minded than reiki practitioners or acupuncturists? This just sounds like prejudice to me -- the idea that believing in cranio-sacral therapy or a psychic bird makes you more "spiritual" and "developed" or something...

If a peer-reviewed publication isn't admissible, then obviously only the personal evaluation of the critic is ... and of course, since there's not enough time to review and analyze everything, it's OK to dismiss unsubstantiated "silliness."

Yup, science is inherently biased. I guess we'd better establish the truthfulness of scientific claims by just seeing if they make us feel good, right? Or maybe we should ask our spirit guide, or consult some runestones?

Get your emotional baggage out of the trunk, dude. We need the room.

Now, don't get me wrong. I'd love to strap a tiny brainscan helmet onto Nkisi and see what kinds of things have developed.

But when I read "The conversation that most astonished Aimee Morgana", all I can think of is data mining -- that there could be thousands of gibberish conversations and endless repetitions of inappropriate non-sequitirs that we never know about, and a few lucky strings of words that "astonish" the credulous owner.

Video editing made Terry Schiavo look intelligent and responsive to human interaction, too. Was she psychic, or were other people simply trying to see what wasn't there?

There's some kind of mockingbird in my neighborhood, and I am enchanted every time I hear him. He runs through a string of different, beautiful little riffs -- one of them is a bluejay, but most them are little melodies. He's like the iPod of birds. I love that bird. But I don't imagine he's making conscious decisions about what to say next. It's instinct. Somehow, Nkisi's instincts were warped so that she synthesizes sentence-like phrases. That's wonderful. But let's do some tests before we start mistaking reasonable skepticism for closed-mindedness.

John- The italicized part of my comments were my responses to other comments. Not sure if that was clear. The snippets you responded to were made by falcon previously. Sounds like we're on the same page.

Ever watch how people learn languages?...when they are living in an environment where everyone around them speaks the language (like a child acquiring language frinstance), the first thing they pick up is strings of sounds--especially those associated with easy to pick up emotional content--like anger, frustration, surprise, pleasure--which they assign meaning to--rightly or wrongly--by linking it contextual activity/interactions. They don't actually understand the meaning of individual words or the strings at all. They just know "this is a communication you can make when xx happens". Over time they tune their verbal gestures to more closely match what others say/do. How different is that stage of learning "speech" in humans from the example of the parrot teasing Maggie the dog above? The parrot may not grok the individuals words in the string, but it's nailed the affect and intention. Can anyone argue animals are incapable of sensing and responding to the chemistry of basic emotions and physiological states like hunger, arousal, etc? Is it such a leap then, for those animals capable of verbal gesturing (verbalization, signing), to use those cues to facilitate expression. And are those expressions so very different from what we might observer in a child first learning to talk or with someone very new to our language?

Scylla - for very young children, it's not so different. For foreign language learners, it is actually quite different.

Language is not just the ability to match sign A to outcome B. My cat can do that. That's classic stimulus-response behaviorism. Real language use is the ability to break apart utterances, recognise individual units as parts that belong to classes, and productively combine them to make novel utterances.

For babies, it's kind of how you're describing. They get bombarded by adult talk. Maybe a neuron will fire, and the baby will say 'Baby', without even referring to anything. Then the adult will say something like "Yes, baby. That's what you are." and so on. The adult is doing exactly what I think Aimee is doing with N'Kisi, but in this case it has the happy effect of contextualising the word for the baby, and lexical acquisition takes place.

For second language learners, it seems to be a bit different. They already have the benefit of knowing a language, and they try to fit words into categories in the languagy way. There may be some phrases that go unanalysed in the early stages (like 'where is the bathroom?', but if they keep speaking the language, they'll try to unpack them.

Animals do feel hunger and arousal. They may use signs to communicate. It is a very big leap to say that they use language.

We get caught up by the idea that animals may be able to use human language, becuase it's really cool. I mean, Dr. Dolittle was a huge hit for that reason. But should we go around testing every fantastical story?
Sure, maybe it could happen but once it is proven not to be accurate should we still believe it is possible? Should I continue my childhood search for faeries at the bottom of my garden? The fact is, that there has been no scientific evidence that animals can use human language. This is not to say that animals are not intelligent or that animals cannot communicate with humans; just that they do not have the neural mechanisms necessary to create novel utterances as we humans do. A human and an animal can have a great relationship and indeed get to 'know' each other very well and even sense the most miniscule of cues from one another that can trigger a behaviour; like a dog 'knowing' when their owner is coming home.
May I suggest that if you are serious about this, then get serious testing it. You can not just say "it's happened too many times to ignore". In these situations people tend only to remember the times it does happen and unintentionally forget the others, it's just natural to observe what we're looking for and ignore the 'irrelevant' information.
It seems that we should all know about falsifiability and that psychic claims often are unable to be tested as they only work under certain circumstances: Little green men work my brain but you can't prove it becuase when you open my head they're invisible!
It concerns me that so many clearly intelligent people seem unable to think critically.

Sorry faerie, but your assessment is just wrong. First, there is evidence that animals can use language, published in peer-reviewed journals from Nature to Journal of Comparative Biology. Pepperberg is not the only one, either. There is evidence in chimps, greater apes, and bonobos. A simple PubMed search could show you. Also, it is baffling to me that you declare that animals do do have the neural mechanisms to make novel utterances. The simplest of animals have these, and those regions are not solely responsible for language anyway. Us humans have Broca's and Wernicke's areas, birds and apes have similar regions that when lesioned, result in aphasia. I mean, if you believe in evolution (hopefully!) you must believe that neural structures evolve from simpler ones. Therefore your notion that humans share no neural similarity with other species is just silly. If you read my post fully, or the comments, you would know that I think the psychic claim is bunk. But her other claims regarding the communication abilities of Nkisi, I do believe. This is mainly due to the PREPONDERANCE OF EVIDENCE that exists in the publication record from other parrots, other greys, other researchers, and other species. This is not an isolated incident as you suggest. This research has been going on for 30 years, or more. I encourage you to check out the bonobo Kanzi, and see if you are still so cynical.

Shelley -- on what basis do you reject the psychic claim? That was a double blind study, it was published in a journal, and the authors were able to attract funding -- all of which you used to support animal language studies. I mean, they used to say animals couldn't use tools or language, and now they say animals don't have psychic powers. Why don't you see that these are just arbitrary barriers thrown up by prejudiced cynics?

Claiming that animals use language and claiming that animals are psychic are really two sides of the same argument. Both rely on the ability of the owner to project their own thoughts and language onto the subject. There's no difference between the two.

As for me, I am very critical of ape and bird language studies (yes, I have examined them) because the evidence is not very compelling. The way to evaluate animal language is not to look at the interaction and say, "Yeah, that looks like a conversation to me." We need to control for the very human tendency to project intentionality onto the subject.

Here's one way to do it. Take N'Kisi, who knows how to say "There's a square" and "Look at the square". Take an object that he hasn't seen, and give it a made-up name (like a 'glark'). That way we're sure he hasn't heard that before. Next we show the object to N'Kisi, and play a recording of Aimee saying "There's a glark" over and over -- and only that sentence. Now if N'Kisi can hear that sentence and generate "Look at the glark" on his own without having heard it before when presented with the object, then I am willing to admit that N'Kisi is using human language, including syntax and semantics. He would have figured out the meaning of 'glark', and could break words out of their original sentences and replace them with new ones to form novel utterances.

No animal has ever done this to my knowledge.

faerie: I think you are spot on.

Fodor, the "peer-reviewed" study can be dismissed pretty easily if you read the methods. He scored 23hits out of 176 trials. They did some tricky statistics, and alas, was still published in a journal with an unranked impact factor which publishes UFO sightings. THAT's why i'm skeptical. Pepperberg publishes in the Journal of Comparative Psychology or the like and her results have been used in Nature reviews, etc. More rigor in the process, at least.

The task you're describing is pretty close to the Kanzi experiments (http://www.iowagreatapes.org/bonobo/language/). They used an arbitrary lexicon which bonobos used to communicate. In fact, they were trying to train Kanzi's mother while he was a small baby. Guess who learned to use the lexicon? Not his mom, who had passed the critical period, but Kanzi did. Novel experimenters' tasks, whom Kanzi had never seen before, elicit the same accuracy (and this is true for Alex too). Kanzi seems to have some grasp on syntax, but that in tenuous at best.

A compromise seems to be "complex communication" as opposed to language. Because I do not think that most animals tested would be able to manipulate syntax and grammar to the extent required to prove equivalent language use. On the other hand, there is intent, content, and flexibility in the modality.

To me the important question is: "Where did human language evolve from, and why?" rather than "Can x species meet the exact same requirements as humans in language usage?" It seems to suggest that our way is best, or at least better, than other forms of useful communication. Such research will lead us to learn more about lateralization (both bird and ape brains are for song/gestures/communication) and the evolution of neural structures contributing to human language and thought. What we really need to do is stick Alex in a fMRI during some tasks, although i doubt he'll sit still. :)

Shelley and others, let me defend myself here...

"Sorry faerie, but your assessment is just wrong. First, there is evidence that animals can use language, published in peer-reviewed journals from Nature to Journal of Comparative Biology. Pepperberg is not the only one, either. There is evidence in chimps, greater apes, and bonobos. A simple PubMed search could show you."

Firstly I enjoy how you really take other people's opinions on board and exercise critical thinking as demonstrated in the first delightful sentence of your reply. Secondly, you really need to define here what 'language' is because I'm not disputing the fact that chimps can use some signing to communicate or that nkisi can say 'look it's a square'.

"Also, it is baffling to me that you declare that animals do do have the neural mechanisms to make novel utterances."

Show me an animal who can talk and think like a human being, or some real evidence to suggest that this is the case.

"I mean, if you believe in evolution (hopefully!) you must believe that neural structures evolve from simpler ones. Therefore your notion that humans share no neural similarity with other species is just silly."

I did not suggest that we share no neural similarity with animals, that indeed would be silly.

"This research has been going on for 30 years, or more."

Yes it has, and I think it's about time we said, 'Yes, animals are smart and are great communicators, but they're not humans!' and gave it a rest.

"I encourage you to check out the bonobo Kanzi, and see if you are still so cynical."

Thankyou for your encouragement, I was familiar with Kanzi and many other animal communication studies prior to my previous post. It did not help me be less cynical, sorry.

Also, memorising a lexicon and putting random words together does not mean that an animal is using language. I suggest you re-read some of fontors posts.

Animals do feel hunger and arousal. They may use signs to communicate. It is a very big leap to say that they use language.

The biggest leap here, by far, is the assumption that all "animals" are somehow the same just because we call them "animals". We are talking here about a parrot, not about a nematode. Please!

memorising a lexicon and putting random words together

And grammar?

Read the part on "flied" again.