Machines Who Think
Art of Science
The World's Scariest Rollercoaster
After recess, it's time for math class. On Scientific American
Frontiers ... find out how smart animals really are. Also...
People -- and computers. Can you always tell them apart? An
endangered woodpecker gets a new lease... on life. And engineers
put a new twist into the rollercoaster. Come along for the
ride, on Scientific American Frontiers.
FLOWERS Hi. I'm Woodie Flowers, and welcome to Scientific
American Frontiers. Daisy, come. Now most people teach their
dogs to bark on command - you know, I say "Speak" and Daisy
says "Woof". But we like it quiet - so we tried another tack.
Daisy -- whisper. Cute, huh? And it didn't take her long to
learn that. In fact, I think Daisy learns faster than we do.
Watch this. She figured out, before we did, that returning
the toy just out of reach would get her some extra attention.
I guess everyone who has a pet must sometimes wonder just
what's going on inside that animal's head. The problem is,
how would you ever find out? After all, you can't very well
just ask an animal. Actually, it turns out you can, if you're
clever enough. And our first story is about two people who've
come up with some ingenious ways of getting animals to reveal
just how smart they can be.
PEPPERBERG I've got to go eat dinner. Gonna put you in, okay?
You be good.
You be good.
PEPPERBERG Okay. I'll see you tomorrow. I'll see you then.
Irene Pepperberg is a professor at the University of Arizona.
ALEX I love you.
PEPPERBERG I love you, too. Bye.
And one of her top students is an African gray parrot named
Alex. Like many parrots, Alex is a virtuoso mimic.
I'm sorry. You're a good boy. I love you.
He doesn't understand what he's saying, he's just parroting.
Or is he?
Come on, what is it?
PEPPERBERG Good birdie. Good parrot.
The extraordinary thing about Alex is that very often, he
does understand what he's saying.
PEPPERBERG What is it?
MAN Good boy.
PEPPERBERG Yeah, good birdie. Alex, what toy?
PEPPERBERG Nail, that's right. You're a good birdie. You're
a very good boy.
MAN What toy?
PEPPERBERG That's right.
You're a very good birdie.
PEPPERBERG Tell me what color. What color?
PEPPERBERG Yellow, that's right.
Good. That's right. Very good.
PEPPERBERG How many? Good boy. How many?
PEPPERBERG Good parrot. Good boy. One. Two.
Alex is even smart enough to answer different questions about
the same objects.
PEPPERBERG Can you tell me what's different? What's different?
PEPPERBERG Good boy. All right. What same? What same?
PEPPERBERG Good boy, good birdie. What color bigger? You know.
What color bigger?
PEPPERBERG Good boy. Good birdie.
If you still think Alex is just parroting, watch this.
PEPPERBERG Look. What matter four-corner blue?
Alex has never been asked this question before. To answer
it, he examines all the objects on the tray. There are several
4-cornered objects, and several blue objects, but only one
that's both 4-cornered and blue. Alex's job is to say what
that one object is made of.
What matter four-corner blue?
That's a good boy. You're right.
It makes sense that animals in the wild know something about
colors and shapes. But who would have suspected that they're
capable of this kind of original, logical thinking? Irene
Pepperberg's achievement is teaching Alex a way to show us
how smart he is. And she's starting the whole process again
with this baby parrot, Alo.
PEPPERBERG Can you tell us what's here? ALO Cork.
PEPPERBERG Yeah. Good girl. Good girl.
Alo has just learned her first word. But for her, cork means,
give me something that I want.
PEPPERBERG What's here - what's this? ALO Cork.
PEPPERBERG No. This is cork. What's this?
FLOWERS (NARRATION) So teaching her a second word is much
Irene, what is this?
Paper that's right. Say better.
Irene and her student model the answer while Alo looks on.
This is very similar to the way a baby parrot would naturally
learn how to sing.
PEPPERBERG In the wild these birds probably learn by observing
interactions between the other birds in the flock. They learn
their vocalizations by watching two adults duet with one another,
and that is what we're trying to replicate in the laboratory.
No. Denise, what's here? ALO Ay-er.
PEPPERBERG No. Say better. What's here?
It's only through this painstaking word by word process that
Irene will teach a new generation of parrots to tell us how
they think. Halfway across the country, at Ohio State University,
Sally Boysen is trying to find out how chimpanzees think.
One. Two. See that. One, two. Where's the two? That's it.
She is especially interested in whether or not they can do
One. Two. Good work, that was good. We're not done yet. Try
Bob is only 4 years old. And like a small child, he's beginning
his education with the basics.
He's learning just those first two numbers and once - I knew
he was going to do that. His attention span is a lot shorter
than a child's, and we end up investing a lot more time in
playing and social interaction than in actual learning. That's
kind of the opposite of what we see in children. Where they
spend a couple hours at school and 15 minutes in recess, we
spend 15 minutes in school and two hours of recess. You hear
that vocal laughter. Oh, boy. Tickle you good, aren't I?
FLOWERS (NARRATION) If Bob is in counting kindergarten, then
10 year old Sheba is a graduate student. She is learning her
numbers from 1 through 7.
How many is that? How many is it? Five. It's the right answer.
So you can eat them all. Okay. Good work.
Sheba also understands zero.
Okay. Look, Sheba. How many things do we have here? None.
There are no candies.
She helps herself to a reward.
She knows zero. Yeah. Oh, you think you should get reinforced
anyway. How many is that? You can do your own trials. It was
one. That was good. Well, gee, I don't even have to be here.
Now that Sheba's mastered her numbers, Sally has come up with
a new challenge. She puts 3 peaches in one box, and 3 peaches
in another. For the first time ever, a chimp will try to solve
an addition problem.
Okay. How many peaches? How many? Show me. Yes! Six is the
right answer. Good girl, there's six peaches out there.
The test involves counting the first set of peaches, holding
that number in her head, and then continuing the counting
with the second set.
Get the right answer. Five. Good.
Without ever being trained to add, Sheba gets it right the
first time. But what happens when Sally makes the task even
harder, by substituting numbers for the peaches?
Do you see that one. Hey. There we go. Okay. How many was
that. Can you pick? Show me. Four. Perfect, four. It was once
thought that only humans could work on this level of abstraction.
To find capabilities in the range that we have with numerical
concepts is not surprising to me. It means we have to rethink
our ideas about what humanness is all about, and also what
champanzeeness is all about.
To test Sheba's number skills in a new way, Sally has devised
an unusual game. One of the players is Sarah. Her job is to
eat candies. Sheba's job is to choose how many candies Sarah
One, two, three, four down here. Are you watching? Miss Priss.
Here's how the game works: whichever set of candies that Sheba
points to first goes to Sarah. Sheba gets what's left.
We're going to put two in here. Give those to Sarah. Sheba
gets two. Sara gets four and Sheba only gets two. Oh, too
So if Sheba understands the game, she should pick the smaller
Put five in that one and we'll put one in there. Now which
one do you want Sarah to have? Oh, you want Sarah to have
these. It's okay.
But no matter how many times they try it, Sheba always points
to the largest amount first. She just can't figure out how
to win this game. Sally has a hunch that making the task more
abstract might help.
Let me ask you a question. Which one do you want to give -
you want to give two to Sarah.
And remarkably, Sheba now picks the smaller number first.
And you get six. Oh, you lucky girl. One, two, three, four,
Sarah's not that happy about this new turn of events.
We have four - now wait, wait until I show you what else we've
got. Now what shall we give to Sarah?
Sally's theory is that actual candies trigger an instinctive
greedy response, but number symbols engage Sheba's mind and
help her play the game intelligently. Unlike Sheba, the males
in Sally's group are too big and too aggressive to let out
of their cages. Still Sally has found a way to play with them...and
work with them.
How much of a peach is this? It's a half of a peach. Right.
Darrell is working on fractions.
It's a half of a peach. That's right. What's that in your
mouth, a peach pit? Thank you. Yummy, yummy. How many of these
do I have? Two bananas. Right. There's two bananas here. Now
watch what I'm going to do. This. And one more time, and now
I have this little weensy piece. It's a fourth of a banana.
When Sally cuts the fruit in front of him, Darrell has no
problem choosing the correct fraction.
I've got some left, too. I have another fourth. That's right.
There it is. Darrell has to understand that there are whole
things and there are parts of things; and that we can assign
a name for different parts of an apple, for example. So that
there's a special name for a half of an apple, a special name
for a fourth of an apple.
When Sally cuts the fruit in front of him, Darrell has no
problem choosing the correct fraction. The chimps don't like
it very much when Sally turns her attention to the camera.
But they figure out a way to get revenge. Soon, they're calm
enough to start work again. Darrell's next challenge is to
determine what fraction of a pear this represents, without
actually seeing the fruit cut up.
Guessing all of them, guessing all of them. I want to show
you something. Look what if I took a whole pear like this.
Oops. This is one pear, isn't it? And I'm going to cut it
up for you. Darrell's still not able to reliably pick the
correct fraction. It really helps him still at this phase
of the understanding of the concept to see a whole fruit divided
into portions. But we're working towards that. Darrell, look
how many pieces I cut this into - one, two, three, four. That's
right, this is only a fourth. That's good, a fourth. Good,
do you want it? You don't want it? Okay. Would you rather
have a piece of candy? Alright. There you go. You can stay
here. We're still working. Are you going to be afraid of dogs
your whole life?
Like Alex the parrot, the chimps are actually teaching us.
Look at - he's a little goofy guy. Oh, he got you. It's two
miles an hour.
We're learning how animals think, and that's a lesson that
never fails to surprise us.
Uh-oh. Don't hurt him. Don't hurt Skylar.
FLOWERS (NARRATION) Our next story is about absolutely cutting-edge
technology - but its roots go back more than fifty years.
That's when this man - a brilliant British mathematician named
Alan Turing - began thinking seriously about a thinking machine.
And just a few years later - in the bleakest days of World
War II - he found a way to put some of his ideas to the test.
This is the famous Enigma encoding machine. The German military
used it to protect their most secret messages. At least, that's
what they thought! But Turing and his colleagues built this
- the world's first electronic computer - and used it to crack
the Enigma codes. Along the way, Alan Turing asked a profound
question about computers: Can a machine be intelligent? He
even devised a practical test to answer this question. How
does this "Turing Test" work? That's what we're about to find
What is your favorite Shakespearean play?
King Lear. I also like many of the comedies. How about you?
Sonnets. Also the comedies, but the tragedies are my favorites.
Are you familiar with Hamlet? **
The college kid who came home and found his mom had married
the guy who murdered his dad just a little month before? You
might say so.
Do you think William was a sexist?
Of course he was. At that time everyone was. Even the Queen.
Haven't you read The Taming of the Shrew?
We're eavesdropping on a conversation that's actually taking
place on the screen of a computer
TERMINAL Emy Gugsa is at one end of this lively exchange,
asking lots of questions about Shakespeare. But Emy is really
trying to find the answer to an entirely different question.
You see, Emy doesn't know whether she's talking to another
person or to a computer program. Her conversation is being
tapped...not by the KGB or the CIA but by an audience of experts
and enthusiasts gathered at Boston's Computer Museum. These
people are hot on the trail of the question ..."Can a machine
think?"...When Alan Turing tackled that question, he came
up with a test to answer it. Today we're witnessing the first
attempt to apply this test. If a computer program can pass
the Turing Test, these experts will consider it a thinking
Rarely in philosophy, certainly, is there a clear, decisive
event like this, where things that have been ideas for years
and years actually enter the real world and so it's an exciting
moment for all of us.
The Turing Test is set up like this...In one part of the museum
are ordinary folks who, like Emy, have no particular computer
expertise. They are the judges-- and it's up to them and not
the experts to interview all the contestants and decide which
are human and which are computer programs. Every 18 minutes
the judges rotate and begin another interview. There are two
groups of contestants. The first group, the human contestants,
are hidden away in the bowels of the museum. Their job is
to conduct normal conversation with the judges over the computer
TERMINALs. The other group of contestants are computer programs.
They're also talking to the judges over computer
TERMINALs. This is the Turing Test. There are eight contestants.
The judges have to figure out which ones are human and which
ones are computer programs. If one of the programs fools the
judges into thinking it's a human, we call it "intelligent"
and it passes the Turing Test. The experts and enthusiasts,
watching from the sidelines, will have their own ideas about
who is human and what is not. Now open-ended conversation
is far beyond the capacity of even the most sophisticated
programs. (Graphic) So, to give the computer program contestants
a fair chance, this will be a restricted Turing Test. Each
contestant has chosen a conversational specialty. And the
judges' questions have to stick to these narrow topics. Let's
look at this topic, "Problems with Romantic Relationships."
We're going to give away the contestant's identity in this
one...to get you warmed up.
Where's the best place to meet men?
TERMINAL Perhaps you're
not getting enough respect in the relationship.
But I don't have a relationship.
TERMINAL You're not getting
enough security in the relationship.
But I don't have a relationship.
TERMINAL Perhaps you're not
being appreciated enough.
No one is fooled by this contestant. STUDENT That's obviously
a computer program.
ENTHUSIAST There's just no interaction
going on there at all. They're just about three miles apart
all the time.
So the Romantic Relationships program fails the Turing Test.
TERMINAL I'm an eight year old girl, a gymnast and a Baptist.
Do you go to a Baptist school?
TERMINAL Be careful with non-metals,
they're often brittle.
Do you work with metals in second grade?
TERMINAL Machines can only do what we have the skill to tell
them to do.
STUDENT Well I'm in fourth grade and I don't think that's
a human at all. Second graders just don't talk like that.
PSYCHOLOGIST/ROBERT If I interact with you, and all of a
sudden you make a remark that's sort of off the wall, I forgive
you that remark. I can forgive you several remarks. If you
keep doing too many of those, I'll say "uh oh we've got a
little problem here."
Making too many off-the-wall remarks, missing the point, giving
short patterned responses, these are the giveaways that distinguish
the computer programs from the humans. But some of the other
conversations are proving much harder to judge. Shakespeare's
Plays is driven by a human contestant. This is Cynthia Clay...She's
a real Shakespeare buff. Now remember... Cynthia is hidden
away...you know she's human...but the judges don't. Let's
see what they're making of her conversation.
Do you know much about the authorship to Two Noble Kinsmen?
That is not among the 37 plays, unless that is some subtitle.
Supposedly it was written by Fletcher and Shakespeare.
BITE Oh. That's interesting.
TERMINAL But I'm required to
only talk about Shakespeare. Why don't we just stick to the
Well, unlike the computer programs we've seen so far, Cynthia's
on the ball. She understands the questions and her answers
are to the point. So...all the judges recognize that the Shakespeare
contestant is human...or do they?
EMY Which of his plays is your favorite?
BITE Let's see I'll give 'em a bizarre answer. Let's see...Pericles.
I'll bet they've never even heard of that one.
TERMINAL Let's see Pericles.
EMY Why is it your favorite?
CYNTHIA Because he was obviously called in to play Dr. Somebody's
awful script. It was one of the few maybe only two plays written
with somebody else.
EMY BITE It was just specific information that sounds like
something that would be programmable. So I just thought it
was a little more on the machine-like side than human-like.
Wow...one judge actually thinks Shakespeare is a computer
program...and the audience is split right down the middle.
SCIENTISTS Some of the answers seem too studied. As if they
were somehow canned opinions that came from a large data base.
I think the Shakespeare one's human because of the way it
answers questions. It can understand the larger meaning of
the question. It's not just responding to one key word.
It seems overly intellectual. The responses are too perfect.
So it makes me worried that they might just be big canned
ENTHUSIAST I thought I identified one human and
I think it's the Shakespeare one.
Ironically, Shakespeare is hard to judge because Cynthia knows
too much. Her command of Shakespeare is encyclopedic. Having
vast amounts of information instantly available is something
people usually attribute to computers.
They thought I was a computer because someone couldn't know
that much. It's like...people go to school! People learn things!
Think how much Doctors have to know!
There's one more conversation that's a real puzzler and by
the end of the day quite a crowd gathers around the "Whimsical
Conservation" contestant. GRAPHIC It's judge Roseann San Martino's
turn to put this one to the test.
TERMINAL Do you have a pet?
ROSEANN Yes, I have a cat. Her name is Bon Ami. Do you have
TERMINAL Honeybunch the cat has been our pet for about
eight years now. She's a regular little kitten factory.
BITE It was personal in a certain sense. I mean we talked
about our pets and it seemed to have a life.
TERMINAL I used
to live in Connecticut before all the riff-raff moved in.
But now I call this little townhouse home.
ROSEANN So you
TERMINAL Everything's been figured out except
how to live. How ya' doin' judge?
ROSEANN BITE I'd like to
have gotten to know this person, if it was in fact a person.
For some reason I assumed that it was a guy. It just seemed
guy like. You know, I would have liked to have met his cat,
and seen his little townhouse, or whatever. And just chatted
with him a little bit more.
Roseann clearly hit it off with the Whimsical contestant.
What do you think? Is this a human or a computer program?
The last round is over. No more conversation. Every judge
has interviewed every contestant. Now the judges have to decide.
For each conversation, were they talking to a human or to
a computer program? It's these ratings that will determine
whether any program passes the Turing Test. Dr. Daniel Dennet,
one of the philosophers who organized this contest, will announce
DENNET The winner of the 1991 competition is Whimsical Conversation,
computer contestant Joseph Weintraub. Whimsical Conversation
didn't just win, it was judged to be a human being by five
out of the ten judges. Speaking on the topic of Shakespeare,
Ms. Clay was judged to be a computer by two judges. But perhaps
it will soften the blow for her if she realizes as well that
she was judged the most human on the mean. So she wins both
most human and most easily confused with a computer.
So what has this first ever Turing Test revealed about computers
CONTESTANT I think we're learning more about the question
of what it means to think than we're learning about what the
capabilities of a computer are.
It raises questions about what it is to be human, what's special
about being human.
You learn about the limits of intelligence, not just the limits
of artificial intelligence.
I think computers can obviously engage in dialogue and they
can draw on a data base. But I don't think they can originate
a thought. That certainly wasn't demonstrated today.
The results of today's restricted Turing Test are not so important.
What matters is the proof that the test itself actually works.
We now have a meaningful way to ask and to answer the question,
"Can a machine think?"
FLOWERS (NARRATION) Usually on the "Art of Science" we bring
you some of the hottest animation produced on today's supercomputers.
But this time we've got a real classic - a film called "Powers
of 10," made by Charles and Ray Eames way back in 1977. Its
age shows a little, but "Powers of 10" is still a great movie.
In fact, for me it's the best illustration ever of a very
abstract idea - the relative size of everything in the universe,
from galaxies to subatomic particles. In our clip, we start
at a picnic beside Lake Michigan. Then, every ten seconds,
we'll zoom ten times farther away - until we fly back in like
In the summer of 1989, the Francis Marion National Forest
in South Carolina was a haven for one of America's rarest
birds, the red-cockade woodpecker. The woodpecker is an endangered
species because old growth pine forests like this, once common
in the south, are themselves disappearing. The Francis Marion
National Forest was special because here the population of
the red-cockade woodpecker was actually on the increase. Then
came the night of September 21st. 1989's killer Hurricane
Hugo struck South Carolina with almost unparalleled fury -record
high seas and one hundred fifty mile-an-hour winds. Watching
the news reports with mounting horror was ornithologist Jerry
JACKSON I was keeping a very close eye on where it was headed.
I had been on the Francis Marion National Forest only about
two weeks before the hurricane hit and I knew what was in
store and it was disaster, I mean absolute disaster.
In just a few hours, most of the Francis Marion was leveled.
Of the forest's seventeen hundred red-cockade woodpeckers,
one thousand were dead or missing. Many trees snapped where
the woodpeckers had drilled their holes, crushing the birds
sheltering inside. It's this destruction of the woodpecker's
homes - even more than the loss of the birds themselves -
that was the most devastating of Hugo's blows to the species.
Because the red-cockade woodpecker's home is definitely not
just a hole in a tree. To find out why the red-cockade woodpecker's
nest is so special, Frontiers came here, to Mississippi's
Noxubee Wildlife Refuge. It's early in the morning - the best
time to catch a woodpecker. Noxubee - another of the few scattered
old growth pine forests of the south - currently is home to
just forty-one woodpeckers. White bands mark trees with woodpecker
holes. The question for Jerry is whether the bird in this
tree is sleeping late.
JACKSON Hey! We got one!
If this is a female, then she's the only one at home. That's
because males and females keep separate living quarters -
and it's the father who nests with the babies, while the mother
sleeps off by herself.
JACKSON Oh, this is a bird that has bands on it, so we've
got us a recapture. We'll be able to tell how old this bird
is and when it was banded.
And its sex - because only the males have the red cockade.
JACKSON They only show these red-cockades when the birds are
aggressive or courting. This bird's a little aggressive right
now because he doesn't really like being caught.
Because it's a male, there are probably babies in the tree
- and their hungry cheeps give them away. Both male and female
share the feeding chores during the day.
JACKSON Well, get back to your nest, fella.
So here's the first reason the holes are valuable - each family
needs two. The second reason is visible around the hole itself
- sticky sap, looking like candle wax. The sap is there because,
unlike other woodpeckers, red-cockade woodpeckers choose to
make their holes in live trees, not dead ones. And making
a hole in a living tree is hard - very hard.
JACKSON The average length of time that it takes for a red-cockade
woodpecker to excavate a cavity is four point seven years.
As a result, those cavities are extremely valuable pieces
of property, and they're passed down from generation to generation,
and the males inherit them.
Four point seven years' work per cavity is quite a mortgage
to pay off - and it raises the question: Why? Why do red-cockade
woodpeckers drill through live sapwood, while other woodpeckers
just peck their way through dead wood? And here's the answer
- the gray rat snake. It's a superb tree climber, its scales
giving it an excellent grip on the bark. But around the nest
hole, the woodpecker has pecked the bark to keep the sap flowing.
As the snake approaches its goal - a succulent meal of eggs
or nestling - the waxy sap gets thicker and thicker. The sap
works its way between the snake's scales, sticking them together.
But if the sap protects against snakes, it doesn't keep biologists
away. Fortunately, Jerry's motives for climbing the tree are
more benevolent. While the parents are away gathering food,
Jerry plans to check on the nestling. This is a three-week
old female. She's banded, then weighed -- during which she
gets her first real look at the world she'll soon be joining.
Jerry's work is part of a detailed monitoring of the birds
here in Noxubee from birth to death.
JACKSON That about does it - forty-five point three grams.
This bird is very close to fledging. You can see it has very
well-developed wing feathers, but that the wing feathers are
still growing. And those feathers will be completely grown
probably within the next week, and this bird will be out of
the nest by then.
Out of the nest - and needing one of her own. Her home nest
will be inherited by a brother. Fortunately, in Noxubee, there
are still a few trees to be drilled, a few homes to be had.
But back in the Francis Marion National Forest, Hurricane
Hugo crushed hundreds of woodpecker homes. With so many trees
destroyed - ironically, snapped in two often because they
contained woodpecker holes - the shortage of homes is acute.
And because it takes so long to make new cavities, wildlife
managers here feared that the seven hundred birds surviving
the hurricane would die homeless. So in the spring of 1990,
they went into the construction business. Biologist Carolyn
Bachler first drills a horizontal hole, just as the bird does.
Then she drills a second hole at an angle from above, to create
a vertical cavity down through the middle of the tree. The
idea is to reproduce in a few hours what it takes a bird over
four and a half years to make.
HOOPER Here's the entrance, here, and it's about nine or ten
inches deep. The light-colored wood here is sap wood, and
it is living wood and the resin actively flows through it.
The reddish-colored wood here is heartwood and it is essentially
dead - there's no resin movement through it. The bird needs
the heartwood into which to put its cavity so that it doesn't
have to contend with resin within its cavity. Carolyn Bachler
scrapes off some bark to set the sap flowing. The tree is
in move-in condition - but it took half a day's work. Here's
a pre-fab option. This time, the tree is excavated by chain
saw. A little putty - and a bird house is slipped into place.
Even sap holes are started for the hoped-for tenants. A little
paint gives the impression that the sap is already flowing.
In all, over six hundred artificial cavities were built in
the Francis Marion Forest over the winter. With spring - and
nesting season - came the chance to find out if they worked.
Eddie Taylor checks one of the bird house implants with a
light and a dentist's mirror.
Bob, we've got some eggs.
Is that right? How many?
That's the very first nest. The same nest, three weeks later
- and hungry chirping signals that the eggs have hatched.
What I'm gonna do is fixing to pull the nestling out of the
nest so we can band them. What I'm gonna use is this little
instrument right here. It makes little nooses, and then I'll
pull it tight and it'll grab them and I'll pull them out gently.
It doesn't hurt the young because their bones have not calcified
at this age, and so they're still malleable. And we can handle
them gently without hurting them.
The chicks' eyes aren't open yet - but they can tell the difference
between light and dark. And dark means their hole is covered
by Mom or Dad arriving with food. At least that's usually
how it is. A nest-by-nest inspection of the trees that remain
in the Francis Marion National Forest this spring showed that
sixty-five percent of all the new birds were born in the artificial
cavities. A band on the bird will help keep track of it in
the future. The man-made nests appear to have given the woodpeckers
here a new lease on life. But even as this most important
population of red-cockade woodpeckers has been rescued from
a rare natural disaster, the erosion of their habitat all
over the south continues. The real threat to the species'
survival remains man, not nature.
JACKSON There's a lesson to be learned form Hurricane Hugo,
and we're very fortunate we have the opportunity to learn
that lesson. And that lesson is that every population is important.
And that we can't count on having a few large populations.
This is probably the best-known woodpecker in the world. We
know what it requires - we know what it needs. We've just
got to make up our minds to do what it needs to protect the
THE WORLD'S SCARIEST ROLLERCOASTER
FLOWERS (NARRATION) I've always wondered what it would be
like to jump out of a plane - so I'm going to find out.
INSTRUCTOR Head up nice and high - hard arch.
FLOWERS This is fantastic - I hope you guys can hear me.
All right give me a left turn, Woodie.
Stop - turn left.
Those are nice, swift turns. This is a 3-D sports car. Here
INSTRUCTOR 3, 2, 1, flare. All right, good job, Woodie. Wow,
I'm sure the guys with the cameras did a great job, but there's
no way pictures can do justice to what I just experienced.
I knew this was going to be a fun job.
Actually, I spent a whole day training for this, and it cost
several hundred dollars. And even though the risk associated
with the jump was probably less than that of driving out here
this morning, it's pretty clear that sky diving's not for
everyone. But a lot of people would like to get the kind of
thrill that I've just experienced without spending the time,
and the money, and the nerve. And that's the reason that amusement
parks are big business - and that's the reason that the designers
of rides are always trying to outdo one another. We wondered
how you would design a ride that was both thrilling and safe
- so we've spent over a year tracking the designing and testing
of what is - at least for now - the world's ultimate rollercoaster.
Three or four major rollercoasters open up every year. And
in a never-ending competition to be the best, each is bigger
than the one before it. Today, coasters can be twenty stories
tall and reach speeds of seventy miles an hour. The coasters
are designed to be safe. But their intimidating looks are
beginning to scare off less adventurous riders -- according
to the designer of this coaster, Ron Toomer.
I really think a lot of people are going to walk up to that
thing, look up there and say, "My gosh, that's too big for
me -- I'm not going to get on it."
FLOWERS (NARRATION) So five years ago, Ron began toying with
a brand new concept in coasters. Here's how all existing coasters
work. The twists and turns and ups and downs in the track
create forces pushing and pulling on the rider. That's what
makes them fun. But always the greatest force is pushing car
and rider against the track -- even when you're upside down
in a loop --and that's what keeps you from falling out. To
demonstrate, we joined a high school physics class as they
rode a coaster with their own home-made force meters.
That's going to register on this. When we're standing here
still, this is our one unit of force. This would mean that
on your rear end, when you're in that car, you'd feel your
normal weight. But as we go around the bottom of that curve,
you're going to see this thing pull down to two, three, or
three and a half times your normal body weight.
FLOWERS (NARRATION) For the experiment, each student was supposed
to look at and remember the force holding them into the car
at different points in the ride. But it proved easier said
Too scared to look at it. We didn't look at it - we forgot.
Gonna do it again.
FLOWERS (NARRATION) We made it easier for us to see by slowing
down the action. Entering a loop, there's a force three and
a half times greater than normal gravity pushing them into
their seats. Even when upside down at the top of the loop,
they're being held into their seats by a force two times gravity.
So at no point do they really feel as though they're upside
You felt like you were on Earth rather than on a rollercoaster
upside down. Didn't have a sensation of being upside down.
Except for the visual sensation of looking upside down, it
felt like you were sitting right-side up. In all existing
coasters, that's just as well - otherwise car and riders would
fall off the track. Only briefly, at the crest of a hill,
does a rider feel like he's coming out of his seat. But what
if a coaster could be designed that did a barrel roll -- like
an aerobatics plane?
We're looking for an advancement of rollercoaster technology
here that would allow us to feel more like that kind of a
sensation where you're able to roll over, fly along upside
down, and you really are upside down. You're hanging upside
down, you look down and there's nothing below you but the
ground - and there's no tracks -there's nothing there to hold
FLOWERS (NARRATION) It was this new twist in rollercoaster
design that Ron and his company, Arrow Dynamics, decided to
explore. This is an early run of a quarter-scale model. To
create the barrel roll effect, the car sits down in the tracks
rather than on them, as in existing coasters - and to hold
it up when it's upside down, it hangs from a second set of
wheels. These are big innovations for a rollercoaster, and
when we joined Arrow in the early stages of development, there
were still plenty of questions to answer - like just how much
initial energy the car would need to avoid the discomfort
- and embarrassment - of a problem like this! The main task
though, was to stop people from falling out - the job of engineer
Would you like to pull the restraint down now?
FLOWERS (NARRATION) The challenge is to come up with a restraint
system that holds people comfortably and securely, no matter
what their shape or size.
Feels great. Really secure.
So you feel more secure hanging onto the handles?
If we're going upside down.
We're going upside down.
I'll hold on.
FLOWERS (NARRATION) Ideally, the restraining system would
be adjustable to fit each rider - but the time that would
take between rides would make the rollercoaster unprofitable.
This system works well for Cindy, an average sized woman.
But Bill, who's a little larger, would clearly have some problems.
The lap bar is meant to fit snugly across his legs.
Obviously, this is not going to restrain his body and offer
the support that he needs when this vehicle turns on its side
or goes upside down.
FLOWERS (NARRATION) Bill can be accommodated only by shortening
the lap bar so it can reach his lap. Now the system works
fine for Bill.
Okay, how do you feel?
I really feel secure. There's no place to go.
FLOWERS (NARRATION) But can it still do a good job on someone
Dusty's size? Arrow's goal is to fit children four feet and
taller. Dusty's four feet four.
Pull that down so it's tight. Feel good? Feel locked in? Okay.
FLOWERS (NARRATION) The shoulder restraint only keeps him
from falling forward. So it's up to the lap bar to hold him
Okay. Dusty, how ya doing?
Good. Things feel OK here.
Okay, your legs are tight, look like they're good and tight,
your arms - you're holding on a little bit here, your head's
clear, OK, and you're off the seat just a few inches. Let's
turn him back up.
FLOWERS (NARRATION) Freeman is increasingly confident his
system will hold riders as different in size as Bill and Dusty,
comfortably and securely.
There's a partial compromise here. If I were setting it just
for him, I'd probably move this lap bar another 1 or 2 inches.
But the way the seat is designed in the lap bar, we got a
range here, where we can lock him in securely, and as well
deal with someone that probably weighs 4 or 5 times as much
as he does.
FLOWERS (NARRATION) Devising an effective restraint system
was a problem they'd anticipated. But the first time they
put a mock-up of an entire car on the track, there was a nasty
surprise in store. Because the car sits down between the rails,
when it goes through its barrel roll, the rear passengers
are brought within a dangerous few inches of the track.
Boy, look at that.
That looks pretty close.
You know, we haven't had to deal with this before, in that
the tracks have always been pretty well under the vehicle.
But this track's coming right up around as we go through the
spiral. We're going to have to do some shielding here, some
more shielding than we normally do. One thing I don't want
to do is close this in with a canopy. We've got to keep the
vehicle open to maintain the thrill of the ride.
FLOWERS (NARRATION) But in the end, a small rear canopy is
the only solution to making sure hair and hands don't get
caught between wheels and track. At last, it's time to erect
the test track -- one quarter of the finished ride. The joins
must be precise to avoid dangerous lurches. But even the greatest
care doesn't mean the coaster will run as predicted. While
expected speeds and forces were all carefully calculated,
they were based on assumptions about wheel friction. And this
is the first time a coaster has alternated riding on top and
bottom wheels. If the friction is less than expected, the
coaster will run faster than planned -- maybe dangerously
so. Which is why the first passenger won't be a person, but
a machine -- an accelerometer. It's much like the students'
home-made force meters, but it records front-and-back and
side-to-side forces, as well as up-and-down. Three years after
work began, the coaster is hoisted into position for its maiden
voyage. The starting height is still something of a guess.
If it's too low, the car may not pick up enough speed to make
it around the track.
There's always that feeling inside - you know - what can happen
that we don't know about yet? And maybe it'll end up just
going back and forth right here in the dip and not make it
all around or something like that.
FLOWERS (NARRATION) The moment of truth. It looks like a perfect
run - but were the forces recorded by the accelerometer in
line with predictions? All seems to be well -- even in the
critical region under the line when the coaster is hanging
upside down. So it's safe for a human rider. Who will it be?
I don't ride them anymore. I get sick, for one thing; and
I don't like to go upside down for another; it's just not
FLOWERS (NARRATION) It's the engineers who'll take the first
spin. After all, they built it!
Bring lap bars down first. Tug on them a little bit. Okay?
Have a nice ride.
FLOWERS (NARRATION) From the sound of it, Arrow's created
the new thrill it was seeking.
This ride concept opens up a whole new dimension to us. And
we can see, instead of the horizontal spirals, we can actually
see a drop, rolling. We think that'd be wonderful. We see
a high speed spiral - just exciting things in our minds and
they've got to be developed and tested, but I can tell you
right now they're going to work.
FLOWERS That's it 'til next time on Scientific American Frontiers.
Please come on back and watch.