[MUSIC PLAYING] Stanford University. Let's get started. Let's see. Once again, game plan for
the coming week or so. Next three lectures will be the
TA overviews-- introductions to nervous system,
endocrinology. Once again-- what I
emphasized the other day is-- if you're not sure if you've
got enough background exposure to these topics for
it to be worth coming, I would suggest you
assume that you don't. Because it could do
nothing more than help, even if you've had
a lot of this before. Following that, some more
advanced topics in the subject, and then barreling
on into the midterm. OK. Today we leap yet
another discipline into yet another one, this one
being concerned with basically the same exact issues that we've
had before-- nature, nurture, interaction, behavior. And this one, as you will see,
with a completely different approach than all
the prior ones. Now the other day, when going
through what heritability means and doesn't mean, one of the key
points that came out at the end was that whole business of,
the more different environments you study something in, the
lower your heritability term is going to be. So what's one of the punch
lines that comes out of that? If you want to understand
the genetics of behavior, you've got to go look
at the behavior in lots of different environments. What today's
discipline is about is taking that one step broader. If you want to
understand anything about the biology of
behavior, number one, you want to study it in lots
of different environments. Number two, you want to study
it in as natural of environments as possible. And number three,
go in there assuming that you are going to have
to translate a language. The quote at the
beginning there, something to the
effect of ethology, the field we'll be
looking at today. Ethology is the process
of interviewing an animal, but in its own language. So what we'll see
here is this is an expansion, in a certain way,
on that notion that if you want to understand behavior, get your
animal out of the laboratory. Now to begin to
sort of appreciate the context this
has in history, we start off-- God help us--
with history of psychology at around 1900 or so. When, universally agreed upon,
the Grand Poobah of psychology was William James. And all I know
about William James is that he was
highly philosophical, and I fall asleep every
time I try to read the guy. But he dominated the period
in a way in which psychology was basically an introspective
business that was sort of a branch of philosophy. And coming out in the
decades after that were the Young
Turks of psychology, who despised this whole
approach, because they wanted to make psychology a science--
a quantitative science, an experimental science. They wanted numbers. Enough of this philosophy stuff. And out of it came this very
deep, deep reflexive distrust of any behavior you couldn't
see and you couldn't measure. And you didn't want to know what
was going on inside anybody's personal life,
inside their head, whether they were
a fish or a human. All you were
interested in is what was going on in
the outside world just before that
behavior occurred and what behavior was it. Simple organisms as
input output black boxes. Forget this philosophy stuff. Give me something I can measure. And this was this
emerging discipline of behaviorism, which
became the dominant school in American psychology by the
middle of the last century. We've already heard from one of
the starting founding fathers of it, and that was John Watson. That was the guy from the
first lecture-- give me a kid from any
background, and I'll turn him into a butcher,
baker, candlestick maker sort of thing, and
that everything could be controlled in environment. And thus, you could
shape behavior to any direction you want. Behaviorism sort of hit its
heyday, its absolute apogee, with its most
famous practitioner, who was BF Skinner. Skinner-- who dominated, I
think, most studies show. Or that he was the
best known, most cited psychologist
throughout the middle part of the last century. Incredibly influential, and
American psychology at the time was all about behaviors. What are the key features of it? First one was this
hysterical, radical, extreme environmentalism. The notion that we are all
table rases, whatever that is. That we are all blank slates
and all that we come in with is this blank slate to be
written on by environment. Don't tell me about genes. Don't tell me about
biology at all. Don't tell me about
anything other than what's going on
in the environment, and how does that shape
the behavior coming out the other end, and
can I measure it? Extreme radical environmentalism
in terms of gene environment. There were no genes. There were no interactions as
far as they were concerned. The next building
principal was this notion of reinforcement theory. Let me control when any
given organism receives a positive
reinforcement, a reward. Let me control when it receives
a negative reinforcement of punishment. Give me the ability to control
those in the environment, and I will produce any behavior
you want in that organism. Once again, whether it is a
fish or a human and everything in between, this utter
reliance on behavior is shaped by the rewards and
punishments of environment. Reinforcement theory. The final piece of it was
this notion of universality. Which is, it works
this way in everybody! It works this way in every
single species out there. And if you want
to study lab rats, you're going to learn
the exact same thing as if you were studying giraffes. But it's a lot harder
to study them in a lab. You sure don't need
to study them out in their natural habitat,
because it's just understanding the reinforcements coming
in, the behavior coming out the other end. They're all the same. This period dominated
by the behaviorists was, historically, a total drag. Because these guys had just
a stranglehold on the field. And all of them,
like, looked the same. If you ever look in one
of those time-life books on the history of
the mind or whatever, there will be the
inevitable picture of BF Skinner with some pigeon. And Skinner was
like your archetypal out-of-central-casting
behaviors. He had this inordinately shiny
forehead and these big glasses. And every picture of
him, for some reason, photographers were always
forced to kneel down and take a picture of him looming
over with his pigeon. And everything considered, this
was an incredibly frightening looking man. And he would then be free to
go on with astonishing quotes like, dog, cat, rat,
child, doesn't matter. It's all the same. This notion of universality
of behaviorism. And Skinner invented all
sorts of Utopian societies-- one, a book of his that
was very influential, a novel called Walden Two. You can build a perfect world
out of reinforcement theory all with environmentalism. And I could have built
the exact same utopia studying this with grasshoppers
as with humans-- boy, cat, dog, rat, whatever. They're all the same. So the building of American
behaviorism-- utter reliance on environmentalism, utter
reliance on the notion that all of behavior could
be explained-- could be shaped by the
rewards and punishments, the positive and negative
reinforcements of environment. And this works the same way
in every single type of animal out there, so you might as
well just study it in a rat, in a pigeon. Those were the two backbones
of behaviorist research. Because they're a
lot more convenient. And you can study it in
the lab, because it's the exact same thing. It is purely a function of the
environment you're controlling. And you understand that. And as long as there is
some behavior coming out the other end that you could
measure and generate numbers, you are doing
scientific psychology, and this is what the
whole field was about. Meanwhile, over in Europe,
there was a totally different tradition emerging. And what it has its greatest
similarities to is-- you know, you read
any of these books on the history of evolution,
and there's always this phase of like all
of these 19th century British naturalists,
where like half of them were Parsons in some
Anglican church or whatever. They always had to be Parsons. And they would always
be out trumping around the woods of England
with a butterfly net. And these were the guys
who would come back with a gazillion different
types of butterflies. This is when
butterfly collecting became this obsession with like
British, upper middle class and Parsons and stuff who would
just go out in the country. And what would you do? You would come back, and you
would generate these whatevers of all the different types of
butterflies and just wallow in the variability. That was the main thing of it. Look at all these
different colors. Look at all these
different shapes of wings. The interpretation at the time
of course was, look at all this glorious
variability of life and God's work showing in it. But the main thing emotionally
that these guys were about was variety-- the
range of differences, the range of different ways
that species could look like. And what this new field
was emerging in Europe was ethology. What ethology was about was
the same exact emotional pulse as in those 19th century
Parsons with the butterfly nets. But what ethologists would do
is go out and collect behaviors. And they would come back, and
they would just wallow and just be buoyant by all of
the different types of behavioral variety. And that's the most
important thing. So we can see right
off the bat, they are not going to be
getting along very well with the behaviorists--
with the rat equals boy equals hippo business there. Because what these
guys were all about was just the sheer variety of
behavior and, implicit in that, how you have to study it
in natural environments. So there were three
guys during this period, beginning work early in
the 20th century, who were the official godfathers
of the whole field of ethology. One was Niko Tinbergen.
Next was Konrad Lorenz. The next was Hugo Von Frisch. And these were the big three. They eventually got
their Nobel prizes. And Tinbergen was basically
a saint, an incredibly sort of accomplished,
interesting man who did all sorts of
extraordinarily good things. Konrad Lorenz, as we
know, was Nazi scum. And Hugo Von Frisch was just
really, really, really old. Somehow he was just really
old from the very beginning. And he was about 89 when
he got his Nobel price. These were the starting three
that began the whole field. So we've got our archetypal
behaviorist villains there with their shiny foreheads
leaning over the camera guy and waving pigeons at them. And what are the
ethologists about? All of them were guys with
these knobby knees and very hairy legs, with big thick
old hiking boots covered in mud and alpine whatevers. And what they were
doing was just tromping around in the outside
world collecting behaviors, just looking at
the sheer variety. OK. It's obvious which team
I'm rooting for here. But even, you know, that they
hung out with Konrad Lorenz. OK. The ethologists are
much more nuanced-- the much broader, much
more multi-faceted, diverse view of life there. The ethologists, completely
unconnected with behaviorism in the United States. Instead, we're developing
this whole field premised on all sorts of
very different notions. Behaviorists, this
radical environmentalism. What ethology was
about, from day one, was gene environment
interactions-- behaviorist rules of positive
negative reinforcement. That explains the entire world. As we'll see, some of the most
amazing things ethologists came up with were examples
of animals acquiring new behaviors that broke every
rule on earth of behaviorism about how learning occurs. Finally, over there
with the behaviorists, with their universality--
cat equals hippo-- and what the ethologists
were about was, every species solves its
environmental challenges, and its history of evolutionary
challenges, in a unique way. And out of that came
Tinbergen's famous quote. Which is, again, ethology is
the study of animal behavior, but where you're interviewing
the animal in its own language. And the whole
premise of ethology was, that's
preferable, and that's not going to happen
unless you're out there in the nature in
the animal's real habitat. And you better be
real open minded as to what counts as
communication, as what counts to be the animal's
language that you are talking to it with. One example of this. And this was somewhat of
a different discipline, but in the 1960s, there was
this very interesting sort of cottage industry of
research that came out. Which is, you take
a normal lab rat. And now what you do
instead is put it in an enriched environment. You put it in a big
enclosure with lots of other rats and beeping toys
and interesting smells and cell phones and all sorts
of sounds of that sort. And what they would show
was landmark-- take a rat and put it in an
enriched environment and-- in 1960s neurobiology,
of a type we've heard about already--
you get a thicker cortex. People went wild
over that in terms of enriching the environment. This prompted a
gazillion studies during the '60s built around
this very Great Society '60s optimism about things. The Head Start program
emerged during that period out of a demonstration that
all sorts of privations in various points of
life, early in life, could be buffered,
reversed, by the right sort of enriched environments. Incredibly influential. Billions of studies done. And hidden away
in the literature was one very interesting study. Which is, when you took these
enrichment sort of Poobahs there-- and now, what
they did in this one study was they studied wild rats. And they brought in wild rats. And what they saw was, even
without the enrichment-- you didn't need to do a
thing about enrichment. You take the best environmental
enrichment laboratory paradigm to get you the
thickest cortex around, and it's not going to be
as thick as in a wild rat. And these guys were not-- these
guys were not ethologists, but this is exactly
the sort of thing that ethology is built around. You were not going to be seeing
normal behavior in a cage, looking at a rat in there. That's like studying dolphins
swimming in a bath tub. You are going to be missing
everything under those. And, whoa, there is
stuff going on out there. If a wild rat has
a thicker cortex than we know how
to come up with, we need to be studying animals
in their natural setting. OK. So that was their basic
sort of call to arms. What ethologists worked
out were a number a very clear experimental
approaches built around a whole bunch of the
W questions-- who, and what, and where, and why, or whatever
it is we did in fifth grade. Here's what they looked at. They looked at, first,
what was the behavior? What was the behaviors? We'll see they had a certain
jargony term for it-- fixed action patterns. And what we'll see
about it is this is a very specific realm
of types of behavior that had a very interesting
subtle relationship with things like instinct. Next. What in the outside
world triggered that behavior to occur? And their jargon
stuff, terminology-- what was the environmental
releasing stimulus that caused this fixed action
pattern to come out? Next. Dramatically different from
where the behaviorists were-- what's going on in
that organism's head so that that releasing stimulus
triggers the fixed action pattern to occur? What are the
intervening mechanisms? What are the innate
releasing mechanism, was the jargon
they came up with. Very jargony, middle
European stuff. But what's going on
in an animal's head? What's the behavior? What triggered it outside there? What's the machinery that
took that releasing stimulus in the environment and
turned it into the behavior? And finally, a realm
which connects them with evolution type
people, what's the value? What's the adaptive
value of that behavior? And what we'll see
is that question meant an utterly different
world to ethologists than it did to evolution people. OK. So starting off, those
fixed action patterns. Fixed action patterns,
when you first begin to learn how
ethologists thought about it, you immediately sense they're
talking about instincts. They're talking about reflexes. They were definitely not. And terms like instinct had
incredibly dirty connotations among psychologists and animal
behaviorists around that time, just because, like an
earlier period, animals, humans had instincts
for maternal behavior, and instincts for
inevitable aggression, and instincts for
entrepreneurship, and instincts for-- and it was
a totally meaningless term, because it described
anything that you wanted to say was something
that's supposed to happen and that I'm good at doing. What you had instead,
in their view, was something much more subtle. A fixed action pattern
is not maternalism. A fixed action pattern
is not entrepreneurship. A fixed action pattern is a
relatively tractable piece of behavior involving a bunch
of interconnected muscles and producing this
characteristic behavior which had a meaning to it. OK. Nice and abstract. We will see example shortly. But what they also had, that
separated it from instincts by miles and miles,
was animals didn't have to learn how to do it. It was hardwired,
but animals had to learn how to do it better. And that was an amazing insight
in terms of gene environment interactions. The word gene was not appearing
in their thinking at the time. But fixed action patterns. In other words,
experience could change the way you do the
fixed action pattern. Or, already harking back
to the molecular lecture and that obsession with if/then
clauses, what experience mostly would do is not so much change
the nature of the fixed action pattern but change the context. Teaching you when to
do it, when not to. If/then clauses of, if
this environmental event is happening, then bring
out this cohesive set of behaviors that have at least
some superficial resemblance to instinct but doesn't
hold up for long. This was really important. OK. So what would be examples? What does this look like-- a
behavior where animals simply know how to do it without
learning, without experience? And they can do it
right from the start and nonetheless
experience shapes that. Obviously, first example was
all of Charlotte's babies knowing how to say salutations
within a few seconds of being born there. And no doubt they would soon
learn who not to say that to. Classic example. And this was one that
Lorenz studied for years and always wanted to have the
pig killed because Nazi scum. OK. But anyway, I digress. And other examples. You take a squirrel. You take a squirrel that's
been raised in a cage without ever seeing another
squirrel-- a squirrel that has subsisted on nothing
but a liquid diet and that kind of thing. And give it a nut, and it
will know how to crack it. It will know to carry out
this fixed action pattern. This is not an instinct
for food acquisition. This is something instinctual
for how you hold this thing and where you chew
on it or whatever it is that squirrels do. It's in there. The squirrel doesn't have
to learn how to do it. But what you see is,
you take that squirrel without any prior experience
with nut cracking, and you look at them over time,
and they get better at it! They do it faster. They get a larger percentage
of whatever's inside there. They have a basic, hardwired
fixed action pattern-- a bunch of coordinated muscles
doing a coherent set behavior, and it is shaped by experience. In this case, they learn
how to do it better. Another example. This is one where you have
any of those intro psych books from anywhere in
1950s, '60s or so, and there was this
inevitable picture where you had-- it was this room. There was this surface, and
it had a checkerboard linoleum on it. And the surface
came to around here, and then it dropped
down about three feet and then continued the
same on the bottom. It was always the same
checkerboard linoleum. Linoleum which was like worth
its weight in gold at the time. Linoleum which was going to
make life better for everyone. So you had this checkered
linoleum thing there. And the critical thing
was right around here. When the wall dropped
down, right here was a sheet of glass
that continued out there. In other words, the actual
floor would continue here on the glass, while it was
also dropping down to this. And what was always in
one of those pictures would be some adorable, cute
baby animal sitting out there on the glass,
totally freaking out. Because it's out there, and
it's looking way down there, and there's stuff-- a yawning
chasm below of three feet-- and you are eliciting what was
called a visual cliff response. Which is, get me out of here! I'm floating up in the air here. You did not have to have animals
who already had experience by trial and error
learning that when you're floating in the air,
that tends not to last for long. And you go to
[INAUDIBLE] afterward. There was no-- it's
there in the first place. You take a baby dog. You take a baby human. You take a baby blue
whale, whatever. And you push them out
on the glass there, and they look down, and
they totally freak out. Unless they're a baby sloth. Oh! Isn't that kind of interesting? Because if you're a baby
sloth who gets freaked out by yawning space
underneath you, you are not going to be
a very functioning sloth when you grow up there. Oh, species specific
fixed action patterns. , Where in this case, this visual
cliff could evoke this whole fixed action pattern of
panic responses, all of that. So where does the
learning come in? You push this poor
animal out there enough times-- the 40th time,
the elephant's out there, and it begins to learn. I don't know what's
up with this, because this gravity business
is making sense otherwise. But it seems to be
OK when I'm out here. You could habituate the
visual cliff response. But all sorts of animals--
except for arboreal ones hanging up there--
all sorts of species didn't have to have prior
experience with not liking to drop down open spaces. It was there already. They learn the context better. Another example of this
that the ethologists showed. These were with
captive primates. You take a monkey who
was raised in isolation, who has never seen
another monkey, and at some point
in adolescence. It's a male monkey. And you sit him down,
and you show him a film. And the film
consists of the face of a huge, scary male monkey
of the same species giving a threatening display,
which is usually displaying the canines there. And this is a monkey who has no
prior experience with monkeys. It has never seen
a monkey before. All of its interactions have
been with Barney or whatever, knows nothing about any of this. And you show it that, and
it will freak out and give a fixed action pattern--
a subordinate gesture crouching down and not
making eye contact. Whoa! Where'd that come from? No learning from
trial and error. It was simply there as
a fixed action pattern-- a whole bunch of
things that the monkey does with its torso,
its face, all of that. Where does experience come in? The monkey needs to learn
that you give a subordinating gesture like that to
some big, scary animal, you don't give it to infants. It needs to learn the
right social context. So that's where it comes in. So you've got all these
examples of behaviors that are hardwired, whatever that means. And what you have, nonetheless,
is shaping by experience. And this is what the
ethologists learned. And what was clear after
awhile was all sorts of animals already knew all about ethology. They understood what a
fixed action pattern was. As follows. Type of monkey in East
Africa called vervet monkeys. If you were a vervet monkey,
there are three things on Earth that terrify you. One is a leopard. Another is a snake, and
the third is an eagle. Leopards come from below. Snakes come from below. Eagles come from above. And what you have
in vervet monkeys are fixed action
patterns of alarm calls that they give when spotting
one of the predators. They have different alarm
calls for each of the species. And what has been
shown is, you fly over the silhouette of a
perfectly nice bird there that's not going to
try to carry it away, and you don't elicit it. It's not just things
going on up there. It's a fixed action pattern
for scary, terrifying stuff up above verses scary, terrifying
stuff below with four legs, versus scary, terrifying
below with no legs at all. And they all know how to do
this without prior experience. But what goes on
is the same thing that makes it a fixed
action pattern, which is it can be sculpted
by experience. What am I saying here? Which is, you are a
young vervet monkey, and you've got these
fixed action patterns. That's great! What you have to learn how to
do, though, is not screw up and say, oh my god, there's an
eagle, when you actually should be saying, there's a leopard. Because if you say,
there's an eagle, everyone's going to
run down the tree. And if you say,
there's a leopard, everybody's going
to run up the tree! And you get some kid
who gets it backwards, and that's going to have
some bad consequences. How do you know adult vervet
monkeys have studied ethology and fixed action patterns? Because they don't
listen to the kid until some adult agrees
with them, because they know this is some dumbass kid that
doesn't know the language yet very well and makes mistakes. Under a certain
age, adult vervets don't respond to alarm calls
other than with vigilance, trying to figure out what
the kid's talking about, until it is seconded
by an adult. So what about fixed
action patterns in humans? Classic example--
infant smiling, involving all sorts
of muscles that who knows what's involved in them? All sorts of muscles
that produce this without prior experience. How do you know? Fiberoptic mysteries
show that fetuses smile at various points. What else? You will see smiling
in blind babies. So there's no visual
information coming in. What's smiling? It's a fixed action pattern. Where's the learning? Learning who to smile at. Learning that mannequins
aren't going to smile back at you and things of that sort. That's it being
shaped by experience. More fixed action patterns. Infants are not having to have
trial and error experience with learning there's this thing
you do with your mouth right after you're born that keeps
you from being really, really hungry, and it's called nursing. Oh, can somebody
demonstrate that for me? That's a fixed action pattern. What happens is, kids get more
and more efficient at it-- the number of calories they
can consume per unit of time. They get better at doing it. More fixed action
patterns in humans. Every culture on earth,
people raising their eyebrows and greeting to somebody. Every culture on
earth, people can recognize what anger looks
like, what fear looks like, what disgust, what
contempt, all of those. These are all fixed
action patterns. And where does the
context come in? With all of them obviously
learning when you interpret this as good news, bad news. When you pretend you
don't see it, all of that. Learning the social context. So humans are absolutely full
of fixed action patterns. Next, the next sort of question
that the ethologists would ask after getting
to this point of, what's the behavior-- what does
learning have to do with it? And now asking the
question-- the same one the evolutionary folks
would ask-- which is, well, what's the benefit of it? What's the adaptive value? And a totally different
way of thinking about it than the evolution people. The evolution folks going
on with adaptive value, which is it allows you to pass
on more copies of your genes. And over the generations,
the prevalence of that trait becomes more common. What the ethologists did
instead were experiments. They simply went and
experimented to see what this information--
what these fixed action patterns-- actually meant. So Tinbergen was famous
for one of those. Which was there was like
some gull something or other that he would study, and
the outside of gull eggs were all speckled. And there would be this very
distinctive fixed action pattern shortly after
the chicks pop out, which is mom would go around
systematically turning over all the fragments of egg
where the white of egg shell-- where the white was facing
up-- and would turn it over so that the speckled
part was facing up. Whoa! That's an interesting behavior. Does that make that gull
more attractive to males and thus pass on more
copies of her genes? But, no, here's
what Tinbergen does. He goes in with these gulls
who just finished doing it and with her new pups,
chicks-- her new chicks just having hatched. And she goes off to get
them food or whatever. And this total vicious,
heartless Tinbergen sneaks in, and he turns the shells
back over to white. And what happens? All sorts of raptors
flying around up there are better at spotting
down there that there's all these baby chicks, who
are gone by the time mom comes back. Tragically. I know! And I'm complaining
about Lorenz? [LAUGHTER] OK. But that's showing, oh, why
do gull mothers, shortly after giving birth, go through
this interesting collection of motoric movements which they
don't have to learn how to do, flipping these things over? Does this make
their groups better to out-compete those with
the next Ice Age comes? No. It's because you turn it
over so that some predator up there in the sky
can't see the egg shells-- the adaptive value. Next. Next example of
this-- and this is like one of the most famous,
iconic sort of stories to come out of
ethology-- this was the work of Von Frisch showing
bee dancing, bee communication. And what bees are about
is-- they're scary, and they go find out
about food sources. And they fly back
to their colony. And what they need to do is
communicate to everybody else that they have
found a food source. And they have to communicate all
sorts of types of information. Which direction? How far? How much? Is it totally exciting,
or is it, I just stumbled into a little bit? And what Von Frisch
was able to work out was bees came back and did
this fixed action pattern movement which has been seen
in every bee species on Earth. Which is, they do this
figure eight, a waggle dance. They're shaking their
rear back and forth. And they're moving
over the floor of the hive, where everybody
is watching and cheering them on, and going around and
doing this figure eight. And what is this about? What is this about? Von Frisch was
able to dissect it, and with brilliant, experimental
tricks showing the bee was giving three
pieces of information. The axis of the figure
eight told the bees, with respect to
sunlight, which direction to go to find the source. The longer the bee danced, the
further away the food source was. And the more frantically it
was wiggling its rear around, the more exciting of
a food source it was. This was information. This was what's
the adaptive value. So Von Frisch is sitting
there saying, well, I think it might mean
this, because I'm finding this interesting pattern. I may speculate. What do you do? You do an experiment instead. Which is, now you
take a beehive, and you put it on
a bar stool thing. And it's sitting out
there in the field, and you grab some bee. And I don't know how he did it,
but he lived to a ripe old age. But you grab a bee, and you
give it some amazing food source out here, and you
let it fly back to the hive where it goes berserk with
the waggle dance stuff, telling everybody there's
something amazing over there. Meanwhile, Von Frisch,
in his hiking boots, is slithering on his
belly towards the beehive with the piano stool, and
comes up underneath it, and rotates it--
rotates it 180 degrees. And what happens then? Everybody in the hive
comes barreling out and goes in the wrong direction! Because they were
given directions as to where to go with respect
to the hive's entrance. That's how you show it
actually contained information. And then he would have the
food source closer or further, and show the length
of the dancing and all sorts of
other cruel things he would do to the hive to
confuse them and wipe out the credibility of that one bee. But this is how you
get the information. He would show this
explicitly-- oh, why do they do this fascinating
fixed action pattern? Because they're
telling them something! And here's how you prove it. And you rotate it around. So you can begin to see
the ethologists were really good, clever experimentalists. And they were just
trying to do it in the animal's own language. This became really clear when
you study their next category of questions-- that
business of, well, what is it that just happened
in the outside world that triggered this fixed
action pattern to occur? What was the sensory trigger? Their jargon-- what was
the releasing stimulus? And they had general
strategies for doing this. OK. You were speculating
that this site is the thing that
triggers whatever this behavior in this animal. How do you begin to test it? Let's remove the-- OK, enough
with the abstract stuff. Baby birds-- baby
gulls or something-- peck at mom's
mouth, at her beak. And that is meant to get mom to
regurgitate some exciting food. How do they know how to do this? What is it that gets
them to do this behavior? On mom's beak, there's
a little red circle. And the early ethologists
noting that, oh, that's kind of interesting. I wonder if they're looking
at the red circle there? I wonder if that's the
thing that's causing them to start this behavior? What were the experimental
techniques for saying, oh, is a red circle on mom's
beak a releasing stimulus for this fixed action
pattern of pecking? Let's do an experiment. And the standard
experimental models they would have would be to
subtract out the thing that you think is the stimulus. Take mom and white out
her little red circle there and see if the
kids no longer peck. Replacement. Which is, now you've
done that to mom. And you draw a little
red circle back, and you see if they
start pecking again. Replacement substitution. Now you take something
completely different. You put some neon
triangle on mom's beak, and you see if you will
elicit the same-- oh, what's the specificity of it? Finally, the thing
that they really specialized in--
which is now you would do what they
called super stimulation. Which is, you wave a piece of
paper in front of the chick there-- that you've drawn
a gigantic red circle on-- and you see if it goes out
of its mind and pecks at it like there's no tomorrow. Experimental
approaches like this. If you take away this thing,
does the fixed action pattern no longer occur? If you put it back,
does it work again? If you replace it
with something else, if you exaggerate
the traits, do you get even more of the behavior? These were the
standard approaches for figuring this out. And ethology types
who were interested in this realm of
releasing stimuli, the thing they have gotten
incredible mileage out of recent years is making
little robotic animals. Because you could program
to do certain behaviors, and you could see if you
could elicit everybody else to respond to the behaviors. And, of course, the
first ones were all sorts of techie folks building
robotic bees who could dance in certain dimensions. And they would
stick him in there and wind up the
gear shaft thing. And they would just
dance away, and they could get the other
bees to go flying off someplace-- the bees not
being very discriminating as to which dancers they
pay attention to. But the point there was,
this was a substitution-- a mechanical substitution
for that releasing stimulus. Or there's some amazing
paper a couple years ago, where somebody made
a robotic cockroach and was able to trick all the
other cockroaches into doing some sort of imprudent behavior. And so you see this is high
tech versions of this approach. Oh, it might be this stimulus
that's-- let's replicate it. Let's make it even
more exciting. So using this approach,
the ethologists would begin to figure
out, what is it in the sensory world of this
animal that caused that fixed action pattern to occur? And it was never more
than in this domain that they were doing that whole
sound byte about interviewing an animal in its own language. Because what it's all about is
recognizing other species out there are functioning
in sensory modalities we can't even guess at. What they began to see. One example-- the sorts
of auditory stimuli that would trigger
fixed action patterns. One example-- deer moose. Big things. What are they called? Elk! Elk! Is elk a female
name for-- bulls. OK. So you get Bullwinkles there. And what he's doing is he
is trying to find a mate. And how does he do it? What they do, when they use
northern Minnesotan beasts, is they bellow. They give these
bellowing calls that are heard for miles and miles. Male deer-- I guess it's deer. OK. They bellow. It could be heard miles away. And if you were the
right kind of female, and you hear some guy
bellowing in the next valley, do you know what you do? You ovulate. Auditory induced ovulation. And not only that, you begin
to go try to find the guy. And when you get there,
you start this whole courting display thing, if
he's the right kind of guy and he expresses himself. And what you have there is--
what's the releasing stimulus? An auditory one. Even more bizarre--
discovering that rats, rats are responsive to
certain types of stimuli. For example, you tickle
them on their rib cage, and they giggle. Yes! Your tax dollars going on that
that could instead be spent on nuking something or other. What you've got
there is rats giggle. They giggle in an
ultrasonic range. So you can't hear it until
you've got some rat giggle decoder thing. And they giggle in
response to logical things. Like you tickle them,
and they giggle. And then they come
back for more. And you show all those
patterns with it. And if you get it at the
right frequency only, and they do this
giggling chirping thing, and you're releasing that
there, and other rats come over to check it out. Because the sound-- the
auditory releasing stimulus, in this case of rat chirping
giggling-- induces this, whoa, let's go see
where the party is response in the other rats. Another example of
this-- interesting thing that you see in various
species, including humans-- which is when
females are ovulating, their voices go a little higher. And studies showing that males
subliminally can detect this. More evidence of modality
supplying information we would never even guess at. OK. Then you can show releasing
stimuli in the visual domain. And one example of this
is that you have turkeys, and turkeys apparently do
their sexual attraction stuff visually. And this prompted one of the
cruelest, most savage studies I've ever seen, which
is trying to figure out what is it that
makes female turkeys attractive to male turkeys. And what they did was,
they would show the male, and they studied his
behavior for a long period. And the scientists
began to guess at what were the releasing
stimuli, the visual information from the female, that
would elicit this behavior? So they did the subtraction
substitution technique. They made themselves an
artificial female turkey. They took a big old
ball of Styrofoam, and they stuck this fake
turkey head on at one end. And they stuck a bunch
of feathery things at the other end and just stuck
it out there in the field. And, of course, the
idiot male turkeys instantly are like all around
the thing and courting it. And all, oh. So interesting visual stimuli. What about it actually
attracts the male? And here's where
the cruelty came in. They would begin to mess
with the artificial females. And I once saw a film of this. And like the empathy you feel
for this poor turkey guy there. Because they have a--
and he comes in there, and they've got the Styrofoam
female with this stuff there, and he's courting it
and having a great time. And now they put him
back there, and they do something or other there. And he comes out,
and the back feathers are kind of perpendicular
to the head. And that doesn't-- so you see
he sort of stops there and kind of circles around for a bit. But, being a male, he
decides what the hell and starts courting the thing. Then they put him back
in the room there. And now they let him out. And the tail and the head
are on the same side. He stands there, looking
a little more confused, before starting to
court the thing. Put him away, come back in,
and the head is over there, and the feathers are there,
and the Styrofoam ball. And you see he's kind
of like looking around and goes up to this and-- well,
that's not going to work-- and goes over to the next one. And you're seeing what
the information-- can you believe how cruel this was? You could begin to see this
sort of information-- what were the building blocks of
that visual releasing stimulus-- in order to find out about that. Then there is, not
surprisingly, a whole world of olfactory releasing stimuli. And we won't spend
any time on that here, because all
that stuff is going to come big time in the
lectures on sexual behavior-- pheromones, information
that triggers behaviors, sexual behaviors in all sorts
of species, including humans. Olfactory information. One example with
humans, which was this very cool study that was
published a couple of months ago. What you need is sweat
from terrified people. So here's what you do. You throw somebody
out of an airplane who's never parachuted before. And the whole thing
is, I guess you can do your first jump in
tandem with the instructor who's attached to you and
who knows what to do. The instructor who's doing
all the physical work while you're just there attached
to them being scared out of your pants. Then you get a control
group-- a control group who, instead, have generated
stress sweat on an exercise bicycle. Totally fine. So quick, you get the guys
lying down in the field. They're hyperventilating. You run at them
with cotton swabs, and push his armpits into it,
and get some of his sweat, and seal it away. And then you get the
person on the bicycle, and you get some of their sweat. And now you sit
down people, and you let them smell the two sources. Objectively, people cannot
distinguish between them. Put the person in
a brain scanner, and give them the odors. And if they're smelling the
sweat from the terrified person, their
amygdala activates. Not if you smell the sweat
from the other individual. And not only that,
you now show them faces where the facial
expression is ambiguous. It's been sort of merged halfway
between a scared appearance and a happy one, between
different facial expressions. And you get somebody
after they've smelled the scared
sweat, and they interpret faces as more frightened. Whoa! Smells? The smells of people's armpits
and whether they were terrified or not changes how
your brain is working and changes how you judge
pictures of facial expressions. We've got stuff going on in
all sorts of sensory domains that we can't even
begin to guess at. Showing this even more, other
species where you go into stuff we don't even know about. Electric fish. You want to understand releasing
stimuli for electric fish, you've got to
understand electricity, and how they sing to each
other in electric fish songs, and how they-- where
is that coming from? Oh, right, they're singing
to each other in electricity. And, yes, you get
territorial songs. You get specific ones. You get males who are
courting the same female, trying to jam each
other's frequencies. This actually is
demonstrated quite readily. You see relatedness of frequency
patterns between siblings. They're going about
this with electricity. Whoa! What is happening
in the outside world there that we haven't
a clue at interviewing an animal in its own language? Vibration. All sorts of insects communicate
with each other by vibrating. Arachnids, there's all sorts of
things where they communicate-- I don't know what they
want to talk about-- but by vibrating a web. They can produce
distinctive patterns of it, which sends information. And oh, OK, weirdo insects
and spiders and stuff. Really interesting work
done in recent years, by somebody who used to be
here on campus named Caitlin O'Connell-Rodwell,
looking at elephants. And it turns out
elephants have all sorts of interesting pressure
transducer receptor little corpuscle
thingies in their feet. Nobody else has stuff like that
in their feet among mammals. These are things that
are supposed to be in other parts of the body. You don't see them at
the bottom of your feet. And what are these about? When an elephant walks,
it's causing tiny, tiny bits of vibration in the ground
that could be picked up hundreds of yards away. And what she has
shown experimentally is elephants can communicate
with each other with vibration through their feet. And she has done some brilliant
classical ethology studies of manipulating it and seeing if
you could change the elephant. They're talking to each
other with vibration. All of this very, very
different than we would guess. Tactile stimulation. And out of this came one of
the iconic experiments and one we will come back to lots
in lecturers to come. This one you've also seen
every single intro textbook out there, which is the
surrogate mother monkeys. You got the monkey who's
raised in social isolation, and all he's got are two
surrogate moms to choose from. And one has this
chicken wire tube torso thing with this little
unconvincing Styrofoam head put on top. And the other one is wrapped
in this nice, warm, terry cloth around its chicken wire
torso, with a head on top. What's the difference? The first one, the
chicken wire one, has a bottle of
milk sticking out. That one supplies milk,
supplies calories. This one supplies warmth. And in the mid-1950s, you
would ask somebody, well, why do babies like
their mothers? Because their mothers positively
reinforce them with milk, says BF Skinner, who clearly
knows endlessly about that. What is bonding between
mother and child about? It's about the mother meeting
the caloric and temperature needs of a child. By reinforcing the child
with milk now and then, the child becomes attached. And this is what
dominated in the field. OK. Nice, abstract question. Interesting realm where it
had an implication-- you take an infant, you take
a very young child, and you put him in
a pediatric ward for a couple of weeks
because of some illness. And what's the
philosophy at the time, derived from
hard-nosed behaviorism? It would say, dog equals
cat equals monkey. It would say, one person
with a bottle of milk equals another person
equals another. Parents, they don't
know the routines here. They never wash
their hands enough. So what we're going to do,
what was standard practice in pediatric wards
at the time, was parents were allowed to
visit 30 minutes a week. And otherwise, all of
the care was provided by people working there. And eventually,
the best thing ever came along that would make
any behaviorist delighted. You could do things with
infants, keeping them going, without even being
touched by human beings. Someone modified
chicken egg incubators and invented human incubators. And an entire just
grotesque literature shows the more incubators
a hospital had, the shorter the life
expectancy was of infants who came in in their ward. But that was the dominant
notion at the time. What do infants like about mom? She gets calories. She gives them calories. So OK. So you've got the
baby monkey there. And she's got two moms to choose
from-- the one that gives milk, and the one that has the terry
cloth with the tactile stuff. And which one is she
going to cling to? It's clear, if you had
BF Skinner in there, he'd be up there nursing
on the milk bottle within a couple of seconds. But what does the
baby do instead? Which one does it bond to? The warm, terrycloth mother. Ah. Love has so much more to
do than just calories, and adequate clothing, and a
vaccine every now and then. This was a study by a man
named Harry Harlow which transformed-- began to get
people to think utterly differently about the
nature of bonding. We will hear tons about
that work down the line. So tactile stuff. Finally, what is the realm of
the most brilliant combination of various releasing stimuli
triggering fixed action patterns out the wazoo? What baby animals look
like in every species out there short of
crocodiles and snakes. All the baby animals and
all the different species have the little short
muzzles, and the big round eyes, and the big
shiny forehead, and the big ears
sticking out there, and maybe a
distinctive coloration that sets them off different
from mom, from adults. And every species, you take some
incredibly cute baby monkey, and polar bears are going
to say, oh, look how cute! And their eyes are going
to dilate, their pupils, and you are going to
elicit cute responses. The term, this set of
fixed action patterns that people and all
sorts of species do in response to the releasing
stimuli of a cute baby. And, I kid you not,
the early generations of Disney illustrators
studied this. They studied what it is
that makes baby faces baby. And an amazing paper, classic
one by Stephen Jay Gould, talking about Mickey Mouse,
the evolution of Mickey Mouse. If you go back to when
Mickey Mouse started off in, I don't know,
the 1920s or so, his name was Steamboat Willie. And he was this rat. He was this like skinny,
angular, ectomorph rat thing with five fingers. And he was basically sort of
this double entendre-ing sort of mildly sexual
aggressive beast, who was always ripping people off. And somehow he turned
from that into our Mickey, who delights people. And what Gould showed
was-- in this paper, he actually showed the evolution
of Mickey Mouse's muzzle as over the decades, showing
how it was getting smaller, and the ratio of the
forehead to this. And somewhere along the way,
Mickey lost one of his fingers. And somewhere along
the way, his voice got real prepubescent,
which it wasn't early on. And suddenly, the oh so chased
mysterious Minnie shows up, and all they ever do
is sing and dance. And this is an amazing paper
on how the Disney artists eventually learned what is
it about baby animals that makes everybody want
to be around them. So we see here the
releasing stimuli. Next thing, after the break,
what's going on in the head in between the
releasing stimulus, and out comes the
fixed action pattern. OK. So five minutes. Two great questions. The first one, going back
to Niko Tinbergen's attempts to get baby gulls eaten--
that whole business about flipping the
eggshells over. Somebody asking, so why
haven't eggs and birds and stuff evolved so that
the inside of the egg shell is speckled also? And the answer is,
I haven't a clue. Does anybody know why? Is there some good reason
why the inside of egg shells have to be white? Calcium or-- well, there's
got to be some good reason. But you go out and test that. OK. So that would have been
an interesting study. I have no idea why. The other one, of
the same sort of bent of the, how do they figure this
out-- how do they figure out that human males can detect,
like the sound of women's voices when they rise around
the time of ovulation? What you do is
you record voices, and then you manipulate
the pitch that it's at. And this is a very
subtle effect, but people listen to it. And if they're men and they
listen to it, on the average, they will rate the voices that
are a little bit higher as more appealing. We will hear plenty about
stuff like that down the line. OK. So what do we got so far? We have looked at interviewing
an animal in its own language and asking what the behavior
is about-- the effect of experience on the behavior. Looking at what's going on
in terms of adaptive value in a totally different way
than the evolutionary folks would be looking at. Looking at what-- in the outside
world, what releasing stimulus triggered the behavior. Huge take home message there
being animals, including us, are communicating
information in sensory realms we could never even
guess at-- interviewing in its own language. And also, along the
way there, training ethologists to come up with
really clever, cool, elegant experiments. So the next step. What's going on
here between when the releasing stimulus happens,
and out pops the behavior from the other side? Old ethology jargon-- what is
the innate releasing mechanism? Which is like this really
unfortunate phrase. But what's going on inside? And it was right
around this point that all the classical
ethologists would say, we don't know! It's obviously
important, but we're not able to understand it. And that has changed now. And what that field is
called is neuroethology. And that is the hottest
part of the field in the last few decades,
trying to understand what's going on in the brain that
turns certain releasing stimuli into behaviors
out the other end. And there have been
some amazing successes in this realm in
terms of working out the nuts and bolts of it. One example, the
one that's probably the most studied,
which is understanding the neurobiology of birdsong. Bird song-- how does a
bird learn its species specific song? How does a bird learn
one that winds up being distinctive for itself? What happens to birds
that have to come up with a new song every
year, because they're seasonal breeders in a way
where they're migrating around? How do they learn that? What sort of sensory stimulation
do they need from around them? If you raise a bird
in complete isolation, does it come up
with a song which is exactly what its species does? No, but it's kind
of close to it. It needs experience to shape it. What's going on in bird
species like [INAUDIBLE] birds that imitate the
sounds of other birds? And people have been
dissecting this down to the individual neurons
for years, understanding what's going on in between. Another domain that seems
far more inexplicable for why people would want to
spend decades on it is-- for people
who took Bio core, I always go over
this one-- a reflex. A reflex you find
in female hamsters. Which is you get
a female hamster, and you put pressure, tactile
pressure, on her rear end. And what she will
do is arch her back at that point and produce what
is called a lordosis reflex. So what's that about? It's obviously a
fixed action pattern. Let's ask an adaptive question. What's the advantage of
having a lordosis reflex? What it does is, by
arching up that way, it exposes the
female's genitals, making it easier for
a male to mount her. Well, when is that
useful for the female? If she's ovulating. You only get this reflex when
estrogen levels are elevated. And adaptive value of
that built around the fact that most things
that are pressing on the rear end
of female hamsters are not grad students trying
to get their degree eventually. [INAUDIBLE] ethologically
relevant context, that's part of the mating
courtship display stuff. And there is a guy I know who is
the king-- the king of pressing on the rear end of
female hamsters. People will not be able
to talk about the behavior for centuries to come
without his name being sung about there. And he and his
lab have gotten it down to individual neurons
in the whole reflex pathway, and which neurons are
affected by estrogen, and how exactly it
works-- neuroethology. Another realm of it,
of neuroethology, is one done with
humans these days-- taking advantage of brain
scanning techniques, and seeing what sort of subliminal sensory
information you give someone, and what parts of
the brain change their levels of activity. And the scared sweat versus
the neutral sweat study is a perfect example
of that, and we will hear plenty of those
in the lectures to come. As neuroethology
really took off, an additional
elaboration happened, which is people began to
try doing neuroethology out in the field with animals
in their real habitat. Because all the ones
I just described were the laboratory
animals trying to do it out in the
field, and there's been a number of approaches. One is, people-- a guy
named John Wingfield used to be at the University
of Washington. John Wingfield studies birds. And I have no idea if he ever
actually wanted to study birds or if he was fated to
do this from the start. But what he does is, he studies
the physiology and the behavior of bird species that
migrate from Baja up to the Arctic each
year and back again. Bird species where what happens
with their migratory behavior, if there's an extra
freeze that's unexpected, if spring comes earlier,
what are the-- and he traps these birds. He nets them, and he's got
like electrodes in their heads, and lets them go to fly off to
Alaska, recording from there. And this is neuroethology
stuff-- figuring out the circuitry going on
precisely with that, with birds in their natural setting. Another example, which
is some of the work I've done over the years, which
is studying baboons out in their natural habitat
and trying to do ethology, neuroethology stuff, out there. What, for example,
does your personality-- if you were a baboon--
have to do with the brain chemistry of anxiety in you? You remember the stuff
by now, benzodiazepines and the benzodiazepine
receptors. Are there personality
related differences in the brain chemistry of
benzodiazepines in baboons out there. And you can squeeze
out a surprising amount of information. All of this is part
of the transition back when of saying,
well, something has to be going on
in between there. You're a behaviorist,
and you say, who cares? You're an ethologist,
and you would say, wouldn't it be great
that someday we're going to be able to study it? And what the neuroethologist
actually can do is study it and
begin to work out circuitry maps and all of
that, what's going on inside. Interviewing the brain of an
animal in its own language. Because-- as you will see
in some of the anatomy, neuroanatomy lectures
to come-- you're not going to make any sense about
which parts of the brain are big, and send
projections to which other parts, until
you know something about the behavior
of that species and what sort of sensory
systems it pays attention to. OK. So these are sort
of the big pieces of the ethological approach. What's the behavior? What triggered it? What are the intervening
neuro endo steps in there? What's the adaptive value? What was also the final piece
of the ethological approach was, what does learning have
to do with all of this? We already know that
learning is relevant in terms of those fixed action patterns. But exploring learning in
all sorts of additional ways. And the ethological
contribution to that field is basically twofold. The first would be
boring, classical, sort of behaviorist
type learning paradigms that behaviorists
had explained centuries before. What was interesting, showing it
in domains of animal behavior, where everybody else
would have said, they have to learn
how to do that? That's not an instinct? Showing realms of
traditional learning that were very unexpected. The second piece
was showing types of learning that break every
rule about how learning works if you're a behaviorist. OK. So what would be a standard
example of normal learning? You get reinforced, you're
more likely to do something. You get punished,
you're less likely. The whole song and dance there. Where does just plain old
normal learning-- learning from trial and error--
come in in places that one would not expect? If you are a female monkey,
it is not instinctual for you to know what you're
doing with your first baby. Maternal competence is
not a strong instinct. It does not produce a
bunch of fixed action patterns in, for example,
female rhesus monkeys. They have to learn how to do it. They have to learn how
to hold on to the kid when they're jumping
up in a tree. They have to learn which end you
try to stick on to the nipple. You have to learn all
sorts of stuff like that. Weird. All of my kids starved to
death, but then today I tried switching this
one around this way. Isn't that cool? Anybody know Niko
Tinbergen's email address? What you wind up
seeing there is they have to learn how to do it. So what sort of
evidence would you see? For example, you have mothers
having a higher likelihood of their kids surviving as you
go from their first offspring to later ones. OK. But maybe you can
come up with all sorts of other things going on there. What else do researchers find? If you have a big sister,
and you are a female rhesus monkey-- or a female baboon,
it's been shown as well. If you have a big sister
who has a baby, when you have your first
baby, your child is more likely to survive. How come? Because you watched her. You watched your
big sister doing it. Or if your mother had another
sister, another daughter, and you got to-- no, another
child of either sex-- and you got to watch,
what was even more clear in the literature was
the more experience you had actually holding on to
your niece or nephew there, the more-- as
termed in the field now-- the more [INAUDIBLE]
behavior that you had, the more likely your first
child was to survive. Because you were learning
how to be a competent mother as a primate, as a monkey. It's not, oh, instinctual. They just know how to do it. It's all hardwired somehow. This is a realm where
you have to learn. Another domain
where this happens, and another one where what
you would immediately say is, wait, they didn't do this just
by reflex or instinctually? Meerkats-- meerkats who eat
all sorts of unlikely things. And one of the things
they eat are scorpions. And mom meerkats have to
teach the kids how to hunt and eat scorpions without
getting yourself in trouble there. What is shown-- like
teaching techniques, they would be teaching three
blocks away in the Ed school, in terms of progressively
making the task harder in small increments. What happens? The meerkat mom goes
and kills a scorpion. And for the first
couple of days, the baby learns how to eat a
scorpion that's already dead. Then, once the kid has
that under control, the mom goes and
captures a scorpion. Which remains alive, but mom is
able to bite off the stinger, and now leaves the kid
with this live scorpion that it can learn how to
hunt, and learn how to kill, and get food. And only when
they're doing that, then mom captures a live
scorpion with the stinger still on and plops
it down for the kid. This is exactly what
they teach you to do if you're being a good teacher. Which is a nice, small,
incremental steps. You don't want to do the
same thing for too long. The kid gets bored. You don't want to
have too big of steps. They get learned helpless. And you just-- they're teaching
their kids how to hunt, because the kids don't
know how to do it. Perfectly clear example
of classical learning, but in a domain where nobody
thought it would be going on. Another example of
it-- and this is one where the behavior itself
was so unlikely that nobody would ever have been saying, oh,
animals just know instinctually how to do this,
because nobody believed this behavior could occur--
which is animals making tools. This has been shown in a
number of species, including, for example, cetaceans. Where it's been best
studied is in apes. Other ape species. Actually, we make tools, too. So all of the ape species. And Jane Goodall, pioneer
of chimp research, was the first one to
discover this-- chimps being able to strip the little bark
and stuff off of a long twig and insert it into
a termite mound and pull it up with
all the termites that grabbed onto the stick
when it was down there. And now being able
to eat the termites. And chimps, various types of
populations of chimps, they use tools to hammer,
to break open nuts. They know how to use
a hammer and anvil. They get a flat rock, and
they do this with this. All sorts of tool
making and tool using. As we will hear, there is
at least one population of chimps that has been studied
where males will actually fashion large pieces of wood--
Where they will work to modify it, to use it as a weapon. They can build tools
that are weapons. So what's going on with that? And all sorts of studies
showing, for example, the more experience you have sitting
and watching somebody else make these things, the faster
you're going to master it. But you actually want
hands on practice. The more the individual,
the other chimp, gives you the
building blocks of it to start doing the
piece and all of that, and you begin to master
it learning by experience. One thing that is
very amusing is that it has been shown
that chimps, the daughters learn much, much
faster than the sons. Because the sons don't
pay attention to mom when she's sitting there
and trying to get them with their hands to do this. The daughters are
much more attentive. What you also see is, the larger
the social group, the earlier in life the kids
learn the tool use. And this is brilliant
ape behavior, brilliant ape learning. Monkeys don't do this. I have seen baboons--
films of them, actually. I have seen baboons on films
where they are watching these chimps doing the
thing with the stick there, and going down,
and out come the termites. And do it again and again. And finally, the chimp is just
gorged with all the termites and staggers off for a nap. And the baboon comes over
and will pick up the stick and shake it at
the termite mound or push it against
the side of it. Monkeys can't do this. Apes can, but they don't
do it instinctually. They gotta learn how to do this. So another domain of learning--
learning purely of the, we know how this
works, we just didn't think it would be
going on in this realm. But then, what
ethology has been most amazing at is showing types
of learning that are simply unexpected, that break all the
classic rules of reinforcement theory. One example. One example of it is what is
now called one trial learning. Which is another
term that I think I used the other
day, the one that's sort of more-- sort of a classic
that everybody learned back when, that animals will
imprint on something after one round of exposure to it. And the classic example
would be Konrad Lorenz and his famous little duckies. That when birds come out, there
is a short period after which where they imprint on any
big thing moving around, on a fairly safe assumption
that, in most cases, that big thing is
going to be mom. That's how they bond to mom. There is a short
window, what is now known as a critical period,
where the bonding goes on during that time. And whatever big thing is
moving around-- and eventually, people were doing
classic ethology studies with subtraction and
substitution of stimuli and running around with
like a stuffed Big Bird. And they imprint on
that, and all sorts of large mobile objects--
basically anything that moves. And that's where you would get
all the little duckies, goose stepping behind Konrad Lorenz
when they had imprinted on him. And this is not by
trial and error. Oh, I initially got
bonded to this thing with sharp teeth and claws. And that was kind of a bummer. That involved all
sorts of blood loss. So I've learned, don't
bond to that sort of thing. Well, then I tried bonding to,
are you my mother bulldozers? And then I tried
bonding to this. And, oh, eventually you
figure out your mother is the one you should
be hanging out. No. This is not a domain where you
can learn by trial and error. And what is seen instead
is this phenomenon of one trial learning. The first thing baby
birds are exposed to in many species, the first
big thing moving around, statistically likely to be mom,
that's who you get attached to. And lots of interesting
neuroethology has gone on over
the years trying to figure out what are the
mechanisms of imprinting. One of the things
that come out is it's nowhere near as dramatic
of critical periods-- if you don't figure out who
mom is in the next 7 and 1/2 minutes, you are up the creek. You will never-- it's more
a sort of relative term. But nonetheless, that's
not how learning works. You're supposed to learn
from trial and error that snakes, and leopards,
and all of those guys are not good mothering models. But it happens. Another realm. Something that came to be
known as prepared learning. Which is, there's
all sorts of ways in which some piece of
information can be learned. Or there's all sorts
of sensory associations you could make with
something, but you are prepared to make certain
of those associations more easily than others. What do I mean by this? A classic example
of it is what's called the sauce
Bearnaise syndrome. And sauce Bearnaise is
some fancy French food that seems very unappealing. And I don't know
what it's made of. But this was first described
by a guy, University of Pennsylvania,
Martin Seligman, one of the most sort
of major figures in psychology in the
last decades or so. We will hear plenty
about him to come. And this was, one evening
he went out for dinner, where he had sauce Bearnaise. And then he went to a concert
of some opera or something, and then he went home. And for some reason or other, he
got himself a terrible stomach ache. Terrible stomach ache. It was a total drag. All night long, he had it. He didn't throw up. It wasn't that bad. But he had a terrible stomach
ache, recovered from it. Weeks, months later,
he goes out for dinner. And he goes off to
this French restaurant, and has sauce Bearnaise,
and is about to take the first spoonful, or forkful,
or handful, or whatever. And he realizes this smells
totally repulsive to him. He could not eat
the sauce Bearnaise. He has never been
able to since then, because he's associating it with
the stomach pains afterward. Remember, he didn't throw up. He wasn't burping up sauce
Bearnaise vapors or anything. So it's not like, oh,
the taste was just like the acidic stuff that came
roaring up the other direction, just associating with
the stomach pain. OK. That's sensible. I mean, that's not great,
but you can kind of see how that happens. But you think about
it, and that's breaking all sorts of
classical rules of behaviorism and classical learning. Which is, if you were going to
make an association-- you can have a choice between an
association of a with b or a with c-- you're going to
make the association more readily with whichever one
comes closer in time to a. So what came closer
in time for him? The opera. He finished dinner, and
then he went to the opera and spent hours listening
to Wagner or something, and he didn't attribute
his stomach ache to that. The next time he went to
an opera, he was perfect. He was just fine with that. He had prepared learning. For pain coming
from your gut, you are far more likely to associate
with food than with music. Prepared learning in us
for pain in our GI tract with gustatory stimuli. More examples are
prepared learning. So now you take
these, and you're trying to teach them where
there's a food source. Experimental stuff. And you give them
some like mark or sign or whatever to tell them,
here it is, and remember this for the next time. Do they make an association
between the food source and whatever marker you put up? And what they show
in the studies is, they can modify the marker. They either modify the
shape of it-- here's a little triangle, square, or
whatever-- the color of it, or they can modify the smell. The smell that they-- some
smell source that they-- And what they show is, bees
have prepared learning. They learn the smell
associations much, much faster than the shape or the
color associations. They can learn of
the other ones, but bees have prepared learning
for making associations by olfactory cues more
readily than making them by color or shape. More examples. One of the great
urban legends is that humans are innately
scared of spiders and snakes. Humans are not, because
there's all sorts of cultures where people love spiders
and where, in fact, they love spiders. Because they love
to chase after them, and pull their legs out, and
then cook them up and eat them, and people are not scared
in front of spiders and getting the
willies from them. It is not innate. But humans show prepared
learning for being scared of spiders and snakes. Which is, you can show in
humans and in monkeys, as well, the conditioning that's needed
to associate, for example, a picture of a spider with
the autonomic response to a mild shock, versus
a picture of Disneyland and Mickey and associate it. That it takes less of
an association for we primates to associate something
like a spider or a snake picture with something
unpleasant than all sorts of other animate stimuli. More examples of how this works. You give people--
or, as it turns out, you give monkeys-- a task. You flash up a complex
picture, and you take it away for a few seconds or whatever. And then you flash it
up again, and one thing has changed in the picture. It's a street scene,
wherever it is. And you need to pick
out which it is. And with the monkey
studies, you need to indicate sort of where
it was to get the reward. Are you able to pick up the
subtle thing that changed? Humans and monkeys
are better picking up for animate objects
that change, are better at spotting if there
are things like snakes, than if they are animals that
we don't associate like that. We are better, in this
very rapid exposure, almost subliminal stimuli, we have
prepared learning for that. Do the exact sort of
thing with humans, where what changes from
one flash to the other is either an animate object
or an inanimate object. Humans have prepared
learning to pick up stimuli of animate objects
much, much more readily. A shorter exposure time, more of
a subliminal cue, all of that. So prepared learning is being
all of these examples where, by any logic, universal rules. You learn to associate this
with this, as BF Skinner, or just as easily with this,
or just as easily with this. And dog, and cat, and
they're interchangeable, and sensory stimuli. But no, all sorts of
species come already wired up to learn certain
associations more readily than others. So completely
violating behaviorism. OK. Final domain where ethologists
think about interesting stuff is one of those ones that used
to be just the backwaters, or one of those were
people would say, there's no way this
could ever be studied. And people finally
got some headway in figuring how to look at
this stuff, which is finally all sorts of
interesting things-- not just studying the ethology
releasing stimuli, neurobiology adaptive advantage of
behaviors, but understanding what the internal cognitive and
emotional life is of an animal. And you can see, it's
right at this point that old classical
behaviorists would just have to leave the room they
were so repulsed by this. Because this was just
like William James navel contemplation. What do you mean what's going on
inside the head of the animal? You measure what it's doing. It produces a behavior, and
that's how you study it. And what instead has come
out is another new branch of ethology known as
cognitive ethology. What's going on there? One of these sort of
landmark ones of it-- of sort of a piece of
it-- came from a guy named Donald Griffin, who
passed away a few years ago. He was in his 90s. He was one of the towering
figures of animal behavior, whatever, over the centuries. What Donald Griffin
discovered was echolocation-- that bats echolocate. Classic, classic ethologist
flavored thing of, whoa! How about we interview the
bat in sound frequencies that we humans don't consider
to be sensory stimuli? And opening up this
whole new world. He was central,
played a key role in World War II in the
development of sonar, working off of some of
the same principles. But he's the person who
discovered echolocation. Just to make all of you who
are seniors feel really upset, this was his senior
honors thesis at Harvard. And, oh! Discovered echolocation. So he passed. And he spent his
entire career on this and doing amazingly
elegant stuff. And people came
to kiss his feet. He was such an amazing
experimentalist. And when he was
about 70 years old, clearly what
occurred to him was, I don't really care what anybody
else thinks at this point. I've got something that
I want to bring up, which is the possibility
that animals actually have awareness. And he published a book
called On the Question of Animal Awareness. And it's clear he published
it then because he had tenure. And he already had tenure
in his nursing home, and he wasn't going to get
booted out by everyone saying, oh my god, did you see what
happened to Donald Griffin? He used to be a real scientist. What happened to the guy? Here he is, sitting
around speculating on, do animals have internal
lives of awareness? And what he did was outline how
what a research program to get at that would be like. And people have been
studying it ever since and showing that animals
have strategic awareness. They are doing all
sorts of subtle really. Really interesting field. So starting it off with that. Closely related to
that was the issue of whether animals
have self-awareness. And we know that dogs
don't, because they will bark at themselves in the
mirror for the rest of time. But do animals have
self-awareness? This was something pioneered
by a guy named Gordon Gallup. And like one of those
brilliant, elegant studies. Here's what he did. He would take a chimp,
a captive chimp, and he would give it
a mild anesthetic, so the chip would be a little
bit drowsy for a minute or two. And while it was
down, he would quickly run in with a magic
marker and put a little circle like on
the forehead of the chimp. Chimp comes to, goes
back to its business, where there is now a
mirror in the room. And at some point, the
chimp comes up to it, because they'll wander
around and encounter this. And here's the critical
question right now. Can the chimp figure out that's
not another chimp, that's me? And what the evidence
was that he produced to suggest this was precisely
the case was showing the chimps were now far more
likely than at the chance level-- you see how
it's being set up, this is a well-designed
experiment-- far more likely to
look in the mirror and scratch here to
see what is this thing? Where did this thing come from? And that has been viewed
as the gold standard for self-awareness. Do you look at that
and respond as if you were knowing that that
is you rather than, that's some other animal? And self-awareness
by this test has been shown in all sorts of species. Elephants by this rule
have self-awareness. I don't know who puts
the magic marker thing up on their forehead. But when you do that, they
will bring the trunk up and investigate it. But interestingly,
marmoset monkeys-- those delightful, pair bonding
for life South American monkeys-- marmoset monkeys
don't have self-awareness. And that's kind of
puzzling, because all sorts of other primates
that in the rough way are as neurobiologically
complex had it. And they don't. And, OK, so Marmoset
monkeys don't. But then an ethologist, a guy
named Marc Hauser at Harvard, thought in the
monkey's own language. And knowing marmosets-- he was
a marmoset researcher-- in order to spot this on your forehead,
if you were a monkey, you have to be looking
in the mirror something very close to looking in
your reflection's eyes. And marmosets don't do that as
part of their social system. They never noticed
there was this up there, because they never look up at
that part of their reflection. So what he would do is, he
would now do the same paradigm and put the little dot here. And marmosets, off
they are a minute later trying to figure out what
this thing was about. Interviewing the animal
in its own language. What else? Another domain of this. So awareness, self awareness. Another thing now shown in other
species called theory of mind. One that psychologists,
developmental psychologists love. Theory of mind. At what point when
you were growing up do you suddenly
realize, number one, there are other individuals? There are boundaries
to individuals. You are not one sort
of ego boundaryless sort of continuation of mom. There's other
individuals who could have different information
about the world than you do. And that usually comes
around age 4 or 5 or so. My children got theory of
mind when they were 3 and 1/2. And we can document
that if need be, but it comes out
somewhere around then. And what you see-- it's a
classic test that it's done. You tell the child a story,
OK, here's the-- some of you will know this. The Mary Ann, the Barbie Ann? The Sally Anne. OK. The Sally Anne test
of theory of mind. So you tell this kid
this story about, here's this kid who has
their doll named Sally Anne. Sally? OK, Sally. OK. So they have this
doll named Sally Anne. And you tell them-- they're
totally attached to Sally Anne. They go to sleep
with it, and always on the bed, and all of that. And the kid goes off every
day to preschool or whatever. And this is the day that mom
looks at Sally Anne there and says, oh no, Sally
Anne is all soiled, and it's time to throw Sally
Anne in the washing machine. So in goes Sally Anne. And Sally Anne is
then on the dryer, eventually sitting
on top of the dryer, when child comes
back from school. And now you say to the
kid you're testing, oh, where's that child going to
go and look for Sally Anne? If you have theory
of mind, you will know the kid doesn't know
that Sally Anne wound up in the washing machine, because
she was off at preschool. And you would then say,
oh, still in the bedroom. On the other hand, if you
don't have theory of mind, it's inconceivable to you
that the child in the story wouldn't know, because you know. And they must know the
same thing, because we all know the same thing. And you would say,
the child will go look on the washing machine. That's how you begin to see
the evidence of beginning to understand that
another individual has different information
than you have. And people were soon
showing this with chimps. Here would be one example. That's not a smile. That's a banana. So you would have one chimp
here and one chimp here. And here's how you would do
this paradigm, which is this would be a high ranking chimp. This would be a low ranking one. Put him in there,
and they're watching. And an experimenter
comes along and puts down a really cool piece
of food and in a way that the chimps can't
necessarily see. But the main thing
here is that there are these dividers in between. In one case the divider,
near the dominant individual, is glass. They can see through it. In the other case, it's opaque. They can't see through it. So now, you release the
chimps to go for the food. And what happens is
the low ranking guy-- if and only if this
was glass here-- doesn't bother trying
to get the banana. Because he knows the
other guy saw the banana. Put up an opaque one, and
now he will go for it. Design is all wrong here. There's no divider here. So release this guy. And if this was
opaque, he will now go for it, because he
knows the other guy doesn't know there's a banana there. He knows that the other guy
has different information than he does. Parts of the controls you
would expect to see now have this with the glass here,
where the dominant individual sees somebody bring in the
banana, shows it to them, put it down behind
the opaque screen. And take out this
dominant guy, and put in a different dominant
guy, let them loose, and he will go for the banana. Because he knows, oh,
it's not simply the rule that dominant guys who scare
me know there's a banana there. They know, he's the one
who saw there's the banana. This guy doesn't know
there's a banana there. Theory of mind. Or now do it with a
lower ranking individual, and it doesn't matter
to you whether or not they could look through
the glass or not. You're going to
go for the banana, because you're going
to get it regardless. The first bits of
evidence of theory of mind that a chimp understands
that other chimps have different information than them. What the studies
since then have shown is that chimps do not know
how to do theory of mind in a cooperative setting. They can only do
it for competition. They can only do
it when they are highly motivated in that way. But still, this is an ape. This is really quite impressive. Birds can do theory of mind. Apparently, the smartest
birds in the entire universe are corvids, which are ravens
and crows who are wildly smart. And all sorts of
interesting ethology studies that have been done on them,
including doing theory of mind. You show, with these
guys, you give them nuts or whatever it is,
seeds or whatever, which they will hide in places. And if there is another
one around looking at them, they won't hide it. Or they'll put it there. They'll dig it down into the
sawdust there or whatever. And when the other
guy isn't looking, they'll quick get it and
move it to someplace else. Because they will know,
when it's over here, that this guy doesn't
have the information. They can do theory of
mind when stashing food. More things that
animals can do that fall within the realm of this
cognitive ethology stuff, which is animals can
distinguish between intentional and
unintentional behaviors. How has this been
done in studies? One example, you take a
chimp, a captive chimp. And it has some totally great
food item sitting out there that it's about
to get access to. And along comes some lummox
of a human walking through. And in one case,
the human leans down and takes the plate of food
and flings it someplace else. And in the other
circumstance, the human comes along and accidentally
trips over the plate, flinging it someplace else. When the human leans
down and flung it away, the chimp will bang on
the walls a lot longer than when the person
accidentally tripped over it. Chimps can distinguish between
intentional and unintentional gestures like that. And dogs know the difference. Dogs absolutely
know the difference between somebody
who has kicked them and somebody who has
tripped over them. All sorts of stuff
like that, as well. What else? Evidence of animals being
able to plan for the future. And one great
example of this has been with those corvid birds. And here you have a
design where every day, there's two compartments
that the bird can go into. And on one day,
you put food here. And the next day there,
and the next day there. You alternate back and forth. The key thing is, any
time you put food in here, you put a huge amount in here. And at the end of the
day, you clear it out. Whereas when you're
putting in food here, you're putting in
only a tiny amount. So on alternating days,
they get a lot here. The next day, a little bit. A lot here, a little. And what you show, what you see
after not that long of a time, is these guys, on the day
that there's food here, will take some of it and stash
it there for the next day. Ravens and crows will do this. This is planning for the future. They have flexible,
cognitive strategies. Here's what's been
shown in bees. OK. So you do your basic
Hugo Von Frisch deal. Which is you show a bee some
interesting food source. They go back to the hive, and
they dance and tell everybody about it. But this time,
what you do is you give the bee completely
ridiculous, implausible information. You take the bee
out to the middle of the lake in your rowboat,
and you give him the food there, the nectar. And they go flying back to the
hive, and they dance like mad, and everything we
know about ethology predicts what the bees are
supposed to do at this point. But the bees don't do that. Because, in effect, they're
sitting there saying, yeah, right. In the middle of the lake? You know what you're telling us? You're telling us there's food
over in the middle of the lake there. That doesn't make any sense. We're not going
to listen to you. They don't respond if a bee,
thanks to an experimental manipulation, is telling about
a food source in a place that cannot possibly be. Flexible cognitive strategies. Finally, some evidence
in various species, but predominately studied
in chimps, of numerosity. Numerosity, as in having
a sense of numbers having meaning in
and of themselves. This is totally cool,
this experiment. OK. What you do is you teach a
chimp a series of three objects. And you show him
the first three, and they need to
recognize that, if I've seen this one before, I press
a lever, I get a food reward. You train them at the
same time to recognize three other pictures--
another trio, another trio, another trio. So the chimp has now learned
a dozen different trios of pictures and
knows how to do that. If I have seen this trio
before, hit the lever, and I get a food reward. Now what you do is
you make a mistake. What you do is you take--
here, instead of HIJ, you take out the J, and
you substitute it with A. In this case, what you
do is you take out the I, and you substitute it with
B. What's the difference? In this first case,
what you're doing is you're removing the third
object in the sequence and replacing it with
something from the first object in a different sequence. In this case, you're
replacing a second object in the sequence with another
second object in the sequence. And when you do this
quickly, tasks quickly, chimps make more mistakes
with this than with this. What's going on in their heads? They're saying, oh yeah,
this one looks right. No, wait, that one's not right. It's a different number two
than the one for this trio. When it matches the
number in the sequence, they make more mistakes,
because they have partially filed it away. Not only have I seen this
picture before-- not only have I seen this picture
before as part of this trio, but when I see it as the trio,
it's the third one that I see. It's the first one
they have had some sort of numerosity information
coming in there. Another study showing this,
and this one with chimps. And in this one, what
you do is a strategy you've heard already,
which is you record the vocalizations of chimps. And what you do is you record
the sounds of big male chimps giving threatening bellows. So now, you hide the
speaker in some bush there, and you've got
your chimps nearby. And you play the sound of
some big male bellowing, and it's a chimp they've
never heard before. Chimps recognize individual
voices, no big surprise. And what happens is
all the males run over to the bush to investigate. Now make it a little scarier. Play the sound of two
strange male chimps bellowing, giving their
territorial calls. And do you run over or not? You have a whole bunch
of males run over. Now make it even scarier. Three voices of them. And what you see is, by the time
it gets up to there, if there are more voices of novel
males that they're hearing, than the number of
males in their group, they get out of there. And what you'll see is, they're
playing the voice there. They begin, and they add in
another one sequentially. And say this will come
up to four voices, and they'll show the three
chimps are like coming up there ready to investigate. And three voices, and
uh-oh, we're out numbered. Numbered. We're out numbered. And they slither back and go
the other way at that point. They have a sense of
the number four is bigger than the number three. Finally, evidence of
transitive thinking. Not just in nonhuman
primates, but as shown here by Russ Fernals in
the bio department in fish. I'm watching a defeat b. Then b defeats me. I go and give a
subordinate gesture to a. And this has now been shown. And a similar theme
again, it's only done in the context
of competition, highly motivated circumstances. OK. So what do we got now? We've got a completely different
way of thinking about behavior and emphasizing here natural
setting-- experimentation rather than coming up with
those just so stories. The notions of all
sorts of species, including us, functioning
in sensory domains that are unheard of. And finally, all sorts
of types of learning that are ways in which
organisms are not supposed to learn that all of us do. For more, please visit
us at standford.edu.
