[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]


Ants, bees and other eusocial insects cooperate to a high degree.  They also
have a very special reproductive process which causes the sterile female
workers to be more highly related genetically than vertebrates.

This is what Bonnan means by them being "clones".  Strictly speaking, they
are not genetically identical, but are highly related.

However, these two facts are NOT an adequate explanation of their
collaborative / collective behavior.

The genetic issue provides the evolutionary explanation as to why they
developed their highly cooperative life style.   It does not mean that there
is anything special about the cognitive abilities they posses.Put simply,
you cannot say that because they are related, the cognitive task of
cooperating is any easier that it is for a wolf. In Hollywood movies clones
are able to share thoughts and have an uncanny way to cooperate, but not in
real life.

The cooperation of eusocial insects is not based on "thinking" but is based
on cognitive tasks - they must recognize their fellow workers, communicate
with them, and act in ways which may be chaotic at the individual level, but
which perform the cooperative acts.  It is "gang" action in the sense that
each ant or bee is making an individual decision based on the information
available to it - not some grand mental plan that anticipates and models
what their companions might do.   However, gang action can produce amazing
levels of cooperation.

Another way to say this is that cooperation is actually not that hard a
task, from a cognitive perspective.  Ants, bees and the like do it quite
well.  A bunch of rules on individual action, and some simple communication
and sensing methods will take you a long way. Eusocial insects have hard
wired these rules into their simple nervious systems.  Vertebrates like
wolves use far more sophisticated general purpose brains to emulate what
amounts to the same gang action. 

Which are "smarter"?  The eusocial insect nervous system is MUCH simpler and
MUCH more inflexible in terms of learning etc.  But it is very well suited
for what it does.  It is the simplest way to implent this set of cognitive
tasks if you count neural circuits.  You "just" need a few hundred million
years of evolution to get this "simple" system. We have big, general purpose
mammilian brains, so we tend to view intellegence in this fashion.  We must
learn our cooperation rules after birth, and we learn them imperfectly.
Judged purely from an ant perspective we are clumsy cooperators at best.

It is a big mistake to say that ants don't "really" act like a pack because
they are clones.  They are just the simplest way to implement cooperation
rules. They do cooperation without the baggage of general purpose
intelligence, so we tend to view their form as some sort of cheap - possibly
related to their genetics.  But this is wrong - the point is that
cooperation does not require general purpose intelligence IF (and only if)
there is the right set of connections in a simple nervous system.  Their
genetic system puts a large premium on discovering these connections, and
after hundreds of millions of years they have picked up quite a few.

As it happens, wolves are almost as close as to the genetic system of ants
as vertebrates get.  A typical wolf pack has one reproducing pair, and the
rest of the pack are behavoirally sterile siblings and offspring of the
reproductive pair.  The dominant female wolf is analogous to the queen of an
ant colony or bee hive.  The analogy is not perfect because wolves do not
have the same genetic system. Naked mole rats are vertebrates which are even
closer to eusocial insects in terms of their social life.

In the case of naked mole rats or wolves, fairly intelligent mammals have
taken an evolutionary trend similar (weakly convengent) with eusocial
insects.  Since they have complex brains, their cooperation rules need not
be hardwired and instead can be learned to some degree, or based on more
general rules. 

The conclusions I draw are:

-  The cognitive ability, or intellegence necessary to cooperate in a group,
is not as high as is sometimes supposed.

-  Eusocial insects demonstrate that simple (but highly evolved) nervous
system can carry out amazing degrees of cooperation.

-  Wolves or other vertebrates (yes, even dinosaurs) do NOT have to be any
"smarter" than an ant to cooperate as well as an ant.  However, it is hard
to compare what you mean by smarter.  Eusocial insects can it in a few
neurons, but it took a long time to evolve this.  If you want to emulate
cooperation with a general purpose mammal brain it can be quite hard.

-  There is NO reasonable way that I know of to judge whether a dinosaur is
"smart enough" to cooperate by looking at an endocast of its brain.  Dino
brains are larger than ant brains, or even naked mole rat brains. But this
is one of those things in life where it isn't the size that counts.  With
sufficient evolutionary pressure, dinosaurs could have been amazing

-  Genetic relatedness is NOT directly relevant to HOW animals perform the
cognitive tasks that enable cooperation.  It is however crucial to the
evolutionary "motive" for cooperation.  If we find cooperating dinosaurs,
then they will be highly related to one another (at least statistically).


-----Original Message-----
From: Matthew Bonnan [mailto:Z966341@wpo.cso.niu.edu]
Sent: Monday, May 17, 1999 9:22 AM
To: dannj@alphalink.com.au; B.Dol@skn.sc.philips.com
Cc: dinosaur@usc.edu

B.Dol@skn.sc.philips.com said:
brainsize would be a limiting factor in the learning abilities of 
animals etc etc. I guess those people have never heard about ants. 
These incredibly small animals have an even smaller brain, but behave

as groups and have structured techniques for defense. There are even

species of ants which hunt as one big organism (sorry; don't know the

name...). They have a highly evolved system of communication with a 
kind of sign-language with their antenae and use all kinds of 
chemical communication too. 
There are, without a doubt, more examples of little, non-mammal, 
small-brained animals which have a kind of behaviour which we could 
call 'pack-behaviour'."

While ants and other social insects are fascinating creatures, this
is a false analogy to pack hunting.  Ants act together as a collective
whole because most of them are literally clones.  There is little
thinking involved, and some ant researchers (E.O. Wilson among them)
have described ant behavior as organized anarchy.  This is not pack
behavior - most ants have no chance at reproduction (unlike the
vertebrates in a pack) because most are sterile, so if a worker ant
dies there's plenty more of that to go around.  In a wolf pack, if a
wolf dies ... different story!

Plus, ants are not vertebrates.  If we want to acertain intelligence
and cooperative behavior in dinosaurs, we need to constrain our search
for analogies to vertebrates, particularly archosaurs.  But read my
recent posting concerning analogies!  The brain-to-body size ratio is
a useful yardstick - it may not tell you how the living animals
interacted, but it gives you a decent idea of higher level
intelligence.  Since when the brain/body quotient is large in extant
vertebrates they tend to have highly developed cerebrums, and the
opposite for vertbrates like 'gators and snakes, if dinosaurs fall
within the extant reptile range, the simplest explanation is that
their "higher" intellectual powers were probably on par with those of
crocodilians and other reptiles.  To argue against this, you need
examples where tiny brains correspond to high intelligence, or some
other test.  

I second Dr. Holtz's last e-mail - no one is saying dinosaurs are
stupid, but we don't have unequivocal physical evidence to say, "Yes,
predatory dinosaurs were pack hunters like extant mammalian
carnivores."  And what is ignoble about not being mammal-like when
dinosaurs were the dominant terrestrial vertebrates for approximately
160 million years?  Whatever they did, they did it very well.  Damn
asteroid ... oops!  Not a another thread! =)

Matt Bonnan