SLOMAN Aaron AGI11
<p>Aug 12, 2011 12:57 PM
 
exclt lecture... worth seeing again in video.
 
Key idea  :  
 
Various environments produce various demands
 
Varied enviros produce varied demands
chem soup with gradients
good stuff, bad stuff, obstacles
things that have to be manipulated to be eaten (disassembled)
controllable manipulators
things that try to eat you
foods that escape from u
mates with preferences
competitors for food and mates
collaborators that need or can supply info
 
John Maynard Smith...  evolution  does not mention above... but the above are critical
 
claws, tongues, jaws, feet...  controllable manipulators
 
octopus can  turn its arm into a virtual jointed arm with elbow and wrist eg.
must represent things that represent you
result of this... 
 
we must also have access to the enviro in which the organism evolved to try to understand it.
compare trying to understand a v sophisticated ALIEN technology... 
we may not have a basis for understanding it
even by cutting it open or scanning it.
julian huxley... the explanatory gap
movie  rotating cube  with painted surfaces  each difft... how to predict what shows?
really helps to know that the cube is 3D
 
18 mont child with broom succession of stills showing problem solving...  disengaging from railing,
swinging handle so that broom cud go thru the door...
 
deb roy's project  on babies  (he mentions)
fascinated by babies...
turning shampoo bottle upside down and squeezing it... must wait for the shampoo
to get to the nozzle before it comes out...
babies are learning abt the properties of objects and their enviro
exosomatic ontolgoies....   ontologies of properties of  your enviro...
like shampoo,  the floor, water  etc.
annette commoroth smith...  beyond modularity...  1992   (she was a student of piaget and
then went off on her own)
she says that piagetian stages are determinede by interaction of child with enviro
gradually gets retransformed into an axiomatic form  (reorganized)
she calls it representaiton redescription
learn it initially empirically and then it gets redescribed in axiomatic form.
diana pitt and michael gibson... a great book on childhood learning
start with empirical learning but then  redescribe in  generative  grammar  a la chomsky
(infinite generativity)
nicaraguan  deaf kids....  illustrates his point that
kids  have  a lot to discuss and will do it by whatever means
*</p>