🚀 go-pugleaf

In article <1hpvrpINNsam@shelley.u.washington.edu>, wcalvin@stein.u.washington.edu (William Calvin) writes:
|> m23364@mwunix (James Meritt) writes:
|> 
|> >Bad example.  Catching a branch after a free-fall requires the
|> >"unconscious" solution of a DE.  Catching a ball is a complicated
|> >set of equations.  Now, being able to express it AS equations is recent,
|> >but the ability to solve them isn't....
|> 
|> All animals seem to have the ability to make at least one particular kind
|> of trajectory calculation, also involved in catching a ball:  that of
|> judging whether an approaching object will hit their head.  It produces
|> the blink replex, and cognitive-level abilities need not be invoked (it's
|> a brainstem-level reflex, based on looming).

Yes, the ability to catch things does not require "solution" of a DE 
or set of DE's in the normal sense of the word, you just need a good
feedback control system ... you may need to tweak the gains depending
on what your catching (maybe adaptive feedback control is the more correct
term) but to say that one must "solve" a complex set of DE's is misleading.

|> Now throwing a projectile, and estimating its trajectory in advance during
|> "get set", is indeed a good candidate for a cognitive mechanism.  All of
|> the ballistic movements (kicking, clubbing, throwing, hammering -- even
|> spitting) require lots of advance planning; feedback arrives too late to
|> do any good, so "get set" must create a complete movement command
|> sequence, start to finish.

While throwing activities are better candidates for cognitive mechanisms
I still don't think they *require* anything more than the ability to
interpolate or extrapolate from past experience.

Larry Edwards