Toward a Science of Consciousness 1998

Neil Rickert rickert at cs.niu.edu
Thu Apr 23 23:18:25 EST 1998


modlin at concentric.net writes:
>In <6hodma$o2j at ux.cs.niu.edu>, rickert at cs.niu.edu (Neil Rickert) writes:

>>[ so I disagree with Bill ]

>No you don't.

Yes I do.

>>modlin at concentric.net writes:
>>>Hardware design is important in a lot of practical ways.  A design must 
>>>provide devices and channels for information to come into the system and
>>>out of it... sensors and effectors, in biological or robotic terms.
>>>Hardware design also determines how fast computations can proceed, and 
>>>how much information can be stored and manipulated... all very important
>>>to the practicality of solving any particular computational problem.

>>>But hardware design has absolutely nothing to do with the kinds of 
>>>things that can be computed,

>>To the extent that that is true, computation is irrelevant to
>>cognition.  You could do all of the same computation on a Turing
>>machine with no devices and channels at all.  And since there are no
>>channels, the Turing machine would have no world to cognize.

>You aren't disagreeing with me here, that is a totally separate point.

>The person to whom I was responding seemed to think that different 
>architectures would allow different types of computation.  That isn't 
>true.

Of course it is true.  I can do all kinds of things on my pentium
with 32 Meg of memory and 4G of disk space that were impossible on
the Z80 system I owned in 1980 (with 64K of memory and two 100K
floppy drives).

>       All computing architectures can compute the same things, which 
>you know perfectly well.   

But the original comment was about what can be computed in practice.  It
was not an in-principle argument.  And what can be computed in practice
can be very sensitive to aspects of the hardware.




More information about the Neur-sci mailing list