From: Boston Robert (rtb198@ecs.soton.ac.ukr)
Date: Wed May 23 2001 - 19:12:23 BST
Skywriting comments on "Re: Searle: Is the Brain a Digital Computer?
Godfrey Steve (Tue May 01 2001 - 13:06:43)
>>>Searle: Is the Brain a Digital Computer? (Yusuf Larry)
>>> SEARLE:
>>> So our question is not, "Is the mind a program?"
>>> The answer to that is, "No."
>>> Nor is it, "Can the brain be simulated?"
>>> The answer to that is, "Yes".
>>Yusuf:
>>The answer might be yes but how much does it help us at the end of the day.
>>Just like a plane simulation can't fly, a simulation of the brain is just
>>that a simulation, i.e an imitation of the behaviour of some exisiting or
>>intended system, or some aspect of that behaviour. Hence, some prediction of
>>how the brain would function under certain circumstances or more precisely a
>>model of part of the brain's functionality. We can build a simulation of
>>almost anything to predict its actions or model parts of it without the
>>simulation being based on or being a replica of what we are modelling.
>>Yusuf:
>>In essence, saying the brain can be simulated is a bit of a hazy statement
>>until the parameters are specified
Boston:
Perameters? What perameters, the Church - Turing thesis states that
everthing can be simulated to a close an aproximation as you like.
>Godfrey:
>I agree that the brain cannot yet be simulated exactly as we do not
>know how it works, but I do think that the simulation of the brain is
>not a pointless activity. By simulating how we think the brain works,
>we can see if the activity in our simulation matches that of a real
>brain. By observing the differences in our simulations we can refine
>our model, and maybe get closer to how the real brian works.
Boston:
If we iterativley update and refine a model of a brain until we end up
with one which produces similar results, all we will have achieved is a
model of the brain that produces similar results, probably not a model
that works like the a real brain.
>>> SEARLE:
>>> The question is, "Is the brain a digital computer?" And for purposes of
>>> this discussion I am taking that question as equivalent to: "Are brain
>>>processes computational?"
Boston:
Brains are not digital.
>>Yusuf:
>>In essense, if brain processes are computational, then like computational
>>processes in a digital computer, the brain becomes the digital computer that
>>manages the effective execution of the required processes.
>Godfrey:
>If this were the case, would it mean that the mind is implementation
>independent? We could then run the brain processes on a digital
>computer, and surley the computer would be exactly equivelent to the
>brain. I still do not think that it would have a mind as it would
>still be perfoming meaningless symbol manipulation.
If brain processes are computational then they are, by the definition
of computation, implementation independant. If brain processes and
consiquently the mind is computation then all we do is perform
meaningless symbol manipulation. We do more, when i think of dogs I
think of dogs. Ungrounded computational brain processes could not think
of dogs.
>>> SEARLE:
>>> It is clear that at least some human mental abilities are algorithmic...
>>> It is furthermore a consequence of the Church - Turing thesis and Turing's
>>> theorem that anything a human can do algorithmically can be done on a
>>> Universal Turing Machine... Now it seems reasonable to suppose there might
>>> also be a whole lot of mental processes going on in my brain nonconsciously
>>> which are also computational. And if so, we could find out how the brain
>>> works by simulating these very processes on a digital computer.
>>Yusuf:
>>Yes, but since only some of our mental abilitites are algorithmic and only
>>some to many of the brain processes might be many, this goes to show us tha
>>the brain is not a digital computer, at least not wholly. This is because a
>>digital computer is basically only capable of performing computation
Boston:
Brains are not digital computers.
>Godfrey:
>I'm not sure what is being said here. I think that Yusuf is saying that
>as only some of the mental abilities are algorithmic, a lot may not
>be. So the brain is probably not a digital computer. Would modeling
>these nonconscious activitys be of much use. They must at some point
>interact with processes that are non algorithmic as we are consious of
>the fact that we are thinking. If only a small part of the mind is made
>from these algorithmic activitys, compaired with the rest of the brain
>would they be able to provide us with enough information to make any
>new discoverys from?
Boston:
Assuming parts of the brain are doing things that are 'non algorithmic'
how could you model them? If by non algorithmic you mean not
computational how could a computational device model it? TUring
machines can only do everything algorithmic we can think of. How would
you model pain? If you did create a model of pain how could you know if
it was the right one? A machine implementing a stratagy to avoid damage
using a mindless model of pain could easily do it without feelings.
>>> SEARLE:
>>> We thus have a well defined research program. We try to discover the
>>> programs being implemented in the brain by programming computers to
>>> implement the same programs. We do this in turn by getting the mechanical
>>> computer to match the performance of the human computer (i.e. to pass the
>>> Turing Test) and then getting the psychologists to look for evidence that
>>> the internal processes are the same in the two types of computer.
Boston:
This sounds iffy, psychologists examine side effects of mental
processes (because of the other minds problem).Is it a property of
intelegent systems that the observable side effects are consistant form
person to person let alone person to machine? nope. For a mechanical
computer that matches the performance of a human and passes T3,
psychologists wont tell you anything useful (unless its insane and bent
on the destruction of the human race, this seems resonable, if you met
god and found he no grand plan and had only created everthing just to
see if it could be done you would be a bit mad), you need doctors and
engineers to get to T4 which is not that usefull anyway.
>>Yusuf:
>>Following Turing's thesis of Indistinguishability, we would therefore be
>>able to deduce/assume that the human brain is a digital computer.
>Godfrey:
>Just because we cannot tell two machines apart, does not mean that they
>are the same machine. An electric and petrol engine both are able to
>power a car but they work in completley different ways. I think that
>all we would ba bale to conclude from this test would be that, we
>cannot differentiate between a brain and a digital computer
Boston:
Godfrey's right, it would only prove we cannot tell them apart and have
no basis for saying the machine is not intelegent (or is intelegent)
>>> SEARLE:
>>> Analogously, we might discover in nature objects which had the same
>>> sort of shape as chairs and which could therefore be used as chairs;
>>> but we could not discover objects in nature which were functioning as
>>> chairs, except relative to some agents who regarded them or used
>>> them as chairs.
>>Yusuf:
>>At the end of the day, chairs are things we sit on, so we do not need to tag
>>an item "Chair" before in becomes a chair in the same way that a digital
>>computer is or isn't
Boston:
Yes but we want to tag things intelegent or not intelegent
>Godfrey:
>I agree that the state of an object is not changed whether we label it
>as something or not. But I think that the object can be interpreted as
>a chair by the individual, and each individaul could have a different
>interpretaion about what the object is.
Boston:
I dont like this chair analogy. Objects can be interpreted as chairs by
induviduals but this does not mean they are chairs they may be wrong,
it may just be a strange table. You can not tell if a chair is
functioning 'chairly' (to function in a chair like manner) unless there
is someone there that is useing it.
You can ask if the chair is being used chairly a reply may come back
saying someone is, great it's a chair with chairlyness ( the property
of acting chairly. See above.). But its possible there isnt really
anyone using the chair and the 'chair' made up the reply to give the
impression of chairlyness. How can you tell if the the chair has
chairlyness? You can't this is the 'Other Chair Problem'.
>>> SEARLE:
>>> One beauty of this research program, often remarked, is that we do not need
>>> to know the details of brain functioning in order to explain cognition.
>>> Brain processes provide only the hardware implementation of the cognitive
>>> programs, but the program level is where the real cognitive explanations
>>> are given.
>>Yusuf:
>>I totally disagree with this statement. We cannot be sure if the details of
>>the brain functioning might have an indirect influence on cognition. To
>>state that " we do not need to know the details of brain functioning in
>>order to explain cognition" is like saying we do not need to know the
>>weather in Alabama to predict the weather in Southampton.
Boston:
I agree, the pathways in our brain are reinforced by the process that
run. We do need to understand the hardware because this is where the
program is stored. It seemed Searle made a lot of assumptions, such as
the physical brain is irelevant and Brain processes arn't cognition
>Godfrey:
>I agree with Yusuf. I don't see how this could possibly be true. We do
>not yet know that cognition is not dependent on the architecture of the
>brain. In Searle's chineese room argument, he shows that understanding
>is not soley computation, but there is something else in our brian that
>allows us to understand.
>>> SEARLE:
>>> The implemented program has no causal powers other than those of the
>>> implementing medium because the program has no real existence, no
>>> ontology, beyond that of the implementing medium. Physically speaking
>>> there is no such thing as a separate "program level".
>>Yusuf:
>>I disagree, if i had a robot, the code implementation that makes the limbs
>>move, aid speech etc are all causal. So this statement is exactly true,
>>however a PC just running code isn't really causal apart from to the
>>hardware.
Boston:
I agree with Searle, programs dont exist (you know what I mean) try and
get one to dig a hole. No cheating, robots arnt allowed and virtual
holes are not holes.
>Godfrey:
>But surley the causal powers of the program running in the robot, are
>all dependent on the implementing medium e.g. the robot. If it had no
>limbs then the program would have no causal powers. A PC is not an
>implemented program but a lumb of hardware, any program runnign on the
>PC would have all the causal powers that the hardware allows it.
>>> SEARLE:
>>> In the brain computer there is no conscious intentional implementation of
>>> the algorithm as there is in the human computer, but there can't be any
>>> nonconscious implementation as there is in the mechanical computer either,
>>> because that requires an outside homunculus to attach a computational
>>> interpretation to the physical events.
>>Yusuf:
>>At what stage, was the differentiation between the brain computer and human
>>computer made, because i must have missed it. And based on this, what
>>exactly is this statement getting at?
>Godfrey:
>I am confused by this as well. Is Searle talking about the brain as a
>piece of hardware and the human computer as the whole human? or is he
>saying that the brain computer is the real brain, and the human
>computer is a man made implementation. As it would make sence that we
>have no conscious algorithm running in my head, but I would be
>conscious of an algoritm that I had designed and written.
Boston:
In our heads (brain computers) we are not aware of algorithms
implementing our intelegence and can't have the nonconsiouse algorithms
performing as the results would need interpreting. T2/3 (human
computers) would be aware of the algorithm being implemented and PCs
(machine computers) have no idea what is going on and we have to
interpret the results. Where did Searle discover conscious intentional
implementation of the algorithm in a human computer(T2/3)?, because of
the other minds problem this would mean Searle had to be the T3 minimum
robot.
This archive was generated by hypermail 2.1.4 : Tue Sep 24 2002 - 18:37:30 BST