Re: Consciousness: what's the problem?

curt@xxxxxxxx (Curt Welch) writes:
Neil W Rickert <rickert+nn@xxxxxxxxxx> wrote:

That approach in my view basically took what humans seemed to do when they
were using language to reason with in an attempt to duplicate our power to
reason, and our power to "know" things and talk about them. It produced
data base systems that could store language based knowledge and answer
questions about the information in the database, but it never came close to
explaining how such a system could create it's own meaning from interacting
with an environment.

I agree with that.

I see the entire problem has one of mapping sensory inputs to effector
outputs.

But it won't be that simple, because the outputs will also depend
on knowledge.

However, the encoding of information about the real world is trivial at the
lowest levels. It's what all our sensors do. They transform events in the
environment into a common internal language which represents those events.

You are going to run into the problem that sensors don't sense.

The perception problem however happens when you try to interpret all this
raw sensory data. The percpetion problem is one of translating this low
level sensory data (which is a lot of "words" that have meaning about the
level of light falling on different sensors in the camera over time), into
words with meanings such as "dog", "cat", "door", etc.

The chances are that there is no easy way to "translate", and it would
depend on a lot of knowledge.

So unlike you, I think creating data from the environment is trivial.
That's what all the sensors do.

A sensor just reports the current state of the sensor.

What does the perception component look like to you? Not a transistor I
guess?

Procedures. Physical procedures, not computational procedures.

Ok, well, can you explain what type of hardware you think we need for your
vision of perception and actions and what laws of simple physics makes that
hardware do something a transistor can't do?

The details of the hardware aren't as important as the procedures
used.

As a simple illustration of the problem consider a clock - let's say
an ordinary analog clock.

The clock doesn't tell you the time. The reading from the clock
merely tells you the position of the clock hands. In order to
tell the time with that clock, you have to know that the clock is
causally connected to the human defined time standards. And that's
procedural. When we periodically reset our clock according to the
time signal on the radio (or wherever else), we are participating
in the procedures for causally linking the clock to society's time
system. And only because of that are we able to use the clock to
determine the time.

It sure feels to me like you require some type of "magic" to show up in the
hardware which suddenly creates conscious awareness.

That "magic" is properly called "knowledge", and it is in the form
of procedures, not just a database of facts.

But until you can
show me the magic your hardware creates and prove to me the function can't
be duplicated with transistors, I'm going to just keep assuming what I
always assume when anyone talks like this.

The problem I ran into, in my theorizing, is that there is a need to
make many pragmatic decisions. You can use transistors for logical
decisions. I don't see how to use them for pragmatic decisions.
In biological terms, what seems to be needed are homeostats.
And neurons might have an advantage there.

But you you do have some very odd views which are hard to follow simply
because they are so uniquely different.

I will admit to having non-standard views. But they do come from some
careful analysis of what I have taken to be the problem.

.

Relevant Pages

  • Re: Thaughts about car reliability.
    ... ECU's, sensors etc. ... Anyone agree, disagree, or think it's a naive idea? ... Mike. ... I usually buy at around 15-18months old with 10-20k on the clock, ...
    (uk.rec.cars.maintenance)
  • Sig change ...
    ... clock. ... park brake warning light came up. ... Apparently some sensors been fitted wrong ...
    (uk.rec.motorcycles)
  • Re: HELP! Twilight Zone "Clock is Broken"
    ... The clock just stops before reaching the current time and the error ... "Clock is Broken" appears when I enter the test mode. ... When the game sets the current time, it winds in one direction, then ... The game only has 4 sensors to tell it where the clock hand is, ...
    (rec.games.pinball)
  • Re: Corvette LT5 problem: No adjustment!
    ... the optos used in the sensors are the same ones ... (and the same circuit) ... that causes so much trouble in the TZ clock. ...
    (rec.games.pinball)
  • Re: Coding style, wait statement, sensitivity list and synthesis.
    ... >> least assume that XST supports this style and create this hardware. ... >>>then the higher precedence clock must be coded first. ... For these very rare dual-edge sensitive register, yes, I think instantiation ... Dual-edge sensitive register elements is not common practice in hardware design. ...
    (comp.lang.vhdl)
Loading