Are Digital Computers and Animal Nervous Systems Really That Different?

[fredreload]

Is the analog output a voltage, current, or a signal? I'm kind of confused.

 

[anorlunda]

Is the analog output a voltage, current, or a signal? I'm kind of confused.

Within reasonable limits, and analog output is usually a constant voltage source. But you can certainly get analog outputs designed to be current sources. For example, 4-20 ma sources.

"signal" is not a word we use in Ohm's Law.

 

[tfr000]

Is the analog output a voltage, current, or a signal? I'm kind of confused.

Analog output of what?
Usually, it just means "not digital". That is, a continuously variable signal, not just 1s and 0s. Like a sine wave.

 

[bubsir]

An analog output is a voltage or current - if it changes (volts or amps) both are also "signals." For a specific implementation you can set up a "current" or "voltage" based output. A current based output is easily converted to a voltage by driving through a resistor. Voltage to Current requires a bit more circuitry.

 

[Babadag]

An analog circuit is a control circuit including an amplifier , resistances and capacitors in order to imitate a mathematic function as divider, summation, integrator or differential-"analog" to this mathematic operations.They are as:

noninverting op amp, inverting op amp, adder circuit, differential amplifier and other. See[for instance]:

http://www.ti.com/lit/an/slaa068a/slaa068a.pdf

The analog output it is the potential [voltage] at the analog amplifier output used as input of another analog circuit or device as a thyristor gate-for instance.

 

[anorlunda]

Is the analog output a voltage, current, or a signal? I'm kind of confused.

Now you have several answers from people who have interpreted your question different ways. I invite you to try again. Better quality questions produce better answers.

 

[Averagesupernova]

Now you have several answers from people who have interpreted your question different ways. I invite you to try again. Better quality questions produce better answers.

Or as one of my favorite wise posters on PF says: A question well stated is half answered.

 

[sophiecentaur]

Or as one of my favorite wise posters on PF says: A question well stated is half answered.

He is a legend!

 

[fredreload]

I am comparing the brain neurons to an analogue circuit in which our consciousness is the analogue output. The neuron does generate a voltage. So I'm thinking we can't exist as current, but more as a voltage. Some speculation on my part.

P.S. Nevermind, it's the same thing as a computer

 

[sophiecentaur]

The Analogue / Digital thing is not clear cut - particularly not in the brain, I think.
Any signal that passes through a system is basically analogue; it will consist of a varying quantity (Volts, Amps, Temperature, Force). This is true, even for 'digital' circuits. The information that is carried, on the other hand, may be analogue or digital. The volts on a loudspeaker lead will deflect the cone proportionally and produce a continuous analogue variation in air pressure - perceived as analogue sound. The volts on a digital transmission line, likewise will take all analogue values from, say 0V to 1V (for a binary signal. The digital information will be carried by that signal in the value of the signal (greater or less than 0.5V, say) at a time when the decoder looks at the input signal. The 'box car' waveform that is shown in elementary descriptions of digital signalling is not usually found in real digital signalling systems. It is an oversimplification of real life; what you will normally see is a wavy line (on an oscilloscope Voltage trace) and the job of the decoder will be to extract the digital values as the signal passes through. This link describes the basics of neurone signals and it makes the point that the information is carried in the form of trains of spikes. The number and frequency of the spikes is what carries the information, rather than their amplitude (as far as I can see). The way the brain actually perceives this information can either be digital with discrete levels - as when we actually make a yes / no decision or fuzzy / analogue when we appreciate how hot some water is (but we can't actually measure the temperature.
So analogue signals carry either analogue or digital information.

 

[fredreload]

So the neuron can work as either digital or analog? You need digital circuit to produce some sort of functional behavior right?

 

[anorlunda]

Neurons are highly nonlinear and adaptive. I do not think it is helpful to compare them to electrical circuits with linear components nor ones with digital components.

Consciousness is not fully understood, but it seems certain that the ultimate explanation of consciousness will not arise from circuit analysis.

 

[sophiecentaur]

The operation of neural 'circuit' elements is an interesting field of study in the same way as the study of electronic components in a processor. But, as the above post says, the way the information is handled in each case is a very different matter.
I would say the distinction between digital and analogue information is not valid, once a system becomes complex.

 

[fredreload]

I do believe that analyzing circuit plays an important role in determining consciousness. We all know that computer seems to "think", in its way with the CPU, and CPU is made up of complicated logic gates that comprises the integrated circuit. Now if we take such a circuit's functional behavior to the next level and allow parallel processing it just might become a real thinking unit. Consciousness might be something that arises from the circuit's functional behavior based on my speculation, but since anorlunda mentioned consciousness does not arise from circuit analysis, maybe you have something else in mind?

 

[anorlunda]

I do believe that analyzing circuit plays an important role in determining consciousness. We all know that computer seems to "think", in its way with the CPU, and CPU is made up of complicated logic gates that comprises the integrated circuit. Now if we take such a circuit's functional behavior to the next level and allow parallel processing it just might become a real thinking unit. Consciousness might be something that arises from the circuit's functional behavior based on my speculation, but since anorlunda mentioned consciousness does not arise from circuit analysis, maybe you have something else in mind?

Lots of computer scientists are trying very hard to achieve artificial intelligence (whatever that means). Most efforts focus on software. Some of the software attempts, try to borrow ideas from neurons, but they are still software.

It is true that software runs on computer circuits, but the same software can run on many different kinds of computers. That leads one to say that the software's property does not arise from the particular circuits. By analogy, PF can be used from a PC or a Mac, and it looks pretty much the same either way. That means that PF's properties do not depend on the details of the underlying circuits.

You might be interested in this https://en.wikipedia.org/wiki/Artificial_intelligence

 

[fredreload]

I know that the current super computer can only simulate 1% of the brain's functionality in terms of neurons and it would be possible to simulate the brain's functionality one day based on the hardware we will be getting. Maybe having a loop structure and a reversible device along with parallel computing. Computer only have 2 states, 0 and 1, but brain can have more states then that. Anyway, I got my idea here http://www.seas.harvard.edu/news/2013/11/synaptic-transistor-learns-while-it-computes .

 

[jim hardy]

Computer only have 2 states, 0 and 1, but brain can have more states then that.

Would you pardon an old guy's ramble ?

I've speculated for a long time that we are hung up on binary logic. ie two state true-false.
Neurons as i understand are 3 state: yes , no , and 'damfino' ( indeterminate) .
Boolean arithmetic is of course two state.
There exists three state arithmetic as well, called of course ternary. Twenty years ago there was very little i could find on the net about base 3 computers , today it's a niche with lots of activity - try a search on ternary or base 3 computing..

Anyhow i figure AI will require base 3 arithmetic. We already have tri state logic gates.

Two other thoughts while I'm rambling.

I once encountered a logic system that used standard TTL logic, 0 and 5 volts. They tweaked a wired-or circuit to detect whether just one or more than one input was active...three logical states in a binary machine, high low and in-between.
That seemed profound to me
because
Once you allow more than two states the floodgates are open - why not four or five or a hundred or an infinite number?
That's what analog is, an infinite number of states.

Hmmmmmmm. so , why stop at base 3 ? For that matter,
Who said the base has to be integer?
3 lies in between e and pi... if i ever learn to convert numbers to non-integer radix i'd like to tinker with some of the basic physical constants...

from http://web.williams.edu/Mathematics/sjmiller/public_html/105Sp10/addcomments/Hayes_ThirdBase.htm

Among all possible ways of writing the numbers up to a million, neither base 1,000,000 nor base 1 seems ideal; as a matter of fact, you could hardly do worse than either of these choices. Minimizing the number of digits causes an explosion in the alphabet of symbols, and vice versa; when you squish down one factor, the other squirts out. Evidently we need to optimize some joint measure of a number's width (how many digits it has) and its depth (how many different symbols can occupy each digit position). An obvious strategy is to minimize the product of these two quantities. In other words, if r is the radix and w is the width in digits, we want to minimize rw while holding rw constant.

Curiously, this problem is easier to solve if r and w are treated as continuous rather than integer variables—that is, if we allow a fractional base and a fractional number of digits. Then it turns out (see Figure 1) that the optimum radix is e, the base of the natural logarithms, with a numerical value of about 2.718. Because 3 is the integer closest to e, it is almost always the most economical integer radix (see Figure 2).
......
As it happens, the first working ternary computer was built on the other side of the Iron Curtain. The machine was designed by Nikolai P. Brusentsov and his colleagues at Moscow State University and was named Setun, for a river that flows near the university campus. Some 50 machines were built between 1958 and 1965.

Idle speculation.

Have fun

hope i didnt intrude

old jim

 

[sophiecentaur]

Would you pardon an old guy's ramble ?

As if?
The reason for the success of binary logic circuitry and processors is that it's so easy to produce binary memory and binary operators. That's all, afaics.

The fact that we so frequently use the term "on the other hand", implies that we are always taking into account more than two outcomes. Fuzzy logic has been around for a long time (formal study of, that is).

Who said the base has to be integer?

Logarithms come to mind.

 

[meBigGuy]

Trying to in any way transition from circuit characteristics to consciousness is an exercise in futility. Even defining consciousness is tough.
But, analog modeling of neurons is possible, as in the book below.

Carver Mead (of Mead/Conway fame) wrote a book on neural networks that are based on analog neurons based on sub-threshold CMOS transistors. It discusses spice modeling, etc. https://www.amazon.com/dp/0201059924/?tag=pfamazon01-20 He was working to build an eye. It is highly recommended for anyone interested in the possibility of using analog computing for modeling neurons and networks

On a much higher behavioral level, I found the following charlie rose brain series episode on state of the art neurological science regarding consciousness to be fascinating.
http://www.charlierose.com/watch/60011362

I have a lot of opinions and personal theories, but nothing scientifically suitable for posting here.

 

[sophiecentaur]

Trying to in any way transition from circuit characteristics to consciousness is an exercise in futility.

Absolutely. Even associating the electronic technology in a processor with the way a computer algorithm runs is pretty fruitless. The same algorithm can run with virtually any type of processor.
OTOH, in an arm waving sense, the consciousness and the circuitry of a brain would have evolved together and you could argue that the structure and the 'intellectual' processes in the brain are probably more inter linked. Computers and Programs were 'designed' from totally different directions and their characteristics would have originated from the conscious 'aims' of Engineers and Programmers. (Intelligent design? haha - not all of my circuits).
Interesting stuff, but as you say, not for PF postings.

 

[fredreload]

It is true that software runs on computer circuits, but the same software can run on many different kinds of computers. That leads one to say that the software's property does not arise from the particular circuits.

Actually that's a good point, we wouldn't know if there is a software running in our mind. We'll need some computer expert here.

 

[anorlunda]

Actually that's a good point, we wouldn't know if there is a software running in our mind. We'll need some computer expert here.

I think you missed the point. PF rules don't allow us to speculate in these public forums. We are supposed to discuss only established science, and in this case science doesn't know how consciousness works.

 

[sophiecentaur]

Many, if not most, digital computers are what are (used to be?) termed General Purpose Computers. Your basic processor, I/O and memory can be programmed to do a variety of tasks / programs. In modern machines there are also specialist circuits that will just do one function ( video, audio, network). And, once built, they remain the same. Animal nervous systems ( the whole thing) tend to build themselves to suit the problems they are presented with and, it now turns out, can adapt even in later life. (For which I am heartily grateful!) It is hard to distinguish between the processor and the process. We are rebuilding and self reprogramming at the same time.
When you get down to it, digital computers and brains have surprisingly little in common, except for the final result - sometimes. Bottom up and top down.