Can Neuroscience Understand Donkey Kong, Let Alone a Brain?

  • Thread starter Sciborg_S_Patel
  • Start date
S

Sciborg_S_Patel

#1
Can Neuroscience Understand Donkey Kong, Let Alone a Brain?

"Two researchers applied common neuroscience techniques to a classic computer chip. Their results are a wake-up call for the whole field."

The human brain contains 86 billion neurons, underlies all of humanity’s scientific and artistic endeavours, and has been repeatedly described as the most complex object in the known universe. By contrast, the MOS 6502 microchip contains 3510 transistors, runs Space Invaders, and wouldn’t even be the most complex object in my pocket. We know very little about how the brain works, but we understand the chip completely.

So, Eric Jonas and Konrad Kording wondered, what would happen if they studied the chip in the style of neuroscientists? How would the approaches that are being used to study the complex squishy brain fare when used on a far simpler artificial processor? Could they re-discover everything we know about its transistors and logic gates, about how they process information and run simple video games? Forget attention, emotion, learning, memory, and creativity; using the techniques of neuroscience, could Jonas and Kording comprehend Donkey Kong?


No. They couldn’t. Not even close.

Even though the duo knew everything about the chip—the state of each transistor and the voltage along every wire—their inferences were trivial at best and seriously misleading at worst. “Most of my friends assumed that we’d pull out some insights about how the processor works,” says Jonas. “But what we extracted was so incredibly superficial. We saw that the processor has a clock and it sometimes reads and writes to memory. Awesome, but in the real world, this would be a millions-of-dollars data set.”

Last week, the duo uploaded their paper, titled “Could a neuroscientist understand a microprocessor?” after a classic from 2002. It reads like both a playful thought experiment (albeit one backed up with data) and a serious shot across the bow. And although it has yet to undergo formal peer review, other neuroscientists have already called it a “landmark paper”, a “watershed moment”, and “the paper we all had in our minds but didn't dare to write”. “While their findings will not necessarily be surprising for a chip designer, they are humbling for a neuroscientist,” wrote Steve Fleming from University College London on his blog. “This kind of soul-searching is exactly what we need to ensure neuroscience evolves in the right direction.”


 

Paul C. Anagnostopoulos

Nap, interrupted.
Member
#2
I'm trying to download the paper but having no luck so far.

If they had the complete transistor map, I'm not sure why they couldn't figure some things out. For example, with enough poking, couldn't they identify an adder? Perhaps only if they collected lots of data and then wrote an analyzer.

I wonder how long they spent on this?

Ah, got the paper.

~~ Paul
 
S

Sciborg_S_Patel

#3
This link works:

Could a neuroscientist understand a microprocessor?

"There is a popular belief in neuroscience that we are primarily data limited, that producing large, multimodal, and complex datasets will, enabled by data analysis algorithms, lead to fundamental insights into the way the brain processes information. Microprocessors are among those artificial information processing systems that are both complex and that we understand at all levels, from the overall logical flow, via logical gates, to the dynamics of transistors. Here we take a simulated classical microprocessor as a model organism, and use our ability to perform arbitrary experiments on it to see if popular data analysis methods from neuroscience can elucidate the way it processes information. We show that the approaches reveal interesting structure in the data but do not meaningfully describe the hierarchy of information processing in the processor. This suggests that current approaches in neuroscience may fall short of producing meaningful models of the brain."
 
#5
A quick read indicates that it's an interesting paper. Indeed, both a microprocessor and the brain: hard.

~~ Paul
And reading a bit more into it, Paul it's not looking very good for materialists like yourself who think that one day "we" are going to understand how the brain creates consciousness, the mind (which it doesn't IMO)
 
Last edited:

Paul C. Anagnostopoulos

Nap, interrupted.
Member
#6
And reading a bit more into it, Paul it's not looking very good for materialists like yourself who think that one day "we" are going to understand how the brain creates consciousness, the mind (which it doesn't of course)
Which parts of the paper make you say this? Especially since consciousness is not mentioned.

~~ Paul
 
#7
Which parts of the paper make you say this? Especially since consciousness is not mentioned.

~~ Paul
Understanding how the brain works to produce consciousness is the ultimate goal of neuroscience, isn't it ?

“let’s see how much closer it gets us to understanding the chip,” says Kording. “If it doesn’t work on the chip, how can we expect it to work on the brain?”

"“But the problem is much harder than I thought it would be. Big data alone isn’t going to save us.”
 
#8
Understanding how the brain works to produce consciousness is the ultimate goal of neuroscience, isn't it ?

“let’s see how much closer it gets us to understanding the chip,” says Kording. “If it doesn’t work on the chip, how can we expect it to work on the brain?”

"“But the problem is much harder than I thought it would be. Big data alone isn’t going to save us.”
Something to chew on.
Brain study suggests consciousness a matter of optimal degree of connectedness in neural network
http://medicalxpress.com/news/2016-01-brain-consciousness-optimal-degree-connectedness.html

Full study here > https://arxiv.org/ftp/arxiv/papers/1509/1509.04304.pdf
 
#9
Something to chew on.
Brain study suggests consciousness a matter of optimal degree of connectedness in neural network
http://medicalxpress.com/news/2016-01-brain-consciousness-optimal-degree-connectedness.html

Full study here > https://arxiv.org/ftp/arxiv/papers/1509/1509.04304.pdf
Basically, they conclude that brain states in chemically induced unconsciousness are different from that of wakeful states. And the "No s***, Sherlock" award goes to....

That study reminded me of this http://scigen.csail.mit.edu/scicach...Sushanandar.Jon+Hopkins.Thomas+Voluminic.html
 

Paul C. Anagnostopoulos

Nap, interrupted.
Member
#10
Understanding how the brain works to produce consciousness is the ultimate goal of neuroscience, isn't it ?

“let’s see how much closer it gets us to understanding the chip,” says Kording. “If it doesn’t work on the chip, how can we expect it to work on the brain?”

"“But the problem is much harder than I thought it would be. Big data alone isn’t going to save us.”
Of course the problem is difficult. But the small amount of progress they made on the chip in whatever small amount of time they spent on it does not prove that they could not make more progress given more time. On the other hand, it's perfectly reasonable and useful to reach the conclusion that neuroscientists need to try other approaches. Perhaps we have to simulate the brain in order to understand it better.

This is similar to progress on understanding what is going on in neural networks. They are an example of a man-made entity that we do not fully understand (as opposed to a computer).

http://yosinski.com/media/papers/Yo...ral_Networks_Through_Deep_Visualization__.pdf

https://research.googleblog.com/2015/06/inceptionism-going-deeper-into-neural.html

~~ Paul
 
#11
#12
Something to chew on.
Brain study suggests consciousness a matter of optimal degree of connectedness in neural network
http://medicalxpress.com/news/2016-01-brain-consciousness-optimal-degree-connectedness.html

Full study here > https://arxiv.org/ftp/arxiv/papers/1509/1509.04304.pdf
Thanks for that, Steve

I'm not sure how that helps your case though, it only strengthens the proponents case.
.........................................................
QUOTE

Scientists (and surgeons) believe that propofol causes people to become completely unconscious, which by definition would mean to become incapable of processing thoughts. The brain should not be able to process pain signals, for example, thus making surgery a pain free experience. To gain a better perspective on the various states of consciousness, the team watched blood flow changes in the brains of the volunteers as they moved from a conscious state, to unconsciousness and then back to consciousness.

In studying the scans, the researchers found that when the volunteers were conscious, there was what they describe as "a flurry of ever-changing activity," with a lot of activity between the various neural networks. In contrast, they found that while unconscious, the brains of the volunteers were engaged in far less interconnectivity and were less variable over time.

These findings, the team suggests, show that consciousness in the brain is merely, in a physical sense, a state where there is an optimal level of neural network connectedness.

..........................................................

Precisely, consciousness according to neuroscience needs a functioning brain (and not just a few neurons somewhere down near the brain stem) to create lucid thoughts with memory formation, like is reported during NDE's. That's why when the brain isn't functioning as in cardiac arrest there should be no experience and yet there is.
 
#13
Of course the problem is difficult. But the small amount of progress they made on the chip in whatever small amount of time they spent on it does not prove that they could not make more progress given more time. On the other hand, it's perfectly reasonable and useful to reach the conclusion that neuroscientists need to try other approaches. Perhaps we have to simulate the brain in order to understand it better.

This is similar to progress on understanding what is going on in neural networks. They are an example of a man-made entity that we do not fully understand (as opposed to a computer).

http://yosinski.com/media/papers/Yo...ral_Networks_Through_Deep_Visualization__.pdf

https://research.googleblog.com/2015/06/inceptionism-going-deeper-into-neural.html

~~ Paul
I don't mean to be awkward here but what exactly is an artificial neuron ? And how can studying anything artificial tall us anything about real neurons ?

" We train an artificial neural network by showing it millions of training examples and gradually adjusting the network parameters until it gives the classifications we want. The network typically consists of 10-30 stacked layers of >>>artificial neurons."
 

Paul C. Anagnostopoulos

Nap, interrupted.
Member
#14
I don't mean to be awkward here but what exactly is an artificial neuron ? And how can studying anything artificial tall us anything about real neurons ?
You could ask the same question about the weather or anything else we simulate. If we can build something that acts the same way as a natural entity, then that suggests we've made progress in understanding it. Especially if we are specifically trying to build something that matches the natural entity.

" We train an artificial neural network by showing it millions of training examples and gradually adjusting the network parameters until it gives the classifications we want. The network typically consists of 10-30 stacked layers of >>>artificial neurons."
Similar to what the brain does, with the environment acting as the feedback for adjusting the parameters. What we don't understand about artificial neural networks is how the representation of the knowledge is stored in the intermediate layers.

~~ Paul
 
#15
Lovely!

I have yet to read the paper, but I have mentioned my own unpublished analogy to fMFRI before. If computers were alien objects and we wanted to understand them, one way would be to stick little thermometers on each chip (possibly an array of thermometers on large chips) and load various pieces of software on to the system. The recordings of heat generation from each chip would generate a wealth of data - with slightly different results for each piece of software and each possible usage pattern, and it would provide zero understanding!

Just one thing that would bedevil any real understanding, would be that the memory chips would get used in an essentially arbitrary order (but probably the same order if the computer was booted from cold) because their memory would get mapped by the paging mechanism. In any case, the view the heat profile would generate would be way too coarse - as presumably fMRI is.

This isn't about materialism vs something else - Paul is right on this - but it is an excellent example of the way science grossly exaggerates what it has achieved in areas like brain science - thereby giving some people the impression that it can speak with authority.

Of course tools like fMRI are useful clinically, but determining that something is not working correctly is not the same as understanding how it actually does what it normally does.

David
 
Last edited:
#16
If we can build something that acts the same way as a natural entity,
We can't learn anything though because "acting" is not the same as "being." We don't really know what a neuron is or where it came from. To change the artificial neuron into a real neuron requires a magic trick and I assume you don't believe in magic ?
 

Paul C. Anagnostopoulos

Nap, interrupted.
Member
#17
We can't learn anything though because "acting" is not the same as "being." We don't really know what a neuron is or where it came from. To change the artificial neuron into a real neuron requires a magic trick and I assume you don't believe in magic ?
Of course we know what they are and we have excellent mathematical models of them. I'm not sure we need to know where it came from to figure it out, although investigating simpler nervous systems could certainly help. I'm not sure what changing an artificial neuron into a biological one has to do with anything.

Do you say the same thing about other kinds of cells?

~~ Paul
 
#18
Thanks for that, Steve

I'm not sure how that helps your case though, it only strengthens the proponents case.
.........................................................
QUOTE

Scientists (and surgeons) believe that propofol causes people to become completely unconscious, which by definition would mean to become incapable of processing thoughts. The brain should not be able to process pain signals, for example, thus making surgery a pain free experience. To gain a better perspective on the various states of consciousness, the team watched blood flow changes in the brains of the volunteers as they moved from a conscious state, to unconsciousness and then back to consciousness.

In studying the scans, the researchers found that when the volunteers were conscious, there was what they describe as "a flurry of ever-changing activity," with a lot of activity between the various neural networks. In contrast, they found that while unconscious, the brains of the volunteers were engaged in far less interconnectivity and were less variable over time.

These findings, the team suggests, show that consciousness in the brain is merely, in a physical sense, a state where there is an optimal level of neural network connectedness.

..........................................................

Precisely, consciousness according to neuroscience needs a functioning brain (and not just a few neurons somewhere down near the brain stem) to create lucid thoughts with memory formation, like is reported during NDE's. That's why when the brain isn't functioning as in cardiac arrest there should be no experience and yet there is.
Quick reminder. Remember that video of camels being slaughtered by having their throats cut. Do you recall where their blood was flowing? On the floor, not their brains, remember? And yet they were still conscious. That act of slaughter effectively simulates a cardiac arrest. More details thoughts later
 
#19
Quick reminder. Remember that video of camels being slaughtered by having their throats cut. Do you recall where their blood was flowing? On the floor, not their brains, remember? And yet they were still conscious. That act of slaughter effectively simulates a cardiac arrest. More details thoughts later
Why have you brought that horrible scenario up again, Steve It wasn't a camel it was a bull (poor creature) and what's that got to do with cardiac arrest ? The heart of the animal is still pumping even though it's throat has been cut so if it is still conscious then it must have been receiving some flow to it's brain.
 
Last edited:
#20
Of course we know what they are and we have excellent mathematical models of them. I'm not sure we need to know where it came from to figure it out, although investigating simpler nervous systems could certainly help. I'm not sure what changing an artificial neuron into a biological one has to do with anything.

Do you say the same thing about other kinds of cells?

~~ Paul
We only know that they exist, we don't know for sure what they are and we don't know how connecting them with a bit of electro-chemistry gives rise to consciousness. I'm not saying anything here which is controversial, it's just a fact Paul. If you know how they create consciousness and what consciousness is composed of, fire away (no pun intended)
 
Top