IBM scientists create most comprehensive map of the brain’s network
July 28, 2010 by Amara D. Angelica
"The Mandala of the Mind": The long-distance network of the Macaque monkey brain, spanning the cortex, thalamus, and basal ganglia, showing 6,602 long-distance connections between 383 brain regions. (PNAS)
The Proceedings of the National Academy of Sciences (PNAS) published Tuesday a landmark paper entitled “Network architecture of the long-distance pathways in the macaque brain” (an open-access paper) by Dharmendra S. Modha (IBM Almaden) and Raghavendra Singh (IBM Research-India) with major implications for reverse-engineering the brain and developing a network of cognitive-computing chips.
“We have successfully uncovered and mapped the most comprehensive long-distance network of the Macaque monkey brain, which is essential for understanding the brain’s behavior, complexity, dynamics and computation,” Dr. Modha says. “We can now gain unprecedented insight into how information travels and is processed across the brain.
“We have collated a comprehensive, consistent, concise, coherent, and colossal network spanning the entire brain and grounded in anatomical tracing studies that is a stepping stone to both fundamental and applied research in neuroscience and cognitive computing.”
The scientists focused on the long-distance network of 383 brain regions and 6,602 long-distance brain connections that travel through the brain’s white matter, which are like the “interstate highways” between far-flung brain regions, he explained, while short-distance gray matter connections (based on neurons) constitute “local roads” within a brain region and its sub-structures.
Their research builds upon a publicly available database called Collation of Connectivity data on the Macaque brain (CoCoMac), which compiles anatomical tracing data from over 400 scientific reports from neuroanatomists published over the last half-century.
“We studied four times the number of brain regions and have compiled nearly three times the number of connections when compared to the largest previous endeavor,” he pointed out. “Our data may open up entirely new ways of analyzing, understanding, and, eventually, imitating the network architecture of the brain, which according to Marian C. Diamond and Arnold B. Scheibel is “the most complex mass of protoplasm on earth—perhaps even in our galaxy.”
The center of higher cognition and consciousness?
Core subnetwork (PNAS)
The brain network they found contains a “tightly integrated core that might be at the heart of higher cognition and even consciousness … and may be a key to the age-old question of how the mind arises from the brain.” The core spans parts of premotor cortex, prefrontal cortex, temporal lobe, parietal lobe, thalamus, basal ganglia, cingulate cortex, insula, and visual cortex.
Prefrontal cortex: integrator-distributor of information
By ranking brain regions (similar to how search engines rank web pages), they found evidence that the prefrontal cortex, while physically located in the front of the brain, is a functionally central part of the brain that might act as an integrator and distributor of information. Think of it as a switchboard.
As they stated in the PNAS paper, “The network opens the door to the application of large-scale network-theoretic analysis that has been so successful in understanding the Internet, metabolic networks, protein interaction networks, various social networks, and in searching the world-wide web. The network will be an indispensable foundation for clinical, systems, cognitive, and computational neurosciences as well as cognitive computing.”
The findings will also help them design the routing architecture for a network of cognitive computing chips, they suggest.
The research was sponsored by the Defense Advanced Research Projects Agency, Defense Sciences Office, Program: Systems of Neuromorphic Adaptive Plastic Scalable Electronics.
Dr. Modha presented the exciting findings of this study in a talk I attended at the Toward A Science Of Consciousness conference in Tucson in April, but he asked us to hold off on covering this until the formal paper appeared in a peer-reviewed journal.
A detailed Powerpoint slide show with voice narration (60 slides, ~52 minutes, ~50 MB) is downloadable here.
Comments (9)
by yack
What matters here is that we are able to map the brain and learn more about its inner workings. This will help to create more efficient artificial intelligences that can resolve the problems that will open the door to immortality.
Best regards.
by Leah
Mmmm…sexy.
by Alex
That certainly was a lot of work both from scientists and monkeys :)
It even looks very interesting and promissing.
But I would like to mention that the 1e person view can only be studied bij the first person itself, not from a third.
This means that you cannot study consciousness or awareness from the outside.
But of course you can analyse and study the Brain.
by Alex
Note :
“1e” means 1st and “bij” means by
:)
Sorry I am native Dutch speaking.
by brainmaps
I have heard from a few people that this paper is significant for cognitive computing. So let me take a moment to correct this misunderstanding.
For cognitive computing, you will need to know precisely how the brain is wired together, down to the single neuron level. I know a few people, like Henry Markram, believe you can forward engineer the brain, but I don’t agree with him on this, and I don’t want to go into the reasons. Let’s just accept that you need to know precisely how the brain is wired together, before you can even think about simulating brain networks and cognitive computing.
Modha is using a public database, called CoCoMac, of monkey (macaque) connectivity that has been around for over a decade. It consists primarily of cortico-cortical connectivity, but also recently added thalamic and striatal connections. CoCoMac is a very coarse resolution database of connectivity, and it is very incomplete. This is a big problem for Bodha, because if the connectivity information that he is relying on is flawed, then his study and any analysis he does on the data is flawed. I am stating, unequivocably, that the connectivity data in CoCoMac is deeply flawed, on many different levels, which I can discuss further if need be. But let’s move on.
It follows that any analysis Bodha does using this connectivity is flawed. That is my primary objection to the paper.
My second objection is that I don’t see any contribution that he makes in the paper, to either cognitive computing, or anything else for that matter. But amazingly, Bodha makes the following statement in his blog:
1) “We have successfully uncovered and mapped the most comprehensive long-distance network of the Macaque monkey brain”
and in the discussion section of the paper, he states:
2) “We have collated a comprehensive, consistent, concise, coherent, and colossal network spanning the entire brain”
Both of these statements are absurd and blatantly false. In 1), he did not map anything. He used a literature collation (CoCoMac) that someone else came up with, based on other people’s mappings. And in 2), he is stating that he, himself, did the collation, which is false. Rolf Kotter, the developer behind CoCoMac, did the collation, not Bodha. And 2) is also wrong because the network does not span the entire brain. Not even close.
Bodha is lucky I wasn’t one of the reviewers for this paper. I’m really amazed that some of the nonsense statements he made in the paper got past peer review. The only novel contribution you can argue for, that this paper makes, is the maximum entropy exponential distribution of the connectivity, but even this loses its significance in the light of my primary objection above.
As far as the relation of this paper to cognitive computing, there is none I can see, in light of the points above. IF the analysis of degree distributions in the network was based on reliable and meaningful connectivity data, then that would have been a legitimate contribution. However, that still would not be significant because anyone studying biological networks knows that degree distributions, in and of themselves, tell you very little about the network. And analyzing degree distributions is the primary analysis of the paper.
Note the last sentence in the paper:
“This hints at an evolutionarily preserved core circuit of the brain that may be a key to the age-old question of how the mind arises from the brain.”
Truly idiotic.
by brainmaps
typo. I mean Modha, not Bodha, in my post above.
by Be Afraid
Knaak,
In what way do the recent advances in Photosynthesis show that quantum effects contribute to “Consciousness” (I am aware of the quantum-walk nature of electron paths to which you are referring in Photosynthesis, so no need to repeat it all)?
I have no doubt that there are quantum processes operating within the brain. I can even see that neural pathways might be selected in the same way that the path of an electron is selected through a channel in photosynthesis.
But, I don’t see how this process would mean that quantum processes were necessary for the formation of a consciousness (how I hate that word too). It just means that without some form of quantum process, a different method of passing data between brain regions (or their simulations) would be needed…
Of course, I am incredibly ignorant of most things to do with Hameroff and Penrose. I got to see Hameroff speak at one of the “Beyond Belief” conferences a few years ago at UCSD, and I was not impressed with the answers he gave during the Q&A (Especially the questions from Sam Harris, a real neuroscientist, and not an anesthesiologist – sorry, but I don’t think his “qualifications” give him the knowledge he claims, and in that talk, he certainly failed to live up to the claims of knowledge)… But, then… ignorant… me, etc…
MB
by Knaak
I’m interested if it will be necessary to understand the function and effect of brainwaves, as well as the ability to simulate them, before we can possibly begin to understand the ability of the brain to work with patterns. A hard map of connections is a start, but then, if I’m not mistaken, each neuron is a bit like a radio tower as well, sending and receiving wave information, even as it exchanges molecules. Not only this, the conjectures of Hameroff & Penrose may be discounted by the scientific community at large, but I’m not convinced that micro-tubules do not play a larger role in the processes of consciousness. Recent research into photosynthesis already has begun to erode some of the criticisms of the work, and I expect further developments will continue to show a higher level of complexity in the brain/mind system than convention accepts at this time.
by Kerry Szymanski
Very exciting development and a major step towards strong AI.