Why did the Human Brain Project Crash and Burn? – Walter Bradley Center for Natural and Artificial Intelligence

The Human Brain Project from 2013 sounded like science fiction in an EU setting: We will build a brain in a decade: And, if we do succeed, we will send, in ten years, a hologram to talk to you.

Well, we all got one thing right. It was fiction. Filmmaker Noah Hutton, a sympathetic observer, chronicled the decline, producing a documentary, In Silico, that focuses on booster Henry Markram who, according to his TED talk bio from 2009, was director of Blue Brain, a supercomputing project that can model components of the mammalian brain to precise cellular detail and simulate their activity in 3D. Soon hell simulate a whole rat brain in real time.

When the project started to wobble, many neuroscientists were angry and disappointed:

But that year (2010), Hutton also started to encounter critics in the neuroscience community. They claimed that the simulation project was premature because too little was known about the different types of neuron in the brain and how they were wired. Anyone can repair a broken watch by putting its known components in the right places, neuroscientist Zachary Mainen at the Champalimaud Centre for the Unknown in Lisbon, Portugal, tells the camera. Try this with the incompletely understood components of the brain, he says, and youll end up with a bunch of parts that doesnt tell the time.

By 2014, about 750 neuroscientists had signed a letter pledging that they wouldnt participate and by 2016, Markram was no longer in charge.

And the whole brain maps? They never happened.

Reflecting on the premiere of the documentary, science writer Alison Abbott tells us,

Hutton hints that the disputes were driven by money. I disagree; my sense is that it came down to leadership style and irresolvable differences in scientific opinion. There is a bolder, even more interesting, story waiting to be told.

Abbott does not spell out the irresolvable differences in scientific opinion but we cant help wondering if they relate to the question of whether the brain can be understood in such a simplistic way.

Some thoughts from The Guardian in 2014:

Central to the latest controversy are recent changes made by Henry Markram, head of the Human Brain Project at the Swiss Federal Institute for Technology in Lausanne. The changes sidelined cognitive scientists who study high-level brain functions, such as thought and behaviour. Without them, the brain simulation will be built from the bottom up, drawing on more fundamental science, such as studies of individual neurons. The brain, the most complex object known, has some 86bn neurons and 100tn connections.

The main apparent goal of building the capacity to construct a larger-scale simulation of the human brain is radically premature, Peter Dayan, director of the computational neuroscience unit at UCL, told the Guardian.

Some thoughts from The Scientist in 2015:

After weathering serious criticism last year, the European Commission-backed effort to map the brains neural connections must reform or die, a review panel says.

Its been a rough nine months for the European Commissions Human Brain Project (HBP). More than 250 of Europes top neuroscientists threatened to boycott the $1.6 billion effort to create a computer simulation of the human brain last July, and now a European Commission (EC) review panel has echoed some of the same concerns voiced by those scientists.

Some thoughts from HBC, a portal dedicated to fast computers, in 2019:

From the outset, HBP was beset by criticism unrealistic goals, un-useful goals, poor organization, waste of scarce research resources said many. Others argued its big goals would lead to big insights as well as myriad useful tools. Its hard to gloss over the HBPs problems, but perhaps too easy to understate its contributions Whether the HBP was and is tilting at windmills is a significant question.

Tilting at windmills along with Don Quixote maybe? (See the illustration of the outcome of the Dons famous attempt to charge a windmill.) What if the brain is not only not a computer but not even like a computer?

As neuroscientist Yuri Danilov has pointed out, Right now people are saying, each synoptical connection is a microprocessor. So if its a microprocessor, you have 1012 neurons, each neuron has 105 synapses, so you have you can compute how many parallel processing units you have in the brain if each synapse is a microprocessor. But as soon as you assume that each neuron is a microprocessor, you assume that there is a programmer. There is no programmer in the brain; there are no algorithms in the brain.

The human brain exceeds the most powerful computers in efficiency. Its also not clear exactly how it works. Lemurs, with brains 1/200th the size of a chimpanzees brain, passed the same IQ test. And this is to say nothing of the little understood relationship between the human brain and the human mind.

Underlying the quarrels and stalemates of the Human Brain Project may be practical problems with the idea of simply simulating the brain on a computer.