Japanese Scientists Build Artificial Brain Cerebellum In The Laboratory

JohnThomas Didymus

Japanese researchers say they have used human embryonic stem cells to build a three-dimension structure similar to the part of the brain called the cerebellum.

In higher animal brains, the cerebellum is responsible for coordinating motor functions and processing sensory information.

The team, led by Keiko Muguruma, at the RIKEN Center for Developmental Biology, announced the breakthrough in a paper published in the journal Cell, titled "Self-Organization of Polarized Cerebellar Tissue in 3D Culture of Human Pluripotent Stem Cells."

According to Muguruma, the cerebellum structure they built in the laboratory survived intact for only for a short time.

"The cultured cell aggregates (called spheroids) stop to develop at certain time point and start to degenerate. There may be some missing things for complete development."

But despite the fact that the brain built in the laboratory was transient and disintegrated soon after it was assembled, researchers in the field are unanimous that the achievement represents a significant step in the ongoing research efforts to construct an "artificial brain."

Discovery reports that several teams of researchers around the world are currently working to construct an "artificial brain," variously from biological tissue, electronic circuits, and computer programs. They hope to be able to construct a system that can perform much of the tasks humans are capable of performing with ease.

One of the foremost ongoing projects in this field is the European Union's Human Brain Project, involving about 183 top researchers from more than 20 countries. The goal of the $1.2 billion project is to map the human brain as a prerequisite to building systems that are able to simulate brain function.

Researchers involved in the project worldwide hope that insight gained into the way the brain functions would help medical researchers to better understand the mechanisms of brain diseases and thus enhance their ability to develop effective therapies.

Ongoing efforts, albeit on a smaller scale, involve a small start-up in La Jolla, San Diego, California, called Brain Corporation. The company is developing a computer program that will allow robotic systems to be trained to perform routine domestic tasks.

According to Eugene Izhikevich, founder and CEO of the company, the computer program being developed simulates the learning capabilities of mammalian brains, such as rats and dogs, rather than human brains.

Everyone knows how highly trainable dogs are. Izhikevich believes that the ability to develop a program that can match the learning capabilities of a dog, for instance, will be a major achievement in software engineering.

"If we can build a brain for robots and steal as much as we can from biology, [then you] can train your robot by showing examples of desired behaviors."

Already, the company has developed a robotic pooch than can learn to obey simple commands given by hand gestures.

Brain Corporation says it is making its software available to robot trainers. According to Izhikevich, the software will allow trainers to teach robots how to perform repetitive domestic operations such as those involved in using a dishwasher or operating a washing machine. Although he argues that the program does not threaten people at their workplace -- it's designed only to perform domestic chores -- it is easy to see how a breakthrough in teaching a robot to perform routine domestic chores can be applied to routine tasks in the workplace.

"It's like training a dog. It requires certain skill, but you don't need a PhD in robotics. You can spend a day, a week or a month."

Izhikevich said that the software only allows a robot to learn to perform tasks that it has been trained to do and no more. It does not equip robots with the ability to initiate complex operations and decision-making processes on their own.

However, when properly trained to perform specific tasks, a robot could become a fully autonomous system.

Other research teams are trying to develop robots that can perform more complex functions involving complex decision-making and exercise of original insight, such as humans do when they process and use language, categorize things, create and play music, guess what other humans are thinking, or what other humans want with the help of subtle cues such as facial expressions.

Although, these are cognitive capabilities we take for granted because we, as humans, are able to perform them easily, researchers attempting to simulate these functions in machine systems are being confronted with their largely unrealized complexity and sophistication.

Discovery reports that an IBM team is developing a computer chip called TrueNorth, an artificial system that mimics human brain pattern recognition capabilities by replicating the dense interconnections of the brain neural network using transistors.

But some teams of biologists are approaching the same problem that machine and software engineers are tackling by looking at ways to actually build brains using biological tissues.

According to Discovery, a team at Tufts University led by Robert Caplan is working to build three-dimensional tissue scaffolding of functional mammalian brain cells that can process electrical stimuli.

The ultimate goal, according to Caplan, is to build brain structures that are complex enough for scientists to use to test new types of drugs for treating brain diseases, such as Parkinson's and Alzheimer's.

"Ideally we would like to have a laboratory brain system that recapitulates the most devastating diseases. We want to be able to take our existing toolkit of drugs and understand how they work instead of using trial and error."

The Japanese team is also working to a similar goal. But efforts are facing a major obstacle. The brain structures being developed in the laboratory are unable to maintain stability for extended periods of time.

Researchers admit they still are far from developing stable brain structures in the laboratory and even farther from developing structures that can replicate the cognitive functions of the human brain.

[Images: Healthline; Wikimedia Commons]