SPAUN Brain Model Created Inside A Computer
Michael Harper for redOrbit.com – Your Universe Online
[ Watch the Video: Spaun Brain Activity Only for Sample Tasks ]
No matter how disturbing or even frightening as the thought may be, many scientists are working to unlock the mysteries of the human brain, create a model, and then place this model in a robot.
Chris Eliasmith, a professor from Canada’s University of Waterloo, has been working to complete his own human model to think and perceive just as a human would. However, as explained in his recent paper, “Spaun: A Perception-Cognition-Action Model Using Spiking Neurons,” Eliasmith’s model is able to perceive and then act. Called Spaun, (for Semantic Pointer Architecture Unified Network) this model can take in information, such as numbers and shapes, remember the information, and then move an arm to draw out the numbers and shapes it’s seen.
Currently, Spaun only lives inside a simulated environment deep inside a computer, but Eliasmith and team hope this model will be a huge step for understanding how the brain reacts to different drugs and treatments as well as how the brain ages and learns.
The average human brain is comprised of nearly 100 billion neurons, working together to perform the tasks of our everyday life.
“Human cognition isn’t interesting because we can recognize visual patterns … move our arms in an integrated way … [or] solve a particular task or puzzle. It’s interesting because we can do all of this with the same brain, in any order, and at any time,” explained Eliasmith in a July press statement from the University of Waterloo.
While we may not make use of all 100 billion neurons, we’re still able to call upon them when needed.
Eliasmith’s model only simulates around 2.5 million of these neurons, meaning it’s not yet capable of long division. The proof of concept, however, is a promising step forward.
To test Spaun, Eliasmith and team show the simulated model a series of images in 8 different tasks. Spaun then has to decide which actions to take after viewing these images, such as copy the numbers and shapes or recall a number in a list. Spaun is even capable of recreating whole number lists that it’s already seen.
Spaun isn’t able to react quickly in these simple tests, however. It takes a great amount of computational muscle to perceive and react accordingly. As such, it takes Spaun 2 hours to think and process the information it’s been given before acting on it.
Spaun’s 2.5 million neurons do not make it the biggest model brain around, but according to Eliasmith, it is the first to complete tasks and exhibit behaviors accordingly.
Though Spaun was created to mimic the human brain, Eliasmith has said he was most surprised at just how closely it resembled a human mind. For example, Spaun sometimes has difficulty when recalling a particularly long list of numbers. Furthermore, Spaun is more likely to remember numbers at the beginning and ending of lists, just as humans do.
“We can show that the performance on (an) intelligence task mimics the kind of performance that you see in people; it gets worse in the same sort of proportion that you find in humans,” said Eliasmith, speaking to CNN.
Spaun’s spartan computer existence works to its benefit for now. As a simulation, it can be used to study the effects of different drugs as well as the effects of Alzheimer’s disease and strokes.
“There are not only deep philosophical questions you can approach using this work — such as how the mind represents the world – but there are also very practical questions you can address about the diseased brain,” said Eliasmith.
“I believe that critical innovations are going to come from basic research like this. I can’t predict what specific industry or company is going to use this work or how — but I can list a lot that might.”
redOrbit.com
offers Science, Space, Technology, Health news, videos, images and
reference information. For the latest science news, space news,
technology news, health news visit redOrbit.com frequently. Learn
something new every day.\”
2012-11-30 16:00:23
Source:
