To understand linguistic ability of a human mind, a computer has been trained to understand how words and mind images of words are linked. Researchers of Machine Learning Department of Carnegie Mellon University at Pittsbrugh have recently published their findings in journal Science. The computer can ‘read’ people’s mind by looking at their brain scans as they imagined or thought about specific words. The main aim of the research is to understand how and where brain stores information. This study might also help in treatment of language disorder and learning disabilities in people.
Brain, knowledge organization and linguistic ability have challenged researchers for centuries. But only now with great strides in medical and computing technology an interdisciplinary team work is hoping to crack the brain puzzle. Using functional magnetic resonance imaging, a brain scan of real time activity of brain is generated when nine students think of 58 different words. They further calibrated the computer with these brain activity scans. Not all students thought of same images for specific words, so a kind of averaging was needed. These averaged images of brain scans for specific words were fed in to computer. Now the computer was asked to choose a brain image when a person thought a certain word. And it worked. The study is further being extended to study brain scan images of certain phrases. Thus the complete language image mapping of human mind and “reading” a mind by computer is the ultimate goal. But the question remains are we so predictable and uniform in thought?
There are limitations of using machines to understand human brain. First of all, real time activity brain scan is a difficult part. A patient has to be ‘hooked’ up thinking to a scan machine for several minutes without mind straying over to other thoughts. Also, human brain has huge imaging ability which is impossible to be mapped. People are not identical in thought. ‘Reading’ and ‘predicting’ human mind will impossible task for any computational machine. Such experiments can never be done in controlled environment and repeated. And science always doubts experiments which cannot be repeated. But on the science has always seen ambitious experiments and efforts being perfected over decades. Nobody believed Galileo and his telescopic observations but now we have ‘Phoenix’ on the Mars. So, maybe years ahead we’ll have a machine friend reading over minds perfectly and never hurting us!
Source: Yahoo
