
Predicted fMRI images for celery and airplane show significant similarities with the observed images for each word. Red indicates areas of high activity, blue indicates low activity. Credit: Courtesy of Science
For many years, scientists have been trying to find ways to decipher human thoughts. It took several algorithms and neuro-scans to get to the bottom of how the brain really works.
In their most recent study, a computer scientist, Tom Mitchell, and a cognitive neuroscientist, Marcel Just, both from Carnegie Mellon University, used fMRI data to develop a sophisticated computational models.
These models were designed to predict the brain's response in relation to concrete nouns, or things that we experience through our senses.
The researchers created models for 60 concrete nouns which have been taken fMRI activation patterns. These models were also used to analyze text corpus, a set of text containing a trillion words noting any relationship of each noun to a set of 25 verbs associated with sensory or motor functions.
Combining fMRI data and analysis of the text corpus, the computer was able to predict the brain activity pattern of thousands of other concrete nouns.
Using this method, the researchers have determined that using their computational model is significantly better than chance. An important implication to understanding brain-related diseases and memory losses.
Click here to read more about Brain Imaging and Computational Modeling, Mind Reading using fMRI




















