A new computer tool can predict the intelligence of a person



[ad_1]

Be careful if you lie about your IQ because the researchers have come up with a machine learning algorithm that can tell you how intelligent you are by examining a scanner of your brain. The study showed that the new computational tool can predict the intelligence of a functional magnetic resonance imaging (fMRI) of its resting brain activity

The fMRI develops a map of brain activity by detecting changes in blood flow to specific brain regions.

"We discovered that if we let people lie down in the scanner and do nothing while we measure the pattern of activity in their brains, we can use the data to predict their intelligence," he said. co-author Ralph Adolphs. Institute of Technology (Caltech).

To train their algorithm on complex patterns of activity in the human brain, researchers downloaded brain scans and intelligence scores of nearly 900 individuals.

The researchers found that after data processing, the algorithm was able to predict intelligence at statistically significant levels across subjects.

"The information we derive from brain measurements can be used," says Julien Dubois of Cedars-Sinai Medical Center.

To predict the intelligence of brain scans, the algorithm does something that humans can not do, because even an experienced neuroscientist can not look at a brain scan and tell how intelligent a person is, according to the researchers.

The researchers also conducted a parallel study using the same population and the same test approach.The personality test that they used divides the personality into five scales – openness to experience, awareness, extroversion, amenity, neurosis.

However, he It's been much harder to predict personality using the team's method, used to predict intelligence.

Both studies still need to be published in the separate journals Personality Neuroscience and Philosophical Transacti ons of the Royal Society.

Source: IANS
Image Source: Shutterstock

Posted: June 30, 2018 5:30 PM | Updated: June 30, 2018 17:32


[ad_2]
Source link