The idea of “mind-reading” has commonly been associated with magic or the supernatural. However, research done at a lab in the University of Toronto Scarborough (U of T Scarborough) recently took a significant stride forward in achieving this possibility.
Dan Nemrodov, a postdoctoral fellow working in Assistant Professor Adrian Nestor’s lab at U of T Scarborough, is the pioneer behind a newly developed technique.
When human eyes detect an object, the brain automatically generates a mental image of that visual stimulation. Through the use of electroencephalography (EEG) data, Nemrodov was able to successfully translate this mental image, or percept, into a tangible and physical illustration of what is taking place inside the brain.
EEG is essentially a popular neurological test that picks up electrical activity in the brain by attaching small electrodes to the scalp. After the connection is established, the activity of brain cells in the form of electrical impulses is seen as wavy lines on an EEG recording.
To carry out the study, the test subjects’ brain activities were first measured through EEG. Afterwards, Nemrodov’s team transformed the data into concrete images with the help of various machine learning algorithms.
Although this is not the first time a similar feat has been accomplished, Nemrodov’s research has valuable underlying implications.
First, this is the first time that EEG has been used to reconstruct brain images. In the past, researchers had only generated similar images using an expensive neuroimaging technique known as functional magnetic resonance imaging (fMRI).
This technique detects brain activity via monitoring changes of blood flow in the brain, but EEG is a more portable and economical method of achieving the same means and hence has more practical applications. EEG imaging also has another advantage.
“fMRI captures activity at the time scale of seconds, but EEG captures activity at the millisecond scale,” Nestor said, according to ScienceDaily.
The researchers estimated that it takes a typical human brain only about 170 milliseconds (0.17 seconds) to process visual signals, in this case faces, and form a representative image of those signals. This makes EEG a valuable tool that can accurately record and adjust to rapid changes in brain activities.
Additionally, this new technique has social implications. Oftentimes, people with autistic or other psychotic diagnoses struggle to communicate or understand others’ minds.
However, brain reconstruction could potentially help people who experience such troubles in communication or verbal expression.
“Not only could it produce a neural-based reconstruction of what a person is perceiving, but also of what they remember and imagine, of what they want to express,” Nestor said.
Nestor also believes that this brain image reconstruction technique can have forensic and even law enforcement applications, such as generating more accurate facial images based on eyewitness information.
Such an advancement has practical utilities because it would likely decrease the need to focus extensively on eye witness’ verbal accounts, since people can sometimes provide incoherent or biased information immediately after the stressful nature of witnessing a crime scene.
The findings in Nestor’s lab are anticipated to be published in the journal eNeuro.
The research project has received generous funding up to date, partly by the Natural Sciences and Engineering Research Council of Canada (NSERC) as well as from a Connaught New Researcher Award.
“What’s really exciting is that we’re not reconstructing squares and triangles but actual images of a person’s face, and that involves a lot of fine-grained visual detail,” Nestor said.
