The human mind's ability to recognize objects is truly astonishing. The scientific community is keen to understand the entire process better, not only for its speed but also for its precision. This led a group of researchers at the University of Washington to conduct an experiment in which they managed to decode brain waves in real time, recognizing with a certainty greater than 90 percent what kind of object each participant was observing.

Decoding the Mind in Real Time
Brain

Every day we are more amazed by what certain algorithms can do in recognizing gestures, faces, and other elements. Twenty years ago we didn't even have decent speech recognition, but artificial intelligence and machine learning continue to crush limits. However, the human mind remains far superior, especially when we consider its potential. The so-called "neural networks" need to be trained, while we only need a few milliseconds to recognize and differentiate objects. This "speed of perception" is what leads us to a recent experiment carried out by the University of Washington.

Decoding the Mind in Real Time
The image shows the differences in signals according to each electrode and the observed material.

The participants, seven epilepsy patients who received electrodes in their temporal lobes for a week to determine the origin of their seizures (making this study a kind of "secondary task"), were limited to watching a series of images composed of houses and faces, at intervals of 400 milliseconds. The brain waves captured by the electrodes were processed by a complex algorithm that sampled and digitized the signals about a thousand times per second. By combining these data with the signal type and the location of the electrodes, the software achieved an accuracy of 96 percent when recognizing (within a margin of 20 milliseconds) what each participant was seeing: a house, a face, or a gray screen.

Although the algorithm involved received prior training, this technique could represent a huge step toward creating a "brain map" to identify which brain regions are most sensitive to certain types of data. Another future possibility to consider is a communication system that would allow completely paralyzed patients to get in touch with other people.

Official announcement: