A Computer Predicts Your Thoughts, Creating Images Based on Them

-By Monika Rajput

Summary: Scientists and researchers have come up with a very new and surprising technique that enables a computer to monitor the brain signals of humans and then model perception visually. In laymen’s terms, a computer can now try to imagine and guess what thought goes on in a human’s mind. Because of this intriguing ability of imagination, their computer has been able to put together new and unique information. This information has been in the form of various images, all of which are fictional and are something entirely new and different. The researchers have based this innovative technique of theirs on a new and unique brain-computer interface.

Full Story:

Scientists and researchers have come up with a very new and surprising technique that enables a computer to monitor the brain signals of humans and then model perception visually. In laymen’s terms, a computer can now try to imagine and guess what thought goes on in a human’s mind. Because of this intriguing ability of imagination, their computer has been able to put together new and unique information. This information has been in the form of various images, all of which are fictional and are something entirely new and different.

The researchers have based this innovative technique of theirs on a new and unique brain-computer interface. Prior to this brain-computer interfaces similar to this one have gotten successful in performing communication but only one way- from a brain to a computer. This new technique is the first one of its kind where methods of artificial intelligence were used to simultaneously model both, the presentation of the matter and new information by the computer as well as the signals of the human brain. This study and technique were first published in September 2020 in the Scientific Reports journal, which is an online open-access journal.

Unconscious thoughts and attitudes may be revealed:

This technique has a number of varied uses other than the generation of fictional and new images of human faces. This study gives way to more research in the same field that can eventually lead to the creativity of humans being increased with the help of a computer. Furthermore, the developers have also stated that with this technique we may even be able to gain a deeper understanding of how perception and its process work in a human brain.

Michiel Spape, the senior researcher of the project has stated, “The technique does not recognize thoughts but rather responds to the associations we have with mental categories. Thus, while we are not able to find out the identity of a specific ‘old person’ a participant was thinking of, we may gain an understanding of what they associate with old age. We, therefore, believe it may provide a new way of gaining insight into social, cognitive and emotional processes.” Spape also thinks that this technique will also prove interesting in the field of psychological advancement. He gives an example of how a person may perceive an elderly man differently than the other. They can both have two different images and ideas of that elderly man. He further says that they are still researching and looking for clues or proof that this technique is able to reveal such unconscious thoughts and perceptions.

Neuroadaptive generative modelling:

This technique has been called Neuroadaptive generative modelling by it researchers. The researchers conducted a study to evaluate and determine how effective this technique was and thirty-one voluntary participants took part in it. They were all given more than a hundred images that had been generated by artificial intelligence and the researchers then recorded their EEG.

The volunteers were made to specifically pay attention to a certain feature such as a smiling face but of an elderly person. They had to see all these images rapidly and simultaneously their EEG was recorded and sent to a neural network which then recognized and stated if any of the generated pictures was similar to any one of the pictures shown or not.

An estimation of the faces and features that the people had in their minds is then made by the neural network through this information. The final images produced by the computer were then given to the volunteers to analyze and almost every single one of them was a perfect match for the features they had in their minds. 83% accuracy was achieved by the experiment.

Tuukka Ruotsalo, Academy of Finland Research Fellow at the University of Helsinki, Finland and Associate Professor at the University of Copenhagen, Denmark said, “The technique combines natural human responses with the computer’s ability to create new information. In the experiment, the participants were only asked to look at the computer-generated images. The computer, in turn, modelled the images displayed and the human reaction toward the images by using human brain responses. From this, the computer can create an entirely new image that matches the user’s intention.”

Journal reference:

Lauri Kangassalo, Michiel Spapé, Tuukka Ruotsalo. Neuroadaptive modelling for generating images matching perceptual categoriesScientific Reports, 2020; 10 (1) DOI: 10.1038/s41598-020-71287-1

TechThoroughFare

Share on facebook
Share on twitter
Share on linkedin
Archives

Leave a Reply

Your email address will not be published. Required fields are marked *

Science

Latest

Trending