In a world where we are bombarded by unwanted messages, annoying calls and advertisements, privacy has become a very important issue. Picture: Joshua Woroniecki/Pixabay
In a world where we are bombarded by unwanted messages, annoying calls and advertisements, privacy has become a very important issue. Picture: Joshua Woroniecki/Pixabay

Tech News: Computers can now read our inner emotions

By Opinion Time of article published Feb 22, 2021

Share this article:

IN A world where we are bombarded by unwanted messages, annoying calls and advertisements, privacy has become a very important issue. The recent WhatsApp controversy showed how important privacy really is to users.

Users were dismayed with the latest WhatsApp update of its privacy policy and its intention to share private data with Facebook. Users were no longer able to opt out of sharing their personal data such as phone numbers, contacts, pictures and business communication content with parent company Facebook. This resulted in a mass exodus of millions of users to rival platforms such as Signal and Telegram.

Users want to be in control of their personal data and the use thereof. Luckily our deepest emotions and thoughts are still private, although spouses often believe that they know what the other is feeling and thinking. But our emotions and thoughts may not be private for long, since innovative breakthroughs in artificial intelligence (AI) and neuroscience may soon change this “last” space of privacy.

Detection of inner emotions

Researchers from the Queen Mary University of London on 3 February 2021 published novel research in PLoS ONE 16(2) under the title: “Deep learning framework for subject-independent emotion detection using wireless signals.” In the article they describe the application of wireless detection and a deep neural network (DNN) to determine a person’s inner emotions even in the absence of any other visual cues such as facial expressions.

Harmless radio signals were sent towards the subjects while they were watching a short video clip of three to four minutes and were then measured when the radio frequency (RF) reflected off the bodies of the subjects. After noise filtering, this made the measuring and analysis possible of minute variations in the person’s body movements in reaction to emotion evoking stimuli (in this case videos).

Based on this “hidden” information, the AI algorithm could determine basic emotion types evoked by the video, such as scary, relax, joy, and disgust. RF signals are thus able to describe underlying emotions of a person with the same accuracy as an electrocardiogram or ECG (recording the electrical activity of the heart) with the added benefits of being wireless and more practical.

Application of the technology

The application of this breakthrough is wide and in addition to psychological and neuroscientific studies of human behaviour could contribute to the management of the overall health and emotional wellbeing of people, as well as detect depression. As robots become a standard feature of the modern workplace, wireless emotion detection could be an important tool in understanding and managing the human/robot interaction.

The researchers are even exploring if low-cost devices, such as Wi-Fi routers or smart building sensors could detect emotions when a large group of people are gathered in an office or other environment. Except for detecting the overall state of mind of employees, human resource practitioners would also be able to determine the reception of new decisions and policies, regardless of what the users may express verbally. Even the police would be able to use this technology during crowd control to determine important emotional changes, which may lead to violence.

Until now, emotion detection has relied on the assessment of visible and audible signals such as speech, facial expressions, body gestures, or eye movements. Unfortunately, these methods can often be unreliable in displaying the true internal emotions of people. This is where this non-invasive method of detecting non-visible signals of emotions becomes valuable.

The future

State-of-the-art and low-cost emotion detection through a novel deep-learning neural network and wireless system is now possible without the use of invasive, bulky or restrictive physiological measurement systems such as an ECG and electroencephalogram or EEG (measuring the electrical activity of the brain).

However, depending on the particular use, emotional detection will certainly be met with fierce opposition from certain groups. Ethical, social and privacy concerns will have to be addressed before the public will accept this privacy invading technology.

But in future we will certainly see much more of the integration of human and machine so that the computer, AI, and the Internet will almost seamlessly weave into a “second self” for many people.

Professor Louis C H Fourie is a technology strategist

*The views expressed here are not necessarily those of IOL or of title sites

BUSINESS REPORT

Share this article:

Related Articles