In Liu Sixin’s famous science fiction series The three-body problem, human beings make their first contact with the Trisolarans, an alien civilization incapable of deception or lying because they broadcast their thoughts. This ability creates a deep distrust of human beings on the part of the Trisolarans, as our thoughts and plans are private from our own subjective reality.
Human beings have no foolproof way of knowing if someone is telling the truth. Sometimes our intuition can give us clues; strange body language, inconsistencies in what people are telling us or their behavior. But beyond that, we are largely left to take the word of our peers as we have no way of accessing their inner world.
That doesn’t mean we haven’t tried. People haveand we continue to develop techniques that can basically detect when someone is lying.
How did lie detection evolve?
One of the first known methods for lie detection comes from 1000 BC China. A person suspected of lying had to stuff his mouth with a handful of dry rice. After a while they had to spit out the rice. If the rice remains dry, the suspect is considered guilty of the charges against him.
Read more: The science of spotting a liar
The method is based on the physiological assumption that the experience of fear and anxiety is accompanied by reduced salivation and dry mouth.
More modern attempts to detect deception similarly rely on physiological signs, or “biomarkers,” as they are commonly known, that accompany lying.
One example is the common polygraph or lie detector test, which has been used in criminal cases for the past several decades, often providing evidence of deception to the prosecution. It works by picking up changes in blood pressure, heart rate, breathing and skin conductance that may be indicative that someone is not telling the truth.
The truth of detecting a lie
However, there is considerable uncertainty about the efficacy of polygraphs such as these physiological changes can also be induced by conditions other than lying. Furthermore, the supposed physiological cues that accompany lying may not be conserved across cultures, ethnicities, genders, and ages. Interpretation of lie detection results may be based on previous findings from one particular group that may not transfer to another.
“Addiction is certainly a major problem. To interpret the results of a lie detector test, the collected data is compared with other results. If these other results were conducted on people of a different race, ethnicity, or gender, the analysis could indeed be flawed, and different demographic groups could experience an increase in false positives,” says Jo Ann Oravec of the University of Wisconsin.
Oravec recently published a report which raises ethical and functional concerns regarding modern lie detection techniques. She has found problems with the “biomarker” approach to lie detection.
“False positives, which categorize people as liars, often have devastating economic and social consequences. Individuals identified as liars are not “innocent until proven guilty” but the opposite […] they have to provide support for their innocence, possibly in the form of yet another poorly validated lie detector test,” adds Oravec.
Can we use AI for lie detection?
More recently, researchers have been harnessing the potential of AI to detect lies. In 2021scientists at Tel Aviv University have developed a new AI-powered lie detector that measures people’s facial micro-expressions to determine the authenticity of their statements.
According to the researchers, the technology provides a 73 percent success rate compared to humans, who can spot a lie roughly 54 to 60 percent of the time. However, significant concerns questions have been raised as to whether such technologies may infringe upon human rights to privacy, fairness and discovery bias.
“AI-enhanced systems that collect data from a distance and that are rooted in complex constructs (such as ‘deception biomarkers’) may produce results that have a less transparent and obvious connection to the entity they are associated with,” says Oravek.
For the past few decades, neuroscientists have been working on the project of decoding mental content from brain-based imaging technologies. By combining fMRI data and machine learning, scientists have been able to infer with high accuracy semantic and visual mental content, a form of “mind reading”. This has led some to suggest that such techniques could be applied to lie detection.
However, in a 2019 paper, researchers at the University of Plymouth have shown that people using mental countermeasures can cheat fMRI-based lie detector tests. This has led some to question the legitimacy of implementing such technology for use in criminal cases.
Certain sectors of society, such as crime, business and immigration, have good reasons for wanting to use such lie detection technologies. However, there are several ethical and practical reasons to question whether or not such technology should be used.
“In the US and some other nations, legislation has restricted the use of lie detection processes in many settings. However, what constitutes ‘lie detection’ is evolving with these emerging AI-enhanced systems,” warns Oravec.
Read more: Lies, fraud and the criminal justice system