Social Engineering Blogs

An Aggregator for Blogs About Social Engineering and Related Fields

The Latest Science about Behavioral Indicators of Deception

A Message for Professional Interviewers, Investigators, Therapists, and Negotiators

I wanted to let our Humintell community know of a recent paper that was published in a top-tier, scientific journal on behavioral indicators of deception.

In the scientific community, there have been debates about whether behavior can reliably differentiate truths from lies, and if so, which ones.

Fortunately, many studies published in the last two decades have demonstrated that nonverbal behavior (NVB) can differentiate truth-tellers and liars fairly well.

One reason for the emergence of the wealth of these findings is that these studies have examined situations in which people are actually interviewed about meaningful events and where there’s consequences for not being believed.

These are precisely the types of situations in which professionals – investigators, therapists, counselors, negotiators – work and need to make the best evaluations they can possibly make.

In the remainder of this article, I summarize three main take-aways of the latest paper, and interested readers can read the paper here.

1. Professionals should focus on clusters of NVB produced in multiple channels of behavior

Across the face, voice, hands, and whole body. Examining clusters makes more sense than examining only single behavior because NVB are part of a total communication package that occurs across multiple channels, with and without words.

Four facts support this suggestion:

  1. Human bodies are wired to connect our thoughts, feelings, and behavior (think embodied cognition)
  2. Our thoughts and feelings are blended at any one time and across time
  3. We verbalize only a portion of what’s in our heads at any one time
  4. Different mental states (cognitions, emotions, etc.) map onto different NVB channels (face, voice, gesture, etc.)

2. Professionals should focus on behavioral indicators that have been validated in science and vetted in the field.

Our Humintell community knows that certain NVB have been scientifically validated as deception indicators while others have not.

On one hand, facial expressions of emotion and microexpressions, some types of gestures, fidgeting (in some contexts), and some aspects of voice differentiate truth-tellers from liars.

On the other hand, looking away when answering questions (gaze aversion) has not been scientifically validated as a deception cue. Astute professionals will know not to believe everything about NVB that they may hear or read about.

3. NVB are also important indicators of many other mental states that can be helpful for interviewers.

Our Humintell community also knows that NVB can signal many different mental states, all of which can be useful to professionals as landmarks of meaningful topics and themes.

These include specific, discrete emotions such as anger, disgust, or fear; general affective states such as open or closed, relaxed or tense; specific verbal words or phrases; cognitive processes, confusion, concentration; and others.

Identifying these behaviors can give professionals additional insights to people’s mindsets.

Experienced professionals all develop their own customized interview strategies and tactics. The accurate and reliable observation and classification of NVB can be a crucial aid, and focusing on NVB clusters that have been validated in science and vetted in field work is key.

Equally important is to ignore NVB that have not been validated. By knowing which behaviors have been validated and vetted and which not, professionals can become more efficient by distinguishing meaningful signals from noise in the behavioral mess that occurs in interviews.

Here at Humintell, we base all our trainings on behavioral indicators of mental states that have been validated in science and vetted in the field by practitioners. Thus, please rest assured that we practice what we preach.

 

The post The Latest Science about Behavioral Indicators of Deception first appeared on Humintell.

Can AI Tell Your Politics By Looking At Your Face?

Believe it or not, the 2024 Presidential Election is right around the corner. And according to a new study out of Denmark, AI may be able to predict your political views.

A team of researchers based in Denmark and Sweden recently conducted a study to see if “deep learning techniques,” like facial recognition technology and predictive analytics can be used on faces to predict a person’s political views.

The study was entitled “Using deep learning to predict ideology from facial photographs: expressions, beauty, and extra-facial information” and published as an open access article in March 2023.

The Methodology

Tfigure 5he researchers used a public dataset of 3,233 images of Danish political candidates who ran for local office and cropped them to only show their faces (see example image to the left).

After that, they applied advanced techniques to assess their facial expressions and a facial beauty database to determine a person’s “beauty score.”

Using these data points, the scientists predicted whether the figures pictured were left-wing or right-wing.

According to Business Insider, “The study found that the tech accurately predicted the political affiliations 61% of time.

The model predicted that conservative candidates  “appeared happier than their left-wing counterparts” because of their smiles, whereas liberal candidates were more neutral.

Women who expressed contempt — a facial expression characterized by neutral eyes and one corner of the lips lifted — were linked to more liberal politics by the model.”

In addition, the researchers found that AI correlated the political candidate’s level of attractiveness with their politics.

Women deemed attractive by their beauty scores were predicted to have conservative views, though there was not a similar correlation between mens’ level of attractiveness and right-wing leanings.

The study’s writers say the results of this study, “confirmed the threat to privacy posed by deep learning approaches.”

Attractiveness and Political Ideology

Links between attractiveness and political ideology are nothing new.

One study entitled “Effect of physical attractiveness on political beliefs” examined the relationship between attractiveness and political beliefs.

As reported in the Guardian, “The researchers took data from the 1972, 1974 and 1976 American National Studies surveys which asked people to evaluate the appearance of others and also explored participants’ political beliefs, income, race, gender, and education.

These results were compared with the Wisconsin Longitudinal study which focused on the physical characteristics of more than 10,000 high school students who were rated by others on their level of attractiveness.”

The results of that study suggested that “more attractive individuals are more likely to report higher levels of political efficacy, identify as conservative, and identify as Republican.”

Facial Recognition Technology and Political Orientation

Facial recognitionSimilar research suggests that facial recognition technology can predict a person’s political orientation with 72% accuracy.

Published in Scientific Reports one study suggests that facial recognition technology can accurately predict someone’s political stance from their Facebook profile photo.

Michal Kosinski, an associate professor at Stanford University, applied a facial recognition algorithm to 1,085,795 faces obtain from online social media profiles.

Of this dataset, 977,777 came from dating website users in the U.S., UK, and Canada who had self-reported their political orientation.

The other 108,018 faces were from Facebook users in the U.S. who also self-reported their political orientation and additionally completed a 100-item personality test.

The algorithm compared each participant’s facial features to the average facial features of liberals and conservatives. The technology used these similarity measurements to determine the likelihood that a participant was either a conservative or a liberal.

The results showed that the algorithm was able to predict political orientation alarmingly well and with similar accuracy across countries and social media platforms.

Among U.S. Facebook users, this accuracy hit 73%. Among U.S. dating website users, accuracy was 72%. Among dating website users in the UK and Canada, accuracy reached 70% and 71%, respectively.

The post Can AI Tell Your Politics By Looking At Your Face? first appeared on Humintell.

Can Smiling Improve Your Mood? Research Says Yes.

Emotions and Facial Expressions

We all know that emotions give our lives meaning, and life without emotions is impossible to imagine.

Emotions are a vestige of our evolutionary history and are primarily controlled by an archaic part of the brain.

This is why Dr. Matsumoto describes emotions as immediate, involuntary, automatic, and unconscious reactions to things that are important to us.

Emotions help us react in some situations with minimal conscious awareness and are triggered by a universal, underlying psychological theme.

When triggered, they recruit an organized system of reactions that produce specific physiological signatures, direct our cognitions, and produce specific types of feelings.

Importantly, emotions produce specific, nonverbal behavior in the face, voice, and body.

Different emotions are expressed by different, specific, unique facial configurations (facial expressions) that are universal to all cultures, regardless of race, nationality, ethnicity, religion, gender or any other demographic variable.

Facial Feedback Hypothesis

If emotions produce specific universal facial expressions, can facial expressions in turn affect your emotions? According to the facial feedback hypothesis, they can.

But is this actually true?

Scientists have been interested in the idea of a facial-feedback hypothesis since the 1800s (Source: Betterhelp) and modern researchers have continued to study the hypothesis to this day.

Smiling is Good for Your Heart

One study conducted by clinical psychologists Tara Kraft and Sarah Pressman showed the positive effects of smiling. Turns out, smiling can be good for your heart in stressful situations.

How Masks Hinder PolitenessFor their study, the researchers examined participants’ heartbeats, since stress and heart health are related.

17o participants were split into 2 groups: one knew what the study was about, while the other didn’t.

In the training stage, the researchers taught the volunteers how to either hold their faces in a neutral expression, hold a social smile (upper right hand image), or hold a Duchenne smile (upper left hand image).

The researchers monitored the participants’ heart rates as they performed various tasks; both groups were required to use their hand to quickly trace a star reflected in a mirror, followed by placing their hand in a bucket of ice water for one minute.

While completing these tasks, each person had to hold chopsticks in their mouth which activated muscles corresponding to a forced smile.

They found the participants who were instructed to smile, and in particular those whose faces expressed genuine or Duchenne smiles, had lower heart rates after recovery from the stress activities than the ones who held their faces in neutral expressions.

Even the volunteers who held chopsticks in their mouths, that forced the muscles to express a smile (but they had not explicitly been instructed to smile), had lower recovery heart rates compared to the ones who held neutral facial expressions.

Interestingly, those who smiled genuinely during the trial recovered the fastest, followed by people with fake (social) smiles. Those with neutral smiles had the slowest recovery.

Even Fake Smiles Can Improve Mood

Recent research also suggests that fake or social smiling can make people feel happier.

An international collaboration of researchers led by Stanford University research scientist Nicholas Coles published a study in Nature Human Behavior.

As part of the Many Smiles Collaboration, a total of 26 research groups from 19 different countries and over 3,800 participants were involved. The average age of the participants was 26 and over 70% were women.

The researchers created a plan that included three well-known techniques intended to encourage participants to activate their smile muscles:

  1. One-third of participants were directed to use the pen-in-mouth method
  2. One-third were asked to mimic the facial expressions seen in photos of smiling actors
  3. The final third were given instructions to move the corners of their lips toward their ears and lift their cheeks using only the muscles in their face

In each group, half the participants performed a small physical tasks and simple math problems while looking at cheerful images of puppies, kittens, flowers, and fireworks, and the other half simply saw a blank screen.

They also saw these same types of images (or lack thereof) while directed to use a neutral facial expression. After each task, participants rated how happy they were feeling.

After analyzing their data, the researchers found a noticeable increase in happiness from participants mimicking smiling photographs or pulling their mouth toward their ears.

Interestingly, the researchers didn’t find a strong mood change in participants using the pen-in-mouth technique but the evidence from the other two techniques was clear.

It provided a compelling argument that human emotions are somehow linked to muscle movements or other physical sensations.

For more on how smiling boosts your mood, visit this past blog post

The post Can Smiling Improve Your Mood? Research Says Yes. first appeared on Humintell.