Social Engineering Blogs

An Aggregator for Blogs About Social Engineering and Related Fields

Research Shows Cats Have Nearly 300 Facial Expressions

If you’re a cat lover, you probably already know that your feline friend uses subtle body language and nonverbal cues to communicate with you.

You may even know that cats usually don’t meow to each other, only to humans (a tactic developed through adaptation and evolution).

But how do cats communicate with each other?

Recent research that studied domestic cats in a cat cafe in California shows they do “talk” to each other and that their means of communication amongst themselves are even more complex than we originally thought.

The study entitled “Feline Faces: Unraveling the Social Function of Domestic Cat Facial Signals” was recently published in Behavioural Processes.

In it researchers tallied nearly 300 different feline facial expressions (276 to be exact), used to communicate hostile and friendly intent and everything in between.

What’s even more fascinating, researchers believe cats may have evolved this range of sneers, smiles, and grimaces over the course of their 10,000-year history with humans.

The Cat Study Methodology

Over 10 months, researchers Lauren Scott and Brittany N. Florkiewicz observed 53 domestic shorthair cats in a Los Angeles cat cafe and recorded 186 separate meetings between the animals.

The team recorded 194 minutes of footage after the café closed to visitors for the day and focused specifically on recording and reviewing the cats’ facial expressions when interacting with each other.

After obtaining the recordings, they used the Facial Action Coding Systems designed for cats and compared the complexity and compositionality of facial signals produced in affiliative and non-affiliative contexts.

To measure complexity and compositionality, they examined the number and types of facial muscle movements (AUs) observed in each signal.

Similarly, research scientists who study facial expressions of emotion in humans also use FACS and AUs to code faces and facial muscles specifically.

276 Unique Cat Facial Expressions Found

After observing, recording, coding and studying the cats’ facial expressions, researchers found 276 unique facial expressions (which is not far removed from the 357 expressions produced by chimpanzees).

These different expressions were sorted into 2 major categories

  • Friendly: which constituted 45% of the expressions
  • Aggressive: which made up 37% of the expressions

The remaining 18% were too ambiguous to clearly categorize.

According to Florkiewicz, each cat expression combined about four of 26 unique facial movements, including parted lips, jaw drops, dilated or constricted pupils, blinks and half blinks, pulled lip corners, nose licks, protracted or retracted whiskers, and/or various ear positions.

By comparison, the human face includes over 40 structurally and functionally anatomically independent muscles, each of which can innervate independently of each other. This makes the face one of the most complex signal systems available to humans.

According to Science Magazine, what exactly the felines were “saying” to one another with these expressions remains unclear, Florkiewicz says. But overall, cats tend to move their ears and whiskers toward another cat during friendly interactions, and to move them away from their compatriot during unfriendly interactions. Constricted pupils and licking lips also tend to accompany such rivalrous encounters.

Interestingly, some of the cats’ friendly expressions resemble those made by people, dogs, monkeys, and other animals. This suggests that these species may share “a common play face.”

Although the researchers haven’t been able to compare their results with those of wild felines, they do know that all close relatives of the domestic cat are ferociously solitary animals.

Pet cats might have retained some of that defensive communication, Florkiewicz says, but these domestic descendants probably started to pick up friendly facial expressions as they gathered to await humans’ dinner leftovers.

Historically cats have received less research attention than their canine counterparts and their cognitive abilities were less recognized. However, in the last 10 years, research focusing on trying to better understand our feline companions has grown.

To read more about how good your cat is at emotion recognition, visit this past blog post.

The post Research Shows Cats Have Nearly 300 Facial Expressions first appeared on Humintell.

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.