Social Engineering Blogs

An Aggregator for Blogs About Social Engineering and Related Fields

Social Influence in Investigative Interviews

We talk a lot about how to understand people’s emotions and how to read what they say accurately, but what about learning how we can change our behavior so that they are more forthcoming?

This is the subject of a recent paper by Humintell’s Dr. David Matsumoto and Hyisung Hwang. In this study, they sought to examine the impact that different cues of authority had on information interviews. This was an attempt to determine whether the truth-tellers or the liars in a mock interview context were more likely to convey information, depending on the influence of the interviewer.

After recruiting participants, they were divided into a treatment and control group. The treatment group was encouraged to commit a mock crime but told not to disclose that they had done so, while the control group simply proceeded directly to the informational interview.

This design split the participants into those who had to lie and those who had to tell the truth during the interview, with each being encouraged by a small monetary payout to convince the interviewer that they were telling the truth and had not committed the crime.

However, there were other experimental conditions at play, namely the environmental context and appearance of the interviewer. In order to prime a sense of interviewer authority, for instance, some of the interviewer wore intimidating suits or had impressive degrees and law enforcement posters on the wall.

Nothing beyond those environmental factors was different. All of the interviewers followed the same script and spoke in comparable fashions, but previous research has seen profound differences in perceptions of authority based on these apparently simple differences.

Overall, they found that the authority conditions got truth-telling participants to freely volunteer more information than their counterparts in low-authority conditions. However, there was no effect for the liars.

Interestingly, the effect of the authority condition seemed to hold even for the truth-tellers written statements, even though these were completed after the interview. This suggests that there is some lasting impact from an impression of interviewer authority.

This helps expand on previous work that looked at how to evaluate interviewees’ truthfulness and emotions by looking instead at what can promote efficacy by the interviewer. It appears that these signs of authority, even if limited to clothing, can result in more forthcoming interviews.

While this appeared to only work on people who were already telling the truth, that does not take away from the usefulness of this information. Many interviews are conducted on truthtellers, with the struggle sometimes involving getting tight-lipped people to speak more freely. Wearing a suit or even putting your diploma on the wall might help here!

In the meantime, there is always more that needs to be learned about how to tell if your interlocuter is lying, so check out Humintell’s exciting training procedure!

In Dr. Matsumoto’s Own Words

We’ve been talking about new microexpressions research for the last two weeks, so we’ll all benefit from Dr. David Matsumoto’s own insights.

In this new video, Dr. Matsumoto discussed his recent article on microexpressions and the importance of precisely studying microexpressions and their neurological correlates. He emphasizes the striking difference in neurological patterns between expressions at or above 200 ms and those below.

Digging into Deception Detection

lying-deceit-deceptionLast week we wrote about the importance of distinguishing micro and macro expressions, but we passed too quickly over deception detection.

Microexpressions differ from their longer lasting counterparts in many ways, but one of the most salient is the fact that they can betray underlying emotions. The fleeting microexpression can show anger or surprise where someone may be trying to conceal it, and effective deception detection can often involve noticing these microexpressions, as hard as that might be for a layperson.

As Drs. Matsumoto and Hwang point out, there are many factors that make people better at deception detection. These include personality features like openness to experience and conscientiousness but also empathy and emotional regulation.

One of the major takeaways from the study we blogged on last week was the more precise definition of what microexpressions are, committing to seeing them as incredible fleeting and at least under half a second.

This was elaborated on in a 2018 study by Drs. Matsumoto and Hwang which looked at microexpressions as keys to differentiating truth from lies, operationalizing this particular definition of the length of microexpressions. They point out that previous research found mixed results in connecting microexpressions to deception detection, but that such research had failed to rigorously define how long a microexpression is.

Much of this past literature, though it found the potential for microexpressions, was that it did not properly define microexpressions based on observations of how they actually function. Some of this research categorized microexpressions as lasting for up to four seconds!

Such a revised definition makes a great deal of sense when looking at deception detection. Much of the attempt to conceptualize microexpressions comes from trying to see their role in showing spontaneous and concealed expressions. Longer durations are less likely to be signs of repressed and concealed emotions.

This 2018 study then sought to look at the role of microexpressions in deception detection given this more rigorous definition. First, they looked at whether microexpressions under half a second occurred and if they could reliably distinguish truth from deception. They also looked at longer expressions to see if these could be seen as consistently spontaneous.

They found that such expressions were generally frequent and acted as reliable indicators of deception, much more so than expressions longer than a second. Interesting, very short expressions of less than 0.3 seconds were not helpful indicators.

We are curious what our readers have to say about this, however. Do you notice these fleeting expressions? Do you think that they are uniquely distinct from longer expressions? Please comment and let us know what you think!