Faking Detection

"I am the last and highest court of appeal in detection."

Arthur Conan Doyle's Sherlock Holmes in The Sign of Four

Have you ever wondered if the job applicant sitting across the table from you was completely honest with you? Maybe embellishing their credentials? Or perhaps even outright making them up?   

Applicant faking (or use of deceptive impression management - IM) is potentially problematic for organizations because less qualified individuals may use such tactics to get hired over more qualified ones who do not use deceptive tactics. This situation introduces a systematic source of inaccuracy into the interview process. And, of course, organizations do not want to hire someone who only pretended to be a good fit with the job.

Logically, one solution for organizations is to make sure that hiring managers or interviewers are able to detect when applicants use impression management tactics and to distinguish honest from deceptive attempts. This information should then be used to give lower evaluations to applicants who use deception or eliminate them from the selection process. Interestingly, when asked about it, interviewers generally believe they can detect when applicants use deceptive tactics (i.e., lie to them). But is it actually the case?

What if interviewers acted like Sherlock Holmes and tried to identify applicants who are honest and those who are deceptive? Would they be as successful as the famous British detective?

This is one of the questions I examined in my research on faking detection in job interviews. And unfortunately (for interviewers and organizations) the findings suggests a quite pessimistic reality. 

In a series of lab studies, my colleagues Adrian Bangerter (University of Neuchatel, Switzerland), Julia Levashina (Kent State University, OH) and I have highlighted that detecting IM or faking tactics is a difficult task. In four studies, we found that interviewers correctly detected on average between 12% and 19% of deceptive tactics used by applicants (and between 13% and 25% of all types of IM tactics - see article here). 

Such results have been confirmed in a study in the field, with applicants interviewing with experienced interviewers for actual jobs in recruitment agencies. In that study, we found that interviewers’ perceptions of tactics used by applicants failed to converge with the tactics actually reported by applicants. In other words, interviewers had no idea how honest or deceptive the applicants were during the interview (see article here). 

Although interviewers in general are not great faking detectors, it is possible that some interviewers possess individual traits or characteristics making them better at this task. Therefore, my research also attempts to identify individual differences explaining detection abilities.

So far, my research has mostly highlighted factors that are NOT associated with detection abilities. For instance, in the studies described above we have shown that interviewer's age or their interviewing experience does not predict detection abilities. As an example, experienced HR managers and interviewers did not outperform novice university students at this task. It also seems that interviewer's personality does not play a major role in detection.

However, a combination of higher levels of cognitive abilities and general trust may (to some extent) help interviewers be better detectors. As such, an interviewer might be better to identify applicants who fake because s/he is both more intelligent and more sensitive to whether or not an applicant is likely to be trustworthy. Yet, it is important to note that those characteristics only explain a relatively small portion of detection abilities (see article here).

Interestingly, we also found that emotional intelligence is not associated with better deception detection, suggesting that the concept of "detection wizards" is likely a myth (see article here).

Some of my ongoing projects aim at better understanding how and when applicants fake, and finding ways for interviewers to become better IM and faking detectors. With the support of a research grant from the Canadian Social Sciences and Humanities Research Council (SSHRC IDG 430-2015-00098), I'm working on several studies to explore those issues

For instance, one of the key reasons why interviewers do not perform better at faking detection is that they do not rely on the right cues. Indeed, non verbal cues like gaze aversion, fidgeting, or body and arms movement are NOT good indicators of faking (see article here)!

In a recent paper, my colleague Deb Powell (University of Guelph, ON) and I used a tool from forensic psychology research called Criterion-Based Content Analysis (CBCA), and adapted it to the employment interview context. For instance, we found that CBCA could be a valid indicator of faking/deception if three conditions were met. (1) Applicants were free to be honest or deceptive in their answers, (2) applicants formatted their responses as stories, and (3) an overall CBCA indicator was used.  Using this approach, we could correctly identify about 65% of fakers (vs. 50% for chance level).

Our (long-term) objective is to be able to train interviewers to rely on the right cues to become better faking detectors.