January 23, 2014

Don’t Overthink Your Credibility Assessments

By Dr. Ken Broda-Bahm: 


When it comes to assessing someone’s believability and deciding whether they’re lying to you or not, which works best: your quick “gut” intuition or more sustained and careful thought about it? It turns out, the answer is “neither.” Immediate choices, as well as decisions made after longer deliberation, are both less reliable than judgments rendered after turning our minds away to focus on another task. This finding of a recent study (Reinhard, Greifeneder, & Scharmach 2013) suggests that unconscious processing is a better guide when assessing truthfulness. This advantage for unconscious processing might relate to some of the reasons I have always been a skeptic on human lie detecting ability. Short of Pinocchio’s nose, there are very few reliable physical tells, and very few techniques that work in improving a receiver’s ability to detect deception. Of course, there are potential exceptions, like the Lie to Me inspiration, Paul Ekman, but recent research suggests that human nonverbal response is far less universal and interpretable than previously believed.

Generally, the research supports a healthy skepticism. “Humans have been shown to be generally poor deception detectors,” the Reinhard et al., authors write, noting that individuals tend to achieve an accuracy rate that is only slightly above the rate of a coin flip. The unreliability of these physical cues, in addition to the prevalence of false beliefs about what does or doesn’t reveal a liar, essentially means that the less we rely on our own conscious thoughts about truthfulness, the better we do in assessing actual truthfulness. “Judges who were kept from consciously deliberating outperformed judges who were encouraged to do so or who made a decision immediately.” This finding has implications in many legal contexts including attorney credibility or jury selection. For this post, however, I want to take a look at how this research, particularly within the broader frame of credibility and not just lie detection, should influence the way we assess witnesses.


The Study: To Gauge Truthfulness, Don’t Think About it Too Much

Much of the previous research in lie detection has simply asked people to decide whether the person in the stimulus is lying or telling the truth. That setting emphasizes conscious thought, drawing upon many of the naive biases that each individual brings to that test. To look at how much these false beliefs might reduce the accuracy of lie detection, the researchers varied whether their study participants made truth or falsity determinations right after watching the message, after short periods of conscious thought, or after short periods of unconscious thought which they brought about through a “distraction” exercise of solving puzzles.  

Using three different sets of stimuli groups (e.g., an employment situation in which applicants were reporting either a true or a false internship) and across five experiments, the researchers found that unconscious deliberation did best: “Lie/truth judgments formed after short periods of unconscious thought were significantly more accurate than judgments formed right after watching a message (standard control) or after periods of conscious thought.”

Not only were these advantages statistically significant, they were substantive as well. A three- minute period of distraction led to successful lie detection at a rate that was six times better than the comparison groups. The authors note that these effect sizes “are near to those observed when judges use specific interview techniques developed to detect deception.” 

One alternate reading of the basic study is that it may not be so much that unconscious thinking is good, as it is that conscious thinking is bad. The authors consider this possibility as well, writing that “one could argue that the distraction period merely offers the bliss to forget irrelevant or nondiagnostic information.” To address that possibility, they also added a “mere distraction” condition to compare to their unconscious processing conditions where participants worked on distraction task without knowing that they would be determining truthfulness later on. Even in that experiment, the participants who completed the puzzles, while knowing they would make a truth assessment later on, did better than those who merely completed the puzzles.

The researchers also demonstrated that those allowed to process the task unconsciously used more and better diagnostic cues than the other groups, supporting the idea that there is something going on during that background reasoning time that allows participants to give less attention to unreliable beliefs about what honesty looks like, and to give more attention to what really matters. “The present findings offer a solution to the apparent puzzle of humans’ seeming incapacity to distinguish lies from the truth.” The authors conclude, “Five experiments have not only documented that conditions of unconscious thought help individuals to attain higher accuracy rates in deception detection, but also shed some light on the processes underlying this performance boost.” 

The Implications for Assessing Your Witness’ Credibility

One question the research doesn’t answer is what mode of processing people are likely to rely on in the more natural setting of their day-to-day determinations of credibility. In a trial context, however, there are reasons to believe that the ultimate determinations of credibility or believability would not just be made immediately or consciously, but would instead be made after some unconscious processing during the “distraction” of all the other information presented in trial. In other words, while the witness is on the stand, jurors might certainly be engaged in conscious thought on “is this person believable?” But they aren’t required to give any immediate answer to that question, so as the trial moves on to other witnesses and other content, it is likely that jurors would continue to unconsciously process their assessment of that witness along with all the other accumulated information during the trial. 

Of course, that should be studied specifically in a trial context. But in the meantime, let me suggest a few thoughts on how we approach witness credibility assessments. 

1. Don’t Overemphasize Your Conscious Response

Lawyers predictably have an analytical bias, a preference for what one can adduce reasons for. So you might have a first impression of your client-witness, but then you will have a long opportunity to think of responses to that first impression and other reasons that this client is better or worse than you originally thought. The research reminds us that this sustained analysis is not the most natural, nor the best way to assess credibility. Instead, a better strategy is to save at least some role for our own less conscious gut feelings, or the gut feelings of others . 

2. Crowd-Source Your Witness Evaluation

The less conscious gut feelings of others can be invaluable in helping attorneys avoid the positive groupthink that can sometimes be a by-product of working closely within a team. Because it’s your job to put your case in the best light, you start seeing it in the best light as well, and that can extend to witness evaluations. The best way to check your perceptions is to get reactions to witness video clips in a focus group or a mock trial setting. By doing that, you are more likely to see the full range of evaluations. 

3. Delay the Evaluation

In a research setting, there is a tendency to think that immediate is better. Think about the dials used to record instant reactions while the presentation is still occurring. The belief is that the instant reaction is the more reliable reaction. This research puts a question mark over that. It may actually be better to delay the evaluation. For example, if an attorney’s summary argument used a witness clip during the presentation, ask about that witness in a questionnaire once the presentation is done. That delay isn’t a disadvantage, but instead creates conditions that are more like the most successful group in the experiment, and may also be more like a courtroom setting in which full evaluations are delayed as well. 

Unlike the study authors, litigators and witness consultants are ultimately less concerned with whether evaluations are correct or incorrect, and more interested in whether they are favorable or unfavorable. The study is still important, though, based on the likelihood that your fact finders will unconsciously process witness credibility throughout trial, ultimately relying most on the mode that turns out to be most reliable. By adapting not only to conscious and analytical reactions, but also to the intuitive processes that humans have developed over time, you’ll have a better chance of knowing whether you’re putting a Pinocchio on the stand. 


Other Posts on Witness Credibility: 


Reinhard, M. A., Greifeneder, R., & Scharmach, M. (2013). Unconscious processes improve lie detection. Journal of personality and social psychology,105(5), 721. 10.1037/a0034352

Photo Credit: Sjdunphy, Flickr Creative Commons

Related Posts Plugin for WordPress, Blogger...