Was the runner safe or out? Strike or ball? It may just depend on what team you’re cheering for.
We have all been in these conversations before, whether we are arguing about a penalty in a football game, the acceptable temperature in the car, or whether two people happen to look alike or not. As it turns out, many of our opinions in these circumstances are driven by a variety of cognitive biases. In the basic example of a runner being safe or out, our opinion may very well change depending on which team we are rooting for. If we are complaining about a penalty not being called in a football game, we may be more aware of it (and intentionally looking for missed calls) because we had just listened to a podcast where they discussed bad refereeing. These biases may fuel heated debates at a sporting event or your local watering hole, but when it comes to investigations, they can prove costly for you, your organization, and the subjects you interact with.
In a recent podcast with psychologist Jeff Kukucka, PhD, we discussed the influence of biases in investigations and how it may directly impact the interview process. These cognitive biases, as explored by researchers, are highly prevalent in everyday life—including our stubborn mindset in defending our slanted opinions. These perspectives infiltrate into our criminal justice system, ranging from our interpretation of evidence to our opinion on the credibility of a person in question.
Confirmation Bias
Have you ever been told, prior to meeting somebody, that they are unfriendly and abrupt? Then, when you interact with that person, do you tend to agree with that opinion? There may be a few issues with the reliability of your assessment. First, if you enter a conversation with someone under the presumption that they are unfriendly, naturally, you may avoid a more intimate conversation. Therefore, this person may reciprocate your approach, resulting in their perceived unfriendliness—a reaction that you may have created based on your own behavior. Additionally, even if there were moments of friendliness, confirmation bias often impacts these perceptions by ignoring evidence that contradicts our theory and instead only relying on information that supports it. Regardless of the reality of the interaction, we report back to our allies with, “You’re right; that guy was a jerk.”
This can be problematic in an investigation as interviewers often theorize the crime, including hypothesizing the responsible parties, motives, and methods. Confirmation bias suggests that interviewers may dismiss or explain away evidence that contradicts this predisposed theory while only highlighting information that aligns with their assumptions. A common contributor to confirmation bias is the reliance on observing physical behavior in the interview. If an interviewer observes a person avoiding eye contact, crossing their arms, becoming flush, and acting fidgety, they may wrongly conclude that the person is being deceptive. However, this behavior may be interpreted differently based on the context provided prior to the interview. If you watched a video of this exact behavior but were told this was a victim interview, would you still suspect deception, or would you assume they were just nervous or embarrassed? If our opinion is based on a presumption we’ve decided before the investigation takes place, we are increasing the risk of falling victim to confirmation bias.
Prevalence Effect
The research literature explains this phenomenon as a person being more likely to misidentifying a “target” if the frequency of its appearance is low. Researchers tested this with airport security personnel, determining that the error rate for missing a weapon being smuggled through security screening was higher when officers did not experience a high frequency of this occurring. The implications of this research may be terrifying for those of us who frequent airports, but it also directly influences investigators and the interview process.
For example, if interviewers are instructed that they must talk to a group of ten people and that eight of them are guilty, the opposite effect may occur. Investigators are more likely to become guilt‑presumptive, resulting in more coercive tactics, leading questions, and aggressive behavior. In turn, the interviewee responds more defensively, creates additional resistance, and is less likely to provide actionable intelligence. The higher the frequency or likelihood of something occurring, the more alert the investigator will be—but this also may influence their strategic process.
This same bias can result in profiling based on race, gender, age, or other characteristics where an investigator places a higher or lower frequency rate on one or more of these descriptive categories. As an example, a local convenience store has a sign on its door that says only two students are allowed in the store at a time.
If we assume that this location has had a high frequency of theft or vandalism by youth, then they are becoming more vigilant in targeting this type of customer. The prevalence effect suggests that store personnel may be more successful in identifying concerning behavior from their youthful shoppers but have a higher error rate identifying those of a more senior age. This behavior and mindset are the catalysts for profiling, and then, when combined with “successful” target identification, only fuels confirmation bias.
Does Technology Help?
Maybe. The availability of technology to investigators should be a way to standardize their approaches and remove some of the human elements in which bias is prevalent. However, there are also many biases born out of the use of technology. One concept, known as automation bias, is when people tend to rely on the outcome of predictive technology or other automated decision-making processes rather than seek information that contradicts it. This bias can range from the lack of any alarms or indicators that there is a physical threat at the store or the false sense of data security and digital protection. This can also be found when relying on automated testing to determine if a person is honest or potentially involved in wrongdoing. Additionally, with the increased use of facial recognition technology, investigators may rely too much on automated reporting versus supporting this tool with manual investigation.
Regarding facial recognition technology, there are additional biases that have been shown to impact the human element in identifying whether two people look alike. For example, if you are shown a picture of two people and informed that they are father and son, you are more likely to find characteristics in which the two allegedly look alike. However, if you are shown the same pictures but told that this was a teacher and student, you are less likely to feel they are comparable and may not intently look for those characteristics. This is where the use of standardized technology combined with the human element may assist in both mitigating bias and searching for contradictory opinions.
How Do I Fix This?
The first step in addressing biases is admitting you have them. Bias is human nature, and what we have discussed above is not necessarily intentional but a result of cognitive bias and our brains attempting to fill in the gaps for us. Researchers like Dr. Kukucka suggest that one way to mitigate bias is to have protocols and standard procedures that assist in removing the variability of the human element. For example, when interviewing applicants for a job, hiring managers should use consistent questions and processes to avoid bias influencing the interview process.
In addition to creating processes, researchers suggest that we seek counter-opinions on our theories and perspectives. Operating in an echo chamber only fuels the biases we have (and is a form of confirmation bias). Instead, investigators should always inquire whether their evidence may be wrong, seek alternative theories, and use peers or supervisors to provide critical feedback on their thought processes. If one interviewer feels like a person is being dishonest based on their physical behavior, and everyone agrees, then this just perpetuates the same (potentially inaccurate) assumption. Instead, we should seek opinions that suggest alternative reasons for that person’s behavior: Are they nervous? Embarrassed? Did something else cause them to be concerned today? These types of questions assist us in gaining perspective in our investigations.
Most importantly, if you feel like your rival team is getting all the beneficial calls from the referee, take a minute to think about how you would feel if they were wearing different jerseys—and don’t just listen to your hometown radio station (because they’ll just agree with you and strengthen your confirmation bias). Expand your circle, seek contradictory opinions, and allow other perspectives to help defeat your biases.
Click this link and answer questions about this article to earn CEUs towards your CFI designation or to learn more about the advantages of becoming a Certified Forensic Interviewer.