Over the years we have been consulted on cases in which an investigator was absolutely convinced that a particular suspect was lying when, in fact, the person was telling the truth. In other instances guilty suspects were able to get through an interview without having their lies detected. Behavior symptom analysis is certainly not 100% accurate. However, if proper techniques are employed in eliciting and evaluating a person's behavior, an investigator's accuracy at detecting deception can far exceed chance levels. On the other hand, if no specific procedures or protocols are followed, studies have consistently reported low accuracies in detecting deception based on observing another person's behaviors.

For those who have attended our seminars this web tip will be a reminder of the importance of being aware of variables that may affect a suspect's behavior. For investigators who have not received training in behavior symptom analysis, it should serve as a warning of how easily a person's behavior can be misinterpreted. To increase the accuracy of detecting deception through behavior symptom analysis an investigator should:

(1) Establish the subject's baseline behaviors at the beginning of an interview

On several occasions investigators have reported that the reason they believed a suspect was lying was because of poor eye contact. When these suspects were subsequently interviewed in our office we found that many of them failed to exhibit normal eye contact even when asked non-threatening background questions. In some instances the suspect's poor eye contact could be attributed to cultural differences. In most cases, however, the cause for the suspect's poor eye contact was not readily apparent. It could have been the result of a shy or submissive personality or perhaps an underlying neurological problem. Perhaps the suspect was frightened by the fact that they were dealing with authority figures. Regardless, when a suspect does not establish normal eye contact when spelling their last name or giving their address, the investigator should not use the suspect's poor eye contact as an indication of guilt when the suspect is later asked about the crime under investigation.

It only takes a few minutes at the outset of an interview to establish a suspect's normative behaviors by asking non-threatening background questions. Examples of questions that can be asked include:

"Please spell your last name for me."

"What is your first name?"

"What do most people call you?"

"What is your current address?"

"How long have you lived there?"

"Who do you live there with?"

When the suspect is answering these introductory questions the investigator should carefully observe the suspect's eye contact (normal, poor, forced), communication style (articulate, difficulty comprehending or speaking English, slow or rapid response delivery) and emotional state (composed, unconcerned, frightened, angry, withdrawn). Once these baseline behaviors are established the investigator will be in a much better position to evaluate specific behavior symptoms that occur later during the interview when the suspect answers questions about the issue under investigation.

(2) Clearly define the purpose for the interview

A common cause for misleading deceptive behavior symptoms occurs because an innocent suspect experiences anxiety as a result of some transgression unrelated to the issue under investigation. Consider a situation where a police officer approaches a man in an airport and pulls the man aside for questioning. The officer is interested in determining whether the man is a terrorist but never explains the purpose for questioning him. The man is not a terrorist but does have a number of unpaid parking tickets as well as a pending court appearance for failure to make support payments. If the officer asks general questions without clearly identifying the purpose for the interview, it would be perfectly natural for the man to exhibit apparent deceptive behavior symptoms.

In the airport example, it would be appropriate to explain that the man had been randomly selected for additional screening to verify his identity. During the formal interviews our staff conducts on individuals suspected of possible criminal conduct, after the initial background questions are asked, the purpose for the interview is clearly identified and the subject is then asked to state a position on that issue. One possible statement would be as follows: "John, this afternoon I'll be asking questions about (issue under investigation). Some of the questions I'll be asking I already know the answer to. The most important thing is that you be completely truthful with me before you leave today. Before we go any further, let me ask you, did you (commit issue?)

(3) Question a suspect in a non-accusatory and non-threatening manner

Many behavior symptoms associated with deception represent the suspect's efforts to reduce anxiety or occur from the fear of being caught lying. However, an innocent person may exhibit these same behaviors if he experiences anxiety as a result of intimidation or fear of not being believed. If an investigator appears aggressive in his posture (sitting three feet from the suspect with a forward lean) asks questions in an accusatory tone ("You're the pervert who did this, aren't you!") or asks questions in a rapid-fire approach, forcing the suspect to answer quickly, the innocent suspect is bound to become protective nonverbally (put up barriers, break frontal alignment, exhibit little or forced eye contact), be verbally guarded or defensive and may become confused and offer inconsistent answers.

It is critical that a suspect's behavior be interpreted in the context of the interaction with the investigator. For example, there are a number of behavior symptoms which, if they occur during a non-accusatory interview, are associated with deception. However, these exact same behaviors may be an indication of truthfulness if they occur during an accusatory interrogation. To achieve the greatest accuracy in assessing whether or not a suspect is telling the truth the investigator should evaluate that person's behavior during a non-accusatory interview. An accusatory interrogation should be reserved for suspects whose guilt has been reasonably established.

(4) Elicit a sufficient sample of the subject's behavior

It is amazing how people will listen to a 30 or 60 second segment of an interview on a television news broadcast and form hard and fast judgments about the subject's truthfulness. In most circumstances it is not possible to accurately detect deception based on analyzing a minute or two of a person's behavior. No professional detection of deception examiner would ever render an opinion of a person's truthfulness based on analyzing an interview that lasted only a few minutes.

The interviews our staff conduct last approximately 30 minutes in length and involve dozens of questions relating to the issue under investigation. Many guilty suspects are able to lie convincingly if they are only asked two or three questions about the crime they committed. However, it becomes increasingly more difficult to lie successfully if 15 or 20 questions are asked relating to a crime the suspect committed. An innocent suspect, of course, has no difficulty answering an unlimited number of questions about a crime he did not commit.

An interview designed to assess a person's credibility should incorporate both investigative questions as well as behavior provoking questions. Investigative questions cover such things as the suspect's opportunity and access to commit the crime along with their motivation and propensity to commit the crime. Behavior provoking questions, on the other hand, are designed to elicit different responses from innocent or guilty suspects. For example, when innocent suspects are asked how the investigation will come out on them they tend to express much more confidence than when guilty suspects are asked that same question.

(5) Do not over-evaluate a suspect's nonverbal behavior

It has been said that nonverbal behavior accounts for more than half of the information communicated between two people. This finding does not mean that nonverbal behavior is more representative of a person's truthfulness than verbal or paralinguistic behavior. It merely indicates that there is a tendency to place an inordinate amount of weight on a person's nonverbal behavior.

Of the three channels of communication, the nonverbal channel is the most subject to internal and external influences that have nothing whatsoever to do with a person's truthfulness. During an interview conducted in a controlled environment, the suspect's posture provides perhaps the best insight to that person's level of confidence and emotional involvement. Research indicates that a subject's confidence and emotional involvement are statistically good predictors of truth or deception. The presence or absence of illustrators (the hands moving away from the body) as well as leg and foot behaviors can also provide meaningful information for detecting deception.

Adaptor behaviors (the hands coming in contact with the body, e.g., touching the nose, scratching an arm, wringing the hands)) have the most complex origins and are often erroneously associated with lying. Especially when evaluating adaptor behavior, the investigator needs to establish behavioral consistency before considering a particular movement or gesture as an indication of deception. In other words, if a suspect scratches his neck one time when answering a question about committing a crime the behavior is probably meaningless. However, if a suspect establishes a pattern where he frequently scratches his body when denying involvement in a crime, the consistency of that behavior indicates that it is significant.

In conclusion, detecting whether or not a person is lying based on observing that person's behavior is not a simple manner of identifying certain telltale symptoms of deception. Indeed, there are no unique behaviors that reliably indicate if a person is telling the truth or lying. However, if the preceding guidelines are followed when conducting an interview and interpreting the subject's behavioral responses to interview questions, an investigator can achieve accuracies far above chance levels.