How Evidence Can Be Trumped by a Suspect’s Atypical Behavior


For these individuals, justice might not be done.


  • Sometimes, a suspect’s non-normative or atypical behavior influences decision-making in investigative and courtroom contexts.
  • Autistic adults sometimes display behavioral characteristics that are commonly, and erroneously, associated with an intention to deceive.
  • Reliance by legal decision-makers on a suspect’s misleading behavioral cues can bias the direction of police investigations and court judgmentsIf you were suspected by the police of being involved in a crime and they were interviewing you about what might have transpired, you would hope that they would focus on the “facts of the matter.” In other words, you wouldn’t like their judgments to be biased by what you look or dress like, how attractive you are, your race, where you live, or perhaps by any behavioral idiosyncrasies you display. Similarly, if you were charged with a crime and ended up giving testimony in a courtroom, you would hope the judge and the jury were only swayed by the “hard” evidence. And, of course, if you knew you were innocent, the possibility that such consequential decisions and judgments might be swayed by extra-evidential factors such as those just listed would be extremely disturbing.

    What then if you were a person who, possibly feeling under duress from being interviewed by police or interrogated in a courtroom, looked very nervous, stammered, fidgeted, or didn’t look your questioner in the eye? Or what if you didn’t display the emotions that might typically be expected or you wandered off on a tangent when answering questions? Again, the hope would be that any such behavioral displays would not shape how police or the courts would interpret the overall body of evidence.

    If you think about it, you will likely realize that any of us might display behaviors such as I have described above. Think of the people you know who are terrified at the thought of public speaking, and how they might react under the duress likely to be experienced under police or legal questioning.

    Some individuals, of course, will be more likely to react in such ways than others. For example, some individuals with mental health conditions may display nonverbal behaviors that are considered atypical. Individuals from culturally diverse backgrounds may misinterpret questions or unfamiliar verbal expressions and answer in ways that seem evasive to the questioner. Others with neurodevelopmental conditions may display characteristics during social interactions that are unexpected or unusual.

    Are there any grounds for thinking that individuals who behave in such ways might be disadvantaged in any interactions they have with the justice system? It is extremely difficult to “get into the minds” of police officers or jurors or judges and determine exactly why they really arrived at a particular judgment about a suspect or defendant. It is even more difficult to gather a sufficiently large database on the determinants of such judgments to be able to say confidently that biases are unlikely likely to exist.

    We do know, however, that behaviors that are not normally expected in social interactions—such as gaze aversion, fidgeting and shuffling, inappropriate emotional expressions, and evasive responses—are often reported as being interpreted as a sign of attempted deception or low credibility.

    Autistic Behavior and the Justice System

    Recently, there has been growing research interest in how individuals from certain groups might be particularly vulnerable to biased or negative evaluations should they find themselves thrust into interactions with the justice system. One such focus has been the exploration of how autistic individuals might experience injustices as a consequence of evaluations being influenced by extra-evidential factors.

    Just like non-autistic individuals, there is considerable variability in the behavioral characteristics of autistic adults. However, observational research has highlighted that in social contexts, autistic individuals may be more likely to display characteristics such as gaze avoidance, repetitive body movements, flat or contextually unusual emotion expression, and difficulties with interpreting figurative language. Any of these behaviors might, in certain contexts, be interpreted as challenging the individual’s credibility or perhaps even as a sign that the person is trying to deceive.

    One approach researchers have used to investigate these possibilities is to create and record mock investigative interviews or trials and to use these as stimulus materials to examine how other people react to certain types of behaviors exhibited by the parties involved. Although some criticize the ecological validity of such approaches, they can provide an excellent and meaningful guide—see, for example, Kerr and Bray (2005)—regarding when and how biases or injustices may arise.

    In a recent study using a mock-police interview situation (Logos et al., 2021) in which individuals displayed nonverbal behaviors such as gaze aversion and fidgeting when being questioned about their possible involvement in a crime, those behaviors were relied upon by observers when evaluating the suspect’s guilt—even when compelling evidence decisively indicated the suspect’s innocence. Pleasingly, the research also demonstrated that if observers were informed that the suspect had a diagnosis of autism spectrum disorder (using a simple autism alert card) the biased evaluations were significantly reduced.

    Another study (Lim et al., 2021) found that autistic individuals who were interviewed about the possibility that they may have taken money from an envelope in the research lab were, when telling the truth, judged by observers who watched their interviews as more deceptive and less credible than non-autistic individuals. Perhaps surprisingly, this study did not demonstrate a link between those judgments and systematic behavioral recordings of behavioral characteristics such as gaze aversion, body movements, flat affect, literal interpretation of figurative language, and poor social reciprocity that are considered likely to cue inferences of deception.

    Why this was the case is unknown—perhaps the fact that the sample was sourced from a research database whose members were familiar with (and more relaxed in research studies) contributed, or perhaps some compensatory mechanisms by the participants were at play. Nevertheless, there were features of the individuals’ overall presentation that influenced the judgments of deception—in other words, something clearly misled observers and contributed to erroneous negative evaluations of credibility.

    Research in this area is in its infancy and there are so many unanswered questions. The findings to date should, however, alert us to the possibility that autistic individuals who are simply being themselves may be vulnerable to biased judgments in forensic (and other settings). Moreover, the same may apply to other individuals who when placed in such contexts display behaviors that might be considered non-normative.

Facebook Comments