Body language: police recordings, suspicion and measured movement
Body language can influence a judgment of honesty without making that judgment accurate. Police-interview research has found people maintaining similar eye contact during truthful and deceptive statements, while laboratory experiments show gestures changing how suspicious a speaker seems. More elaborate movement measurements have sometimes separated truth from deception under controlled conditions. These findings concern different tasks and different kinds of evidence. None supplies a dependable everyday rule that eyes, hands and halting speech reliably identify a liar.
Samantha Mann, Aldert Vrij and Ray Bull examined real police interviews in their 2002 study, Suspects, Lies, and Videotape. They analyzed recordings of 16 suspects. Crucially, the researchers selected passages whose truthfulness could be checked against other evidence, including reliable witness statements and forensic findings. They did not label a passage deceptive because the speaker avoided eye contact and then count that same avoidance as a discovery. The external evidence supplied the classification before truthful and deceptive behaviors were compared.
The suspects blinked less often and paused longer during deceptive passages. Their eye contact, however, was maintained equally during truthful and deceptive passages. The researchers also reported large differences between individuals. Thus the findings did include behavioral associations; they did not establish that all movement was meaningless. But an average difference in pausing across selected interview clips does not turn every hesitation into a lie. The sample was small, and the observed variation between people is part of the published result.
Body language also changes suspicion through the expectations of the observer. In a 2020 experiment, Josiah King, Jia Loy, Hannah Rohde and Martin Corley used a game involving treasure supposedly hidden behind one of two pictured objects. A recorded speaker named a hiding place. Participants clicked the object they believed concealed the treasure while their eye and mouse movements were recorded. The researchers manipulated the visual gestures accompanying speech, allowing them to examine how those gestures altered a judgment as the utterance unfolded.
Their first experiment focused on movements of the torso. These had only weak effects on the participants' interpretation. Other trials included gestures directed toward the self or nearby objects, such as fidgeting, tapping fingers and scratching the chin. Those gestures appeared more influential, leading the team to conduct a second experiment specifically about them. The distinction within the experiment matters: changing a posture and making a small self-directed movement did not produce identical effects simply because both involved the body.
In the second experiment, participants selected the object named by the speaker on 80.9% of trials without a gesture, compared with 24.2% when a self-directed gesture appeared. Those percentages describe belief in the speaker's statement within the game. They are not accuracy scores for detecting criminals. The recorded movements were experimentally manipulated, and the study investigated the formation of suspicion. A chin scratch could therefore change a choice even though the experiment had not established that chin scratching naturally accompanies deception.
A different approach measured body language with equipment rather than asking observers to interpret familiar gestures. Sophie van der Zee, Ronald Poppe, Paul Taylor and Ross Anderson reported a 2019 experiment using a motion-capture suit that tracked 23 points on the body. The 90 interviewees included 60 people of South Asian culture and 30 White British participants. They gave truthful or deceptive statements about tasks they had experienced, with interviewers from either their own or another culture. Movement was recorded as a combination of joint displacements.
The team reported classification accuracy of 74.4% from overall movement, rising to 82.2% when data from individual limbs were included. These results involved a defined experimental population, known task instructions and instrumented measurements. They were not the success rate of an officer watching hands across a desk. The researchers identified possible interference with natural movement from the suit itself. They also called for research involving higher stakes and more objective measurements of the emotional and mental demands of deception.
The motion-capture findings also complicate a universal claim that deception makes everyone freeze. Greater overall movement accompanied lying in that experiment, whereas other research had linked deception with fewer movements. The team measured a continuous signal across the body rather than assigning a fixed meaning to a single gesture. Its positive result supports further investigation of that measurement method; it does not transfer automatically to a rule about fidgeting visible to the naked eye.
Vrij and Ronald Fisher's 2020 review offers another distinction: deceptive people can feel nervous while controlling the behaviors that observers expect nervousness to produce. Truthful people can also fear being disbelieved. Taken together, the police recordings, the treasure experiment and the motion-capture work show why body language requires more than a confident impression. The recorded facts concern independently checked statements, experimentally induced suspicion or a measured movement signal. Treating those as interchangeable would erase the conditions that made each result possible.
- Samantha Mann, Aldert Vrij and Ray Bull, Suspects, Lies, and Videotape: An Analysis of Authentic High-Stake Liars, Law and Human Behavior, 2002, original abstract — link.springer.com
- Josiah P. J. King, Jia E. Loy, Hannah Rohde and Martin Corley, Interpreting nonverbal cues to deception in real time, PLOS ONE, 2020, full text at PMC — pmc.ncbi.nlm.nih.gov
- Sophie van der Zee, Ronald Poppe, Paul J. Taylor and Ross Anderson, To freeze or not to freeze: A culture-sensitive motion capture approach to detecting deceit, PLOS ONE, 2019 — journals.plos.org
- Aldert Vrij and Ronald P. Fisher, Unraveling the Misconception About Deception and Nervous Behavior, Frontiers in Psychology, 2020, full text at PMC — pmc.ncbi.nlm.nih.gov