To act truthfully: Nonverbal behaviour and strategies during a police interrogation

Abstract In an experiment, nonverbal indicators of deception in police interrogations of mock crimes were examined. Both vocal and nonvocal behaviours were scored. Thirty participants were subjected to long interrogations (over 9 minutes) conducted by 30 experienced police officers, asked to interrogate as they normally do. Although the liars reported being significantly more nervous, and found the task more strenuous than the truth tellers, no differences in the nonverbal behaviours scored were found. In an analysis of the strategies employed, both truth tellers and liars were found to try to not make excess movements. The principal verbal content strategy for the liars was to keep the story simple, and for truth tellers to keep the story real. The reasons why, in this demanding situation, the truth tellers and liars could not be distinguished by their nonverbal behaviour are discussed.

Laboratory analog study using mock crime paradigm with 30 participants interrogated by 30 experienced criminal investigators (minimum 5 years experience) in unscripted interviews exceeding 9 minutes, with comprehensive coding of vocal and nonvocal behaviors and post-interview strategy reports.

This study addressed a critical gap in deception detection research by creating a high-ecological-validity interrogation environment that better approximated real police practice, moving beyond brief laboratory scenarios with non-expert interrogators. Thirty participants were subjected to extended unscripted interrogations by experienced criminal investigators about mock crimes, with genuine financial motivation to succeed at deception. Although liars reported significantly greater nervousness and stress than truth-tellers, no observable differences in nonverbal behavioral displays were detected between the two groups—a striking null result. Strategy analysis revealed that both groups commonly suppressed excess movements and maintained eye contact, with liars strategically simplifying their narratives while truth-tellers focused on accurate representation. These findings fundamentally challenge the validity of nonverbal behavior as a reliable deception detection tool in realistic interrogation contexts.

This research undermines the evidential basis for law enforcement reliance on nonverbal behavioral cues in deception detection and suggests that interrogators' intuitive behavioral indicators may be unreliable predictors of deception. The findings support a shift toward verbal content analysis and strategic questioning techniques while cautioning against over-interpretation of nonverbal displays in actual criminal investigations.

Background & Context

This 2006 study addresses a critical limitation in deception detection research that had plagued the field for decades. Most prior studies examined short video clips (less than a minute), did not involve actual transgressions, used non-expert interrogators with scripted questions, creating an artificial environment that bore little resemblance to real police interrogations.

The research emerged from the Swedish deception detection laboratory at Gothenburg University, where researchers sought to understand why nonverbal behavioral cues—widely believed by law enforcement to reliably indicate deception—so often failed in actual practice. Prior literature suggested reliable indicators existed, but studies suffered from lack of replicability or low power, and reliance on behavioral indicators over verbal cues often led to erroneous interpretations.

The study's significance lies in its ecological validity: creating a high-stakes, demanding interrogation environment with experienced police officers conducting lengthy, unscripted interviews about mock crimes—conditions that mirror real investigative settings where deception detection truly matters.

Research Design & Methodology

Thirty participants were subjected to long interrogations (over 9 minutes) conducted by 30 experienced police officers, asked to interrogate as they normally do. The participant sample consisted of 21 females and 9 males with ages ranging from 18 to 53 years. All participants received monetary compensation of 300 SEK (about $30) with an additional 200 SEK (about $20) if they could convince the police to believe them, creating genuine motivation.

The police sample included 11 females and 19 males, all with at least 5 years experience as criminal investigators. Participants were assigned to either lying or truth-telling conditions regarding mock transgression scenarios. Both vocal and nonvocal behaviors were scored by trained coders who analyzed the video-recorded interrogations.

Key design elements included:

• Mock crime paradigm involving actual tasks at real locations (convenience stores or bakeries)
• Extended interrogation duration exceeding 9 minutes—significantly longer than typical lab studies
• Experienced investigators conducting naturalistic, unscripted interviews
• High-stakes environment with financial incentives for successful deception
• Comprehensive coding of multiple nonverbal behavioral categories

Results & Key Findings

Although liars reported being significantly more nervous and found the task more strenuous than truth tellers, no differences in the nonverbal behaviours scored were found—a stunning null result that contradicts widespread law enforcement beliefs about behavioral deception cues. Despite experiencing greater internal stress, deceivers successfully controlled their external behavioral displays.

Strategy analysis revealed important patterns:

• The most common nonverbal strategy for both truth tellers (50.0%) and liars (54.5%) was to not make any excess movements
• The second most common was to maintain eye contact (truth tellers 27.8%, liars 27.3%)
• Among principal strategies, the "no excess movements" approach was adopted by 46.7% of truth tellers and 73.3% of liars
• Some participants (truth tellers 16.7%, liars 13.6%) stated they had not used any strategy

The principal verbal content strategy for liars was to keep the story simple, while truth tellers aimed to keep the story real. Liars strategically kept their stories simple, while truth tellers recounted their narrative with an accurate representation of actual events.

Discussion & Significance

This research fundamentally challenges the validity of nonverbal behavior as a reliable deception detection tool in realistic interrogation contexts. The findings suggest that when interrogations are lengthy, high-stakes, and conducted by professionals, both liars and truth tellers become acutely aware of behavioral expectations and actively manage their demeanor. Both truth tellers and liars tried to not make excess movements, and the reasons why truth tellers and liars could not be distinguished by their nonverbal behaviour in this demanding situation were discussed.

The study reveals that suspects—both truthful and deceptive—hold similar beliefs about what behaviors signal deception, leading them to adopt nearly identical impression management strategies. This convergence eliminates behavioral diagnosticity. The greater internal stress experienced by liars did not "leak" through nonverbal channels as traditional theories predicted, suggesting sophisticated self-monitoring capabilities under interrogation pressure.

The finding that liars tend to hold back more details during free recall, especially when unaware of evidence against them, and strategically keep stories simple has profound implications for interview technique development, ultimately contributing to evidence-based approaches like the Strategic Use of Evidence (SUE) technique that focus on verbal content rather than behavioral cues.

Limitations & Considerations

Despite its ecological improvements, this remained a laboratory analog study with mock crimes rather than actual criminal investigations. The stakes, while heightened through monetary incentives, cannot fully replicate the consequences faced by genuine suspects. The 9-minute interrogation duration, though longer than typical lab studies, is still relatively brief compared to many real-world interrogations.

The study examined a limited range of nonverbal behaviors and may have missed subtle cues that more sophisticated coding systems might detect. Cultural context is also important—all participants and officers were Swedish, limiting generalizability to other cultural contexts where nonverbal norms differ. The relatively small sample size (n=30) provides adequate power for detecting large effects but may miss smaller behavioral differences.

Practical Applications

This research delivers a crucial message to law enforcement: relying on nonverbal behavior for deception detection in interrogations is scientifically unfounded and potentially dangerous. Officers should shift focus from "reading body language" to verbal content analysis, inconsistencies between statements and evidence, and strategic questioning techniques. Training programs emphasizing nonverbal cues as deception indicators lack empirical support and may lead to investigative errors.

The findings validate interview approaches that emphasize information gathering over behavioral observation. Investigators should recognize that both guilty and innocent suspects consciously control their nonverbal behavior during interrogations, rendering such cues unreliable. Instead, practitioners should adopt evidence-based techniques that elicit diagnostic verbal cues through strategic questioning and evidence disclosure—methods that emerged from this line of research and have demonstrated superior accuracy in subsequent validation studies.