Detecting Deception In Spontaneous Versus Rehearsed Statements: Direct Deception Judgments for Rehearsed Truths Fall Short While Affective Character Judgments Endure

This study examined how rehearsal affects observers' ability to detect deception in video-recorded statements. Using 147 participants who evaluated spontaneous and rehearsed statements from liars and truth-tellers, researchers found that rehearsal eliminated the ability to distinguish truth from lies through direct veracity judgments, as rehearsed truths appeared as deceptive as lies. However, affective likeability judgments remained diagnostic regardless of rehearsal.

Between-subjects experimental design with 147 participants viewing video recordings in a 2×2 (veracity × presentation type) within-subjects design. Participants made both direct deception judgments and affective character judgments about speakers delivering spontaneous and rehearsed statements.

Background & Context

Deception detection research has consistently documented humans' limited ability to consciously identify liars, with accuracy rates typically hovering around chance levels. Yet this finding presents a puzzle: if we cannot reliably detect lies through direct judgment, how do we navigate the moral complexities of daily social interactions where distinguishing truth from deception is essential?

This 2017 study by van 't Veer, Stel, and van Beest addresses a critical but understudied variable in deception detection—whether statements are delivered spontaneously or after rehearsal. Real-world deception often involves preparation, yet most laboratory research examines only spontaneous lies. This creates an ecological validity gap that may undermine our understanding of how people actually judge veracity in contexts where deceivers have time to craft their stories.

The research bridges two distinct literatures: traditional deception detection focused on conscious veracity judgments, and social cognition research suggesting people form affective impressions—gut feelings about others' character—that may operate independently of explicit lie detection. Understanding how rehearsal affects both channels of judgment has important implications for criminal investigations, security screening, and everyday social interaction.

Research Design & Methodology

The study employed a between-subjects experimental design with 147 participants who watched video recordings of both spontaneous and rehearsed statements from individuals who were either lying or telling the truth. This created a 2 (veracity: truth vs. lie) × 2 (presentation: spontaneous vs. rehearsed) within-subjects factorial design, allowing direct comparison of how rehearsal affects detection accuracy.

Participants served as observers who evaluated video-recorded statements and made two types of judgments. First, they indicated whether they felt the speaker was lying—a direct deception judgment measuring explicit veracity assessment. Second, they rated their overall impressions of the speakers, including affective character judgments such as likeability, capturing more intuitive, impression-based responses.

The key manipulation involved comparing initial spontaneous statements with rehearsed versions of the same content. This within-stimulus comparison controlled for content differences and isolated the specific effect of rehearsal on observer judgments. The design allowed researchers to test whether the cognitive and behavioral polish that comes with rehearsal differentially affects truth-tellers versus liars.

Results & Key Findings

The headline finding revealed a striking asymmetry: during spontaneous statements, liars came across as more deceptive than truth-tellers, but during rehearsed statements this distinction disappeared—rehearsed truths looked like lies. This represents a fundamental challenge to deception detection, suggesting that preparation eliminates observers' ability to discriminate truth from falsehood through direct veracity judgments.

The specific pattern of results demonstrates that rehearsal has opposite effects on how liars and truth-tellers are perceived:

• Spontaneous condition: Observers successfully distinguished liars from truth-tellers, with liars appearing more deceptive
• Rehearsed condition: Truth-tellers were judged as deceptive as liars, indicating rehearsal made honest statements seem less credible
• Affective judgments: Liars were consistently liked less than truth-tellers regardless of whether statements were rehearsed

The dissociation between direct veracity judgments (which failed after rehearsal) and affective character judgments (which remained diagnostic) suggests that impressions of a person's likeability may remain valuable guides in social interaction even when explicit lie detection fails. This finding points to two parallel processing streams in how humans evaluate others' honesty.

Discussion & Significance

These findings have profound implications for understanding real-world deception detection. The results suggest that when individuals have time to prepare their stories—as is common in criminal interrogations, security interviews, and civil litigation—observers lose the ability to distinguish truth from lies through direct judgment. The rehearsal effect appears asymmetric: it makes truth-tellers seem less genuine while leaving liars' deceptiveness unchanged, creating parity in perceived veracity.

The persistence of affective likeability judgments despite the collapse of direct veracity discrimination is theoretically significant. It suggests humans may rely on multiple information channels when evaluating others, with affective impressions potentially operating through different cognitive mechanisms than explicit lie detection. This aligns with dual-process models of social cognition, where intuitive affective responses may capture subtle cues that conscious deliberation misses.

The study challenges the common assumption that practice and rehearsal universally benefit deception. While liars may rehearse to appear more credible, honest individuals who rehearse may inadvertently undermine their own believability. This creates a strategic dilemma for truth-tellers in high-stakes contexts: spontaneity may be more persuasive than polished preparation.

Limitations & Considerations

The study used video-recorded statements in a laboratory context, which differs from live, interactive deception detection scenarios where observers can ask questions and probe inconsistencies. The specific nature of the statements (content, length, stakes) may influence how rehearsal affects credibility, limiting generalizability across contexts. Additionally, the research does not specify what instructions or time participants received for rehearsal, making it difficult to assess whether findings apply to varying levels of preparation.

The sample consisted of observers making passive judgments about strangers in videos—a situation lacking the interpersonal dynamics, relationship history, and strategic interaction that characterize many real-world deception contexts. Professional lie detectors or individuals with specific training might show different patterns. The study also does not clarify whether the rehearsal effect stems from speakers appearing overly polished, losing spontaneous nonverbal behavior, or other mechanisms.

Practical Applications

For forensic interviewers and polygraph examiners, these findings suggest caution when evaluating prepared statements. Suspects who have had time to rehearse their accounts may appear no less credible than honest witnesses who have also prepared, eliminating the diagnostic value of direct credibility assessment. Interviewers might benefit from techniques that circumvent rehearsal, such as asking unexpected questions, requesting reverse-order recall, or focusing on peripheral details that prepared individuals are less likely to have practiced.

The preserved diagnostic value of affective likeability judgments suggests that gut-level impressions about character—independent of specific veracity judgments—may provide complementary information when evaluating rehearsed statements. However, relying on likeability introduces risks of bias based on factors unrelated to honesty. The findings underscore the complexity of deception detection in real-world contexts and the importance of using multiple assessment methods rather than relying solely on subjective credibility judgments.