Research Sources
The research this curriculum is built on.
Every technique taught in Division 6 is drawn from peer-reviewed academic research or from practitioner methodology with a documented, verifiable record. The sources below are the primary references for the curriculum. Where the evidence is mixed or limited, the curriculum says so.
This page is the canonical home for research attribution. The researcher list in How It Works links here for full source detail.
Important context
Behavioral indicators are probabilistic, not diagnostic. The techniques taught in this program are educational tools for developing observational skill, not instruments for making determinations about specific individuals. No behavioral pattern taught here is presented as conclusive, and no cluster of patterns is sufficient on its own to establish truth or deception. The curriculum is explicit about the conditions under which each technique holds and the conditions under which it breaks down.
Academic research
Primary studies and meta-analyses.
These are the studies that most directly inform the curriculum structure, the reliability ratings assigned to each indicator, and the explicit caveats built into the training.
Accuracy of deception judgments.
Bond, C. F., Jr., & DePaulo, B. M.
Personality and Social Psychology Review, 10(3), 214–234.
Meta-analysis of 206 studies and over 24,000 observers. Humans detect deception at an average rate of 54 percent, barely above the 50 percent chance baseline. Training on behavioral cues alone produces minimal improvement. This finding is the empirical starting point for the curriculum: intuition is not a reliable instrument.
Cues to deception.
DePaulo, B. M., Lindsay, J. J., Malone, B. E., Muhlenbruck, L., Charlton, K., & Cooper, H.
Psychological Bulletin, 129(1), 74–118.
The most comprehensive meta-analysis of behavioral cues to deception, reviewing 158 independent samples. Fewer cues are reliably associated with deception than widely believed, and effect sizes are modest. Verbal cues (less detail, more ambivalence, more distancing language) outperform nonverbal cues as a class. No single cue is diagnostic.
Detecting deception by manipulating cognitive load.
Vrij, A., Fisher, R., Mann, S., & Leal, S.
Trends in Cognitive Sciences, 10(4), 141–142.
Deception requires more cognitive resources than honest responding because liars must suppress the truth, construct a plausible alternative, monitor their own behavior, and anticipate detection. Increasing cognitive load during questioning degrades liars' accounts more than honest accounts, producing measurable differences in detail, consistency, and spontaneous corrections.
Detecting Lies and Deceit: Pitfalls and Opportunities (2nd ed.).
Vrij, A.
Wiley.
The most comprehensive practitioner-oriented synthesis of deception research. Covers verbal cues, nonverbal cues, physiological measures, structured elicitation methods, and the documented failure modes of intuition-based approaches. Vrij's analysis of which indicators hold in controlled studies versus which are folklore is the primary reference for indicator reliability ratings in Division 6.
Strategic use of evidence during investigative interviews.
Hartwig, M., Granhag, P. A., & Luke, T.
In R. Bull (Ed.), Investigative Interviewing (pp. 3–36). Springer.
Strategic Use of Evidence (SUE) is a research-backed interviewing technique that delays disclosure of available evidence, prompts the interviewee to commit to a statement first, then introduces contradicting evidence. The technique consistently outperforms conventional confrontational approaches in producing verifiable admissions without coercion.
Psychological aspects of natural language use: Our words, our selves.
Pennebaker, J. W., Mehl, M. R., & Niederhoffer, K. G.
Annual Review of Psychology, 54, 547–577.
Function words, particularly first-person singular pronouns, reveal psychological distance, ownership of events, and cognitive engagement in ways that content words do not. Low first-person singular use in a narrative about oneself is associated with psychological distancing from the described events. This holds across written and spoken language at scale.
Nonverbal leakage and clues to deception.
Ekman, P., & Friesen, W. V.
Psychiatry, 32(1), 88–106.
The original theoretical framework distinguishing deliberate expressive behavior from microexpressions and nonverbal leakage: brief, involuntary facial movements that occur before conscious suppression. Later cross-cultural work established that the basic emotional display patterns are universal, while display rules governing their expression are culturally variable.
Criteria-Based Content Analysis.
Steller, M., & Koehnken, G.
In D. C. Raskin (Ed.), Psychological Methods in Criminal Investigation and Evidence (pp. 217–245). Springer.
Criteria-Based Content Analysis (CBCA) is a structured method for evaluating the credibility of statements based on 19 content criteria derived from Undeutsch's hypothesis: that accounts of genuine experiences differ in specific, identifiable ways from fabricated ones. Genuine accounts tend to contain more contextual detail, more spontaneous correction, more accounts of peripheral detail, and more self-deprecating information.
Source monitoring.
Johnson, M. K., Hashtroudi, S., & Lindsay, D. S.
Psychological Bulletin, 114(1), 3–28.
Reality Monitoring is a framework for distinguishing memories of perceived events from memories of imagined events. Perceived-event memories tend to be richer in sensory and contextual detail, while imagined memories tend to contain more cognitive operations (reasoning, planning). The framework has been adapted for statement analysis to assess the experiential quality of verbal accounts.
National Evaluation of the PEACE Investigative Interviewing Course.
Clarke, C., & Milne, R.
Police Research Award Scheme, Report PRABS/149. Home Office.
PEACE (Preparation and Planning, Engage and Explain, Account, Closure, Evaluation) is the structured interviewing model developed for UK policing as a non-coercive alternative to confession-seeking interrogation. It emphasizes open questioning, active listening, and cognitive interview techniques. Evaluation data show improved information yield without the false confession risks documented in adversarial approaches.
Practitioner sources
Applied methodology with a documented record.
In addition to academic sources, the curriculum draws on practitioner frameworks that have been used in professional and forensic settings. Where practitioner frameworks conflict with the academic evidence, the academic evidence takes precedence and the conflict is noted in the curriculum.
Avinoam Sapir
Laboratory for Scientific Interrogation
Developed Statement Analysis as a practitioner methodology for identifying the linguistic markers that distinguish constructed narratives from genuine accounts, focusing on absence, minimization, and internal inconsistency. Treated in the curriculum as a navigational tool for directing further questioning, not a diagnostic instrument.
Mark McClish
Former U.S. Deputy Marshal
Applied statement analysis focused on the linguistic markers distinguishing genuine denial from constructed deflection. Primary practitioner source for several Foundation tier indicators. All SCAN-derived indicators are presented with explicit caveats about base rates and the risk of false positives.
Chris Voss
Former FBI Lead Kidnapping Negotiator
Applied negotiation framework from the FBI hostage unit. Source for calibrated questioning, tactical empathy, and presumptive framing techniques used in the Operator Series. Voss's practitioner framework is used alongside the academic evidence on open questioning.
Elizabeth Loftus
UC Irvine
Memory malleability and false memory formation research. Establishes that false memories form with the same subjective confidence as genuine ones, which informs how the curriculum distinguishes between a fabricated account and a confabulated one, and why genuine recall of peripheral detail is diagnostically useful.
What the evidence actually shows
Honest about the limits.
Gaze aversion is not a reliable indicator
The belief that liars avoid eye contact is not supported by the research record. DePaulo et al. (2003) found no significant association between gaze aversion and deception across 158 studies. Division 6 does not teach gaze aversion as a deception cue.
Humans detect deception at 54%
Bond and DePaulo's 2006 meta-analysis established 54% as the human baseline, barely above chance. The curriculum is built explicitly on this finding: it exists because intuition is insufficient, not to amplify it.
No single cue is diagnostic
DePaulo et al. (2003) found that even the most reliable behavioral cues produce small effect sizes. The curriculum requires cluster analysis across multiple indicators. No single behavior justifies a deception conclusion.
Verbal cues outperform nonverbal
Across controlled studies, verbal indicators consistently outperform nonverbal ones as a class. This is why the Statement Analysis tier is the Foundation curriculum and nonverbal content is explicitly framed as supplementary.
The techniques taught here are educational tools for personal and professional development. They are not diagnostic instruments and should not be used as the sole basis for decisions about other people. Each indicator in the Response Playbook carries an explicit reliability rating and a documented failure-mode description.
Where the research is genuinely mixed or where practitioner claims exceed what controlled studies support, the curriculum says so. Overpromising what behavioral analysis can deliver is itself a form of deception, and Division 6 does not do it.
Train on the research. Not the folklore.
Seven days of full access to the Behavioral Lab, the Response Playbook, and your first classified case. No card required.
No credit card required. Cancel any time.