What Is the Verifiability Approach to Lie Detection?

By Murder Book Editorial TeamPublished by Gungrounds, the studio making Murder BookPublished 3 October 2026Updated 3 October 2026

The verifiability approach (VA) is a verbal lie detection method built on one core idea: truth tellers naturally include more details that an investigator could check, while liars tend to fill their accounts with details that cannot be verified. Developed by Galit Nahari, Aldert Vrij and Ronald Fisher, the approach was first tested in a 2014 study and has since been examined in more than two dozen experiments. A 2021 meta-analysis found a moderate effect, with truth tellers consistently providing more checkable details than liars. The method is not a truth machine. It works best as one input among several, and it does not tell you with certainty whether any single person is lying or telling the truth.

Key takeaways

The verifiability approach classifies details in a statement as verifiable (checkable by investigators) or unverifiable, then compares the proportions.
Truth tellers tend to provide more verifiable details; liars tend to offer more unverifiable ones.
A 2021 meta-analysis reported moderate effect sizes (around d = 0.49 for verifiable detail counts and for the proportion of verifiable details).
The approach works better when the person being interviewed is told that details will be checked.
No verbal analysis method, including the VA, can reliably determine whether a specific individual is lying.

How does the verifiability approach work?

The method starts with a written or spoken statement. An analyst codes each detail as either verifiable or unverifiable. Verifiable details are those an investigator could, in principle, check: named witnesses, specific locations with CCTV, documented transactions, phone records. Unverifiable details are those that only the speaker could know about, such as internal thoughts, uncorroborated actions or vague references to unnamed people.

Two scores are then compared. One is the raw count of verifiable details. The other is the proportion of verifiable details relative to all details. Research consistently finds that truth tellers score higher on both measures than liars.

The reasoning is strategic. Liars face a dilemma: they want to sound convincing, so they add detail, but they avoid details that could be checked and found false. Truth tellers face no such constraint. They mention verifiable things because those things actually happened.

Why does telling people their details will be checked matter?

One of the strongest findings in VA research is that an information protocol improves the method's accuracy. In the original Nahari, Vrij and Fisher study, half the participants were told that the verifiability of their details would be checked. This instruction had a clear effect: truth tellers increased the number of verifiable details they provided, while liars did not.

Later studies replicated this pattern. When people know their account will be scrutinised, truth tellers lean into checkable facts. Liars, who cannot easily invent verifiable details without the risk of being caught, continue to rely on uncheckable claims. The information protocol therefore widens the gap between the two groups, making the method more diagnostic.

What does the meta-analytic evidence say?

A 2021 meta-analysis by Palena, Caso, Vrij and Nahari, published in the Journal of Applied Research in Memory and Cognition, pooled 20 studies with a combined sample of over 1,500 participants. It found a moderate effect size of g = 0.42 for verifiable detail counts and g = 0.49 for the proportion of verifiable details.

The effect was moderated by several factors. Studies that used the information protocol showed larger differences between truth tellers and liars. Event-related statements (describing something that happened) produced clearer differences than opinion-based statements. The medium also mattered: written statements sometimes yielded different results from spoken ones.

A separate meta-analysis in Applied Cognitive Psychology the same year, covering 28 comparisons, reported a similar effect size of d = 0.49 for verifiable details.

These are moderate effects. They mean that, on average, the groups differ, but the distributions overlap. Some liars provide many verifiable details, and some truth tellers provide few. The method is informative at the group level and suggestive at the individual level, but it is not diagnostic on its own.

How does the verifiability approach compare with other verbal methods?

The VA belongs to a family of verbal credibility tools that analyse what someone says rather than how they behave. Others include criteria-based content analysis, which looks for qualitative markers of experienced events, and reality monitoring, which checks whether a statement has the sensory and spatial richness of a real memory.

What sets the VA apart is its focus on checkability rather than narrative quality. It does not ask whether a story sounds vivid or coherent. It asks whether the details in it can be independently confirmed. This makes it, in principle, less dependent on cultural expectations about how a "good" story should sound.

A 2016 meta-analysis by Oberlader and colleagues, published in Law and Human Behavior, compared CBCA and reality monitoring head to head across 55 studies and found a large overall effect (g = 0.98), with no significant difference between the two older methods. The VA's effect sizes are smaller in absolute terms, but the approach is newer and has been tested in fewer contexts so far.

What are the limits of the verifiability approach?

Several cautions apply.

It assumes investigators can actually check. The method classifies details as verifiable based on whether they could be checked. In practice, checking takes time, resources and access to records. A detail classified as verifiable might never actually be verified.

Sophisticated liars may adapt. Early studies suggested the VA is relatively robust against countermeasures, because liars who try to add verifiable details risk being caught if those details turn out to be false. However, a liar who embeds real, checkable facts into a fabricated narrative could, in theory, score well on verifiable detail counts.

Context matters. The VA has been tested in insurance claims, alibi situations and mock crime scenarios. Its performance varies across these settings. It has not been validated as a standalone decision-making tool in real criminal investigations.

Individual accuracy is uncertain. As with all deception research, group-level differences do not translate directly into confident judgments about any one person. An innocent person who gives a vague account may score poorly. A guilty person who builds a lie around real events may score well.

Can the verifiability approach be used in practice?

Researchers have explored applications in police interviews, insurance fraud investigations and asylum claim assessments. The method is designed to complement, not replace, other investigative tools. Nahari and colleagues have emphasised that the VA is most useful when combined with strategic interviewing, where investigators first encourage detailed accounts and then check the verifiable elements against independent evidence.

The approach does not require special equipment or lengthy training in subjective criteria. Its coding rules are relatively straightforward, which may reduce the inter-rater disagreements that affect more subjective methods. However, coding still requires careful attention, and borderline cases (a detail that is partially checkable, for instance) demand judgment.

For a broader look at how asking for precise, checkable details works in investigative interviews, and how unexpected questions can expose differences between truth tellers and liars, see those articles.

How does Murder Book relate to verifiable details?

Murder Book: The Art of Interrogation is an upcoming narrative interrogation game from Gungrounds, an independent studio in Zagreb, Croatia. Its press kit names Fabrication among the behaviours players examine, alongside Avoidance, Narrative Gaps and Emotional Leakage. Development screenshots show a technique menu that includes Confrontation and Baiting, both of which involve pressing a suspect on the specifics of their account.

The verifiability approach offers a useful lens for thinking about what Murder Book asks players to do: pay attention not just to whether a story sounds convincing, but to whether its details could stand up to scrutiny. A suspect who fills their account with vague, uncheckable claims may be worth pressing further, but that pattern alone is not proof of deception. It is a reason to look closer.

Murder Book is in development for PC (Steam) and mobile, with release currently planned for the end of Q2 2027.