What Is Statement Validity Assessment?

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

Statement validity assessment (SVA) is a structured procedure for evaluating whether a verbal statement is based on a real experience or has been fabricated. It was developed in German-speaking countries, primarily for assessing children's testimony in sexual abuse cases, and it remains the most widely used verbal credibility assessment method in forensic settings worldwide. SVA has three components: a structured interview, criteria-based content analysis (CBCA) with 19 content criteria, and a validity checklist. Meta-analyses show that CBCA criteria can distinguish truthful from fabricated accounts at rates above chance, but accuracy is not high enough for SVA to serve as a definitive test of truthfulness. It is an investigative tool, not a lie detector.

Key takeaways

SVA is a three-part forensic procedure: structured interview, criteria-based content analysis (CBCA) and a validity checklist.
CBCA uses 19 criteria to assess whether a statement's content is consistent with a real experience, based on the Undeutsch hypothesis.
A 2016 meta-analysis of 55 studies found a large overall effect size (g = 0.98), but accuracy at the individual level is not certain enough for court verdicts.
SVA was designed for children's testimony in abuse cases and its generalisability to other populations and contexts is limited.
No verbal credibility method can reliably prove that a specific person is lying or telling the truth.

What is the Undeutsch hypothesis?

SVA rests on what is known as the Undeutsch hypothesis, named after German psychologist Udo Undeutsch. The hypothesis holds that statements derived from genuine personal experience differ in content, quality and expression from statements that are invented or coached. In other words, a real account of something that happened should have distinctive characteristics that a fabricated account lacks.

This idea is intuitive: someone who actually witnessed or experienced an event should be able to describe it with a richness and specificity that is difficult to fake. The challenge has been to turn this intuition into a reliable, structured assessment method. SVA is the most developed attempt to do so.

What are the three components of SVA?

The structured interview

The process begins with a carefully conducted interview, following established protocols for open-ended questioning. The interviewer encourages the person to provide a free narrative without leading questions or suggestive prompts. The quality of the interview matters, because a poorly conducted interview can suppress the very details that CBCA is designed to detect.

Criteria-based content analysis (CBCA)

CBCA is the core analytical step. It applies 19 criteria, grouped into five categories, to the content of the statement:

General characteristics (criteria 1-3): logical structure, unstructured production, and quantity of details. A genuine account tends to contain many details and may not follow a strict chronological order, because real memories are often recalled in fragments.

Specific contents (criteria 4-8): contextual embedding, descriptions of interactions, reproduction of conversation, unexpected complications, and unusual details. These reflect the kind of information that is hard to fabricate convincingly.

Peculiarities of content (criteria 9-13): superfluous details, accurately reported details not understood, related external associations, accounts of subjective mental state, and attribution of the perpetrator's mental state. These capture content that a child, in particular, would be unlikely to include in a coached or invented account.

Motivation-related contents (criteria 14-18): spontaneous corrections, admitting lack of memory, raising doubts about one's own testimony, self-deprecation, and pardoning the perpetrator. A person who freely admits gaps or weaknesses in their account may be more likely to be telling the truth, the reasoning goes, because a fabricator would try to present a seamless story.

Offence-specific elements (criterion 19): details that are consistent with known characteristics of the type of offence described.

A trained assessor reads the statement and rates the presence of each criterion. More criteria present generally suggest a statement based on genuine experience.

The validity checklist

The third component places the CBCA results in context. It considers factors such as the person's cognitive and linguistic abilities, whether the statement could have been influenced by suggestion or coaching, and whether external evidence supports or contradicts the account. This step is meant to prevent over-reliance on CBCA scores alone.

What does the meta-analytic evidence show?

A 2016 meta-analysis by Oberlader and colleagues in Law and Human Behavior examined 55 studies testing CBCA and reality monitoring. The overall effect size was large (g = 0.98), meaning that on average, truthful statements scored substantially higher on content criteria than fabricated ones. Applying all 19 CBCA criteria outperformed any incomplete set, and classification rates were higher when researchers used discriminant function analysis rather than simple sum scores.

An earlier meta-analytic review by Amado, Arce and Farina confirmed the Undeutsch hypothesis at the group level and found that most CBCA criteria discriminated between experienced and fabricated accounts. However, some criteria, including self-deprecation and pardoning the offender, did not reliably discriminate.

Vrij's 2005 qualitative review of the first 37 CBCA studies concluded that SVA evaluations were "not accurate enough to be admitted as expert scientific evidence in criminal courts" but could be useful as an investigative tool.

What are the limits of SVA?

It was designed for a specific context. SVA was developed to assess children's statements about sexual abuse. Extending it to adult witnesses, suspects, or entirely different types of cases introduces uncertainty. The criteria that work well for a child's account of abuse may not generalise to an adult's alibi or an insurance claim.

Subjectivity in coding. Although CBCA provides structured criteria, applying them requires judgment. Different assessors can reach different conclusions about the same statement. Meta-analyses have found acceptable inter-rater reliability for most criteria, but "acceptable" is not the same as "identical."

Verbal ability affects scores. A more articulate, fluent speaker will naturally produce a statement with more detail, more contextual embedding and more conversational reproduction. This means that the amount of detail in a statement is influenced by factors other than truthfulness, such as age, education, language skills and the quality of the interview. The validity checklist is meant to account for this, but it adds another layer of subjective judgment.

It cannot identify deception with certainty. Even with a large meta-analytic effect size, the distributions of truthful and fabricated statements overlap. Some fabricated statements score high on CBCA criteria, and some truthful ones score low. A high CBCA score makes it more likely that a statement is based on experience, but it does not prove it.

The validity checklist is underresearched. Most studies have focused on CBCA alone. The full SVA procedure, including the structured interview and validity checklist, has received less systematic empirical testing.

How does SVA compare with other verbal methods?

SVA, reality monitoring and the verifiability approach all analyse verbal content to assess credibility, but they focus on different features. CBCA looks for qualitative markers of experienced events. Reality monitoring checks for sensory and perceptual richness. The verifiability approach asks whether details could be independently checked.

The 2016 Oberlader meta-analysis found no significant difference in the effectiveness of CBCA and reality monitoring. The verifiability approach is newer and has been tested in fewer studies, but it offers a complementary angle. Researchers including Vrij have suggested that combining multiple verbal analysis methods may produce better results than using any one alone. For the broader question of whether any single cue reliably indicates deception, the answer from the research is no.

How does Murder Book connect to statement analysis?

Murder Book: The Art of Interrogation is an upcoming narrative interrogation game from Gungrounds, an independent studio in Zagreb, Croatia. Its development screenshots show an in-game casebook with entries such as Fabrication, Stonewalling and Narrative Gaps. Players examine suspects' accounts for inconsistencies, missing details and signs that a story may have been constructed rather than remembered.

This mirrors the logic behind SVA: a statement built from real experience tends to contain unexpected details, corrections and admissions that are hard to fake. In Murder Book, a suspect whose account is too smooth, too consistent, or too carefully structured may be worth questioning further. But as with SVA in the real world, the presence or absence of any one feature is a prompt for investigation, not proof of guilt.

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