The Gold Standard Just Failed Its Own Test

JAMA Network Open Confirms What the Instrument Has Measured Since Day One: The Interview Does Not Produce Consistent Results

The Gold Standard Just Failed Its Own Test

JAMA Network Open Confirms What the Instrument Has Measured Since Day One: The Interview Does Not Produce Consistent Results

Every executive assessment on the market starts from the same place: the interview. The behavioral interview. The diagnostic interview. The structured clinical interview. The personality assessment built on interview data.

On May 28, 2026, the assessment field's own peer-reviewed research confirmed a structural problem with that starting point.

A McMaster University meta-analysis published in JAMA Network Open examined 57 studies across 26 countries involving more than 8,000 adults. The researchers tested the most fundamental property of any measurement tool: does it produce the same result when applied to the same person under similar conditions?

The answer is no. Not consistently.

Standardized diagnostic interviews — the tools the field has treated as the gold standard — demonstrate only moderate test-retest reliability. In some cases, individuals received different diagnoses when assessed days apart. The same person. The same instrument. Different results.

The senior researcher stated it directly: the field should reconsider treating these instruments as a gold standard of assessment.

THE TOOL IS INCONSISTENT. THE SOURCE IS UNRELIABLE. BOTH ARE TRUE SIMULTANEOUSLY.

The McMaster study proved the tool does not produce consistent results. My 10,000-case Monte Carlo simulation — validated using the same statistical methodology employed in aerospace engineering and pharmaceutical drug trials — proved the source does not produce reliable data.

81.4% of individuals under sustained structural load misidentify which domain their primary problem lives in. The error is not random. It is systematic, directional, and type-specific. Individuals displace the problem into domains the performance layer can manage — away from where the structural failure actually resides.

73.0% minimize the depth of the problem. 61.1% are wrong about both domain and depth simultaneously. The confidence intervals at 10,000 cases are tight: all headline findings fall within plus or minus 2.3%.

The most severe structural failures produce the most extreme self-report distortion. The person who most needs accurate assessment is the person whose self-assessment is most structurally wrong. This is not a limitation of the person's honesty or intelligence. It is a structural law: a system operating at or near capacity cannot allocate the processing resources required to accurately assess its own state while simultaneously maintaining the performance layer that conceals the degradation.

THE GAP BETWEEN WHAT THEY REPORT AND WHAT IS ACTUALLY HAPPENING

The PE firm evaluating a founder relies on the behavioral interview. The attorney evaluating a client relies on diagnostic assessment. The board evaluating an executive relies on psychological evaluation tools.

All share one structural dependency: the person's verbal narrative.

JAMA just confirmed the tool designed to capture that narrative does not produce consistent results. The 10,000-case simulation confirmed the person producing that narrative cannot accurately describe the condition the narrative is supposed to represent.

This is the gap. The gap between what they report and what is actually happening. Nobody is measuring it — except the instrument that does not start from the interview at all.

WHAT THE INSTRUMENT READS INSTEAD

The Structural Identity Profiler is a 70,000-line diagnostic engine with four-channel biometric integration: EEG, heart-rate variability, facial affect, voice prosody. It bypasses the executive's conscious narrative entirely.

It does not ask the executive how they are performing. It reads the system directly. Load state. Gate architecture. Structural capacity. Risk trajectory. The divergence between the performance layer and the structural reality underneath it.

The result is a 50-75 page engineering report. It sits alongside the forensic accounting report and the financial audit in the client file. Engagement letter. Documented methodology. Professional liability.

The interview reads the mask. The instrument reads the person.

THE FORENSIC ACCOUNTING PARALLEL

A forensic accountant does not ask the CFO to describe the state of the books. The forensic accountant reads the books directly. The diagnosis is independent of the subject's narrative.

The Structural Identity Profiler operates on the same principle. It does not ask the executive to describe their own structural state. It reads the structural state directly. The assessment is independent of the subject's self-report.

Before forensic accounting existed, financial fraud was detected by asking questions. The tools to read the numbers independently had not been built yet. The tool now exists for the human system. The question is whether you continue to assess the executive by asking them — using an interview JAMA just confirmed does not produce consistent results — or whether you measure the system directly.

WHAT THIS MEANS FOR YOUR EXPOSURE

If you are carrying key person risk — if your investment, your fiduciary obligation, or your organizational stability depends on a human system operating at capacity — the JAMA finding is not academic. It is operational.

The tool you rely on to assess that human system does not produce consistent results. The person you rely on to report their own state cannot do so accurately under the conditions that would make the report matter.

The exposure is the gap between the performance layer and structural reality. The exposure is unquantified until it is independently measured.

The $2,500 Demonstration Read is the entry point. Experience the instrument's precision on yourself. See what it reads that the interview cannot.

dongaconnet.com