The science behind
Reflected.

Behavioral interviewing is the most predictive structured technique for assessing how people actually operate under pressure — more predictive than unstructured interviews, and more revealing than questionnaires alone. The core insight: past behavior, described in detail, predicts future behavior better than stated intentions or self-assessments.

The original methodology was developed at AT&T Bell Labs in the 1970s by industrial psychologist Tom Janz, and later systematized by McDaniel, Whetzel, Schmidt and Maurer. Their 1994 meta-analysis of 114 studies established behavioral interviews as significantly more valid than situational or unstructured formats (r = .51 vs .35 for situational, .38 for unstructured).

Reflected uses seven open-ended behavioral prompts designed to surface operating patterns across different pressure contexts: energy, decision-making, conflict with authority, self-perception, unconventional beliefs, failure response, and future orientation. None of these have a "right" answer. They are calibrated to create conditions where people reveal how they actually think — not how they want to be seen.

The Big Five — Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism (we use "Emotional Stability" as the positive pole) — is the most replicated finding in personality psychology. Unlike proprietary frameworks (MBTI, DISC, Hogan), the Big Five is in the public domain, has been validated across cultures and decades, and can be meaningfully inferred from language behavior.

The model emerged from independent work by Tupes & Christal (1961) and was later refined by McCrae & Costa through the 1980s and 1990s. Their NEO-PI-R instrument remains a gold standard. Critically for our approach, research shows that Big Five dimensions can be reliably inferred from natural language — Mairesse et al. (2007) demonstrated meaningful correlation between linguistic markers and personality scores, a finding replicated across many subsequent studies.

Important caveat: Reflected does not administer a validated psychometric instrument. We infer Big Five orientation from behavioral narrative analysis — this produces directional signal, not clinical measurement. Treat the profile as a behavioral fingerprint, not a diagnostic score.

How people tell stories about themselves reveals more than the content of those stories. Psychologist Dan McAdams' narrative identity research (1993–2020) established that personal narratives follow predictable patterns — agency, communion, redemption, contamination — that reflect underlying motivational structures. Leaders who frame past failures as redemptive arcs show meaningfully different risk tolerance and resilience compared to those who frame them as contamination sequences.

We analyze spoken interview responses for structural narrative features: agency language (who acts, who is acted upon), causal reasoning depth, emotional granularity, self-complexity, and counter-evidence acknowledgment. These features are mapped to our 12 behavioral dimensions using a large language model trained to identify behavioral patterns in professional contexts.

The 12 dimensions span decision architecture, authority posture, risk appetite, conflict navigation, emotional regulation, pressure regulation, self-awareness, strategic orientation, achievement drive, learning orientation, negotiation mechanics, and openness to challenge. Each dimension is scored HIGH / MEDIUM / LOW with cited evidence from the interview itself.

The 8 archetypes are derived from cluster analysis of the 12 behavioral dimensions. Each archetype represents a prototypical configuration of strengths across three core dimensions — not a "type" in the Jungian sense, but a pattern of behavioral expression that appears consistently across executives in similar high-dimensional profiles.

Archetypes are not fixed identities. They describe how you most naturally operate given current skill levels — not who you are. A Driver who develops emotional regulation often shifts toward Navigator. An Architect who builds achievement drive becomes a Strategist. The archetype is a snapshot, not a sentence.

We do not claim the archetype framework is validated in the psychometric sense. It is a synthesis tool — a way to make a 12-dimension profile legible to the person being assessed. The underlying dimensions are what carry the predictive signal.

The derailer concept draws on Robert Hogan's foundational insight (Hogan & Hogan, 1997): most leadership failures aren't caused by lack of skill, but by strengths that overshoot under pressure. A leader's greatest asset, amplified by stress, becomes their most dangerous liability.

Reflected does not use Hogan's proprietary scoring system or terminology. We derived our own derailer constructs from behavioral research and independent validation: Volatile (emotional dysregulation under sustained pressure), Perfectionist (standards as bottleneck), Overconfident (authority posture with insufficient feedback integration), Distrustful (adversarial pattern generalization), Approval-Seeking (conflict avoidance as warmth), and Impulsive (speed advantage as decision liability).

Each derailer is scored from the 12-dimension profile. We surface only those that cross a meaningful threshold — most people display 1–2 at most. The goal is not to alarm but to name: a pattern you can see is one you can manage.

Reflected is not a clinical assessment. It is not a validated psychometric instrument. It will not diagnose a personality disorder, predict job performance with actuarial precision, or replace a coach, therapist, or structured 360 feedback process.

What it is: a structured behavioral reflection process that surfaces patterns you probably already sense but haven't articulated. The value is in the articulation — having a language for how you operate creates the possibility of changing how you operate.

The AI analysis is a tool, not an oracle. It processes what you said and identifies patterns. It does not know what you didn't say, what you were consciously avoiding, or what your colleagues would add. Read the report as a starting point for reflection — not a final verdict.