Cognitive Elicitation for Mission-Critical Organizations

What would your best worker do?
Don't wait until it's too late to ask.

MindVail interviews your experts by using an adaptive cognitive elicitation method, maps how they make decisions, and converts that tacit expertise into many uses. One platform for four needs: Knowledge Preservation, Subject Matter Expert Interviews, Requirements Triangulation, and Witness Statement Extraction.

70%
of decisions made are omitted when experts describe complex tasksYates, Feldon & Clark, 2011
0.87
effect size (Hedges’ g) by which training built from cognitive task analysis outperforms conventional trainingTofel-Grehl & Feldon, 2013
47%
how often projects that miss their goals attribute it to inaccurate requirementsPMI, 2014

Sound Familiar?

The gap between a novice and an expert isn't what they know. It's how they decide.

The most valuable data in your organization is in your experts' heads.

Your most experienced people will soon be eligible to retire. New hires take years to reach full proficiency. Your modernization program is building systems around requirements nobody validated with the operators. And every after-action review produces a report that nobody reads twice.

These look like separate problems, with separate budgets, vendors, and owners. They aren't. Each one comes down to the same gap: the judgment your experts use every day has never been captured in a form anyone else can use.

  • Preserve

    Your experts are leaving, and their instincts are leaving with them. Don't let their institutional knowledge follow them out the door.

  • Train

    Your new people take years to make the calls your veterans make in seconds. Unpack your expert's decision making to accelerate knowledge acquisition.

  • Triangulate

    Collecting and synthesizing requirements from multiple stakeholders can lead to losing the plot. Don't create new systems that meet your specifications but don't capture how work is done.

  • Extract

    Scaling to meet the needs of many when a situation or incident arises can lead to delays in understanding when every minute matters. End long waits or shortages of interviewers with digital interviews.

Cognitive Elicitation at the Core

Capture once. Use everywhere.

Organizations often employ solutions that meet specific needs, instead of thinking about solutions that solve multiple problems. A knowledge-management project for retirements, a training contractor for onboarding, a requirements team for modernization, and an analyst cell for debriefs. Each one of those systems requires the information from the same expert but can't share information with any other systems.

MindVail starts with an expert interview. You set the intent and provide context, and the system performs the rest. This cognitive elicitation is captured, mapped, and stored. Your experts give their time once, and their expertise can be valuable in many places.

  1. Elicit

    MindVail first ingests your organization's policies, documents, and SOPs to establish context. From there, the system can identify what is documented and probe for what isn't: the unspoken signals, cues, shortcuts, overrides, and what-if reasoning behind expert decisions. Best of all, dozens of experts can be interviewed in parallel.

  2. Map

    Every decision path, risk threshold, and cue dependency is collected, then structured and visualized with MindVail's Cognitive Decision Mapping. This allows expertise to be updated, queried, and expanded as necessary.

  3. Deliver

    That captured expertise can be used in many outcomes like performance support systems, decision twins, training, requirements, and debrief intelligence.

Traditional expertise capture compared with MindVail
Traditional Expertise CaptureMindVail
Expert timeInterviewed multiple times for various needs and initiativesTacit knowledge is elicited once, then reused across many use cases
Scale1 to 3 experts per studyDozens of experts in parallel
SpeedMonths of interviews, transcription, verification and synthesisMultiple outcomes delivered in days
What you keepStatic reports saved individually in a shared driveA Cognitive Decision Mapping Database, interview exports, and many types of outputs
What it capturesWhat was spoken out loud and written downHow experts deconstruct and decide based on recursive probing cognitive analysis

You've mapped every process. Now map every decision.

MindVail collects human intent, decision making, and expertise.

Most organizations and agencies struggle with WHY they should collect expert knowledge, HOW they can capture expert knowledge, and WHAT to do with it once captured.

M-01Preserve

Keep the knowledge even after the expert leaves.

The situation
When a senior operator with thirty years on the job puts in their retirement package, their replacement inherits the SOPs, but not the instincts. The sound that means a pump is about to fail, the override nobody wrote down, the call they'd make at 2 am when nothing looks right goes with them.
What MindVail does
MindVail captures legacy information while that expert is still around by mapping their decision making logic, cues, overrides, and heuristics.
What you get
  • Searchable databases
  • Legacy expert profiles
  • Cue and heuristic libraries
  • Decision twins
  • Succession profiles

M-02Train

Get new people to expert-level judgment faster.

The situation
Your training teaches procedures, but the reason your experts have succeeded is because they have had the experiences. They know when, and how to follow policy, and in life or death situations; when to deviate from them. That gap is why proficiency takes years instead of months.
What MindVail does
Pinpoints where novice reasoning diverges from expert reasoning and builds training around those friction points. On the job, performance support fills the gap until that judgment is built.
What you get
  • LMS-ready expert guides
  • Novice-vs-expert gap analyses
  • Scenario and simulation seeds
  • Onboarding playbooks
  • Job aids

M-03Triangulate

Many voices, one need. Requirements created in isolation rarely meet the needs of all.

The situation
Your modernization program has requirements, workflows, and a test plan. What it doesn't have is a record of how operators make decisions under pressure. That's why new systems pass acceptance testing and still frustrate the people who use them every day.
What MindVail does
MindVail fits into the systems engineering process you already run. Elicited decision logic comes out as requirements specifications and Jira-ready user stories, so your teams can trace every requirement back to how operators actually decide.
What you get
  • Cognitive requirements specifications
  • Jira-ready user stories
  • System gap analyses
  • Design-decision records
  • The design basis for a performance support system

M-04Extract

Turn debriefs into decisions, not binders.

The situation
After an exercise, incident, or mission, you interview participants one at a time, compile notes for weeks, and still miss where their accounts conflict.
What MindVail does
Runs parallel interviews with every participant, cross-references what they saw and why they acted, and maps where accounts agree and diverge in near real time.
What you get
  • Synthesized debrief reports
  • Consensus and conflict maps
  • Highlight briefs

What You Walk Away With

One elicitation. Outcomes for every team.

Everything MindVail captures lives in one place, so a single round of interviews produces outcomes across the organization, starting with the one that puts expert judgment directly in your people's hands.

Spotlight

Put your best expert beside every operator.

MindVail's outputs (decision maps, cue and heuristic libraries, and decision twins) become a performance support system that represents the tacit expertise captured in your cognitive interviews. It's built from how your own experts decide, not from a generic playbook, so every operator gets that judgment at the moment of need.

  • Ask

    Get an answer to the question every team asks: what would our best expert do here?

  • Guide

    Step through ambiguous situations with the cues and decision points your experts rely on.

  • Check

    Flag the risk thresholds, skipped steps, and warning signs your veterans watch for.

  • Learn

    Reinforce expert reasoning with every use, so performance support and training build on each other.

Spotlight — Decision Twins

A working model of how your experts decide.

A decision twin models how a specific expert, or an entire role, reads cues, weighs trade-offs, and makes the call. Twins can be queried, run against new scenarios, and compared side by side with other experts. They also form the engine behind performance support.

The Full Output Suite

OutcomeWhat It Includes
Performance Support SystemsExpert-driven decision aids, guided troubleshooting, and point-of-need answers built from your experts' tacit knowledge
Decision Twins and ModelsDecision twins of experts and roles; Cognitive Decision Maps; cue, heuristic, and risk-threshold libraries; searchable RAG vector database
Training and WorkforceLMS-ready expert guides; novice-vs-expert gap analyses; scenario and simulation seeds; onboarding playbooks; job aids
Requirements and EngineeringCognitive requirements specifications; Jira-ready user stories; system gap analyses; design-decision records
Debrief and IntelligenceSynthesized debrief reports; consensus and conflict maps; highlight briefs
Talent and SuccessionSuccession profiles; judgment-based hiring and assessment rubrics
Enterprise AI EnablementSOPs that represent how the process is actually completed; elicitation models; secure APIs for LMS, HRIS, and DevOps

The Evidence

What the research says.

  1. 70%

    of decisions made are omitted when experts describe complex tasks

    Research on expertise indicates that experts omit roughly 70% of the decisions they make when describing how they perform complex tasks (Yates, Feldon & Clark, 2011).

    Yates, K. A., Feldon, D. F., & Clark, R. E. (2011, April). Using cognitive task analysis to capture expert knowledge and skills for research and instructional design [Professional development workshop]. Annual Meeting of the American Educational Research Association, New Orleans, LA, United States. (opens in a new tab)

  2. 0.87

    effect size (Hedges’ g) by which training built from cognitive task analysis outperforms conventional training

    A peer-reviewed meta-analysis found that training built from cognitive task analysis outperforms conventional expert-led training, with a large effect size (Hedges’ g = 0.871; Tofel-Grehl & Feldon, 2013).

    Tofel-Grehl, C., & Feldon, D. F. (2013). Cognitive task analysis-based training: A meta-analysis of studies. Journal of Cognitive Engineering and Decision Making, 7(3), 293–304. (opens in a new tab)

  3. 47%

    how often projects that miss their goals attribute it to inaccurate requirements

    When a project misses its original goals, inaccurate requirements management is the primary cause 47% of the time (PMI, 2014).

    Project Management Institute. (2014). Pulse of the Profession in-depth report: Requirements management, a core competency for project and program success. (opens in a new tab)

Compounding Value

Every interview makes the next mission faster and cheaper.

Because all four missions draw from the same place, value builds on itself. Expertise captured to preserve a retiring expert becomes the benchmark trainees learn from. The decision logic elicited for a new system becomes the baseline for its after-action reviews. Every new interview strengthens every program it touches.

Where Decision Advantage Matters Most

Built for any organization where judgment can change the outcome.

  • Defense and National Security

    Close the gap between doctrine and warfighting intuition across operations, exercises, and acquisition.

  • Energy and Critical Infrastructure

    Protect decades of unwritten plant knowledge and safety overrides before generational turnover.

  • Healthcare and Life Sciences

    Capture diagnostic intuition, triage pathways, and procedural judgment.

  • Advanced Manufacturing

    Preserve process-tuning expertise and production-line troubleshooting.

  • Systems Engineering and Technology Transition

    Build digital transformation on how operators actually decide.

  • State and Federal Law Enforcement

    Scale interviewing with ease while collecting real time insights that drive the conversation.

Who we are

MindVail is an MPF Product.

MPF builds AI products for complex work where the stakes are high. Our technology is designed to strengthen human judgment and performance, and people stay in charge of the decisions that matter.

MindVail applies that approach to institutional knowledge. It captures the judgment that experienced experts use every day and turns it into decision models the organization can review, validate, and pass on to the next generation of staff.

Visit MPF (opens in a new tab)

Expertise is not written in the manual. It lives in the cues an expert notices, the options they weigh, and the tradeoffs they make under pressure. When that expert retires or moves on, the judgment leaves with them. MindVail keeps it.

MindVail started with a problem we kept running into. Organizations could document their procedures, but they had little visibility into how their most experienced people actually made decisions. Exit interviews and AI transcripts recorded what experts said. They did not capture how the experts reasoned.

MindVail is built on Cognitive Task Analysis. Trained interviewers use structured protocols to walk experts through real cases. They draw out the cues the expert relied on and the alternatives they set aside, and they produce task diagrams and decision models from the session. The expert reviews and corrects every output before release, so the organization ends up with a validated record of how its best people think.

One company, three products.

Each has its own site. All three are part of the MPF portfolio.

MPFVisit MPF (opens in a new tab)

What connects the three products.

  • Your standards, not ours

    Your experts decide what good judgment looks like. MindVail records their reasoning in their own terms and does not swap in a generic model.

  • Humans in command

    Interviewers lead every session, and experts approve every output. People can review, step in, and override at any point.

  • Evidence you can retrace

    Every decision model links back to the interview segment and the expert statement it came from.

FAQ

Decisions, decoded.

Your experts already have the answers. Make sure your organization keeps them.

In one briefing, we'll map where expertise gaps are costing you across all four missions and show you how elicitation turns your experts' judgment into performance support your whole organization can use.