Missionloops · a research instrument for DRDC
A scientific instrument for studying human resilience.
A self-selected population works its own high-stakes problems on it, under genuine pressure. The AI teaches method but never reads the work, so it cannot influence the results. DRDC gets data no recruited study group can produce. We build and run the instrument and provide the team of researchers.
The gap
The questions defence most needs answered about human resilience resolve only in real people, under real stakes, over months, not in a lab or a recruited cohort. There has been no instrument that could measure them there. This is that instrument. Below are the defence research priorities it can carry.
DRDC research priority · Combat in the virtual and physical environment
Cognitive-warfare resistance
An instrument to test whether resilience to cognitive warfare can be built in the operator being targeted, and measured against a criterion set in advance.
The defence hypothesis
This threat costs the attacker almost nothing, and it runs outside the domains a military is equipped to fight, so military procurement cannot meet it. The defence is training and research, run on the very civilians the attack targets. Taking part is what defends them: they are participants, not subjects. What the research learns feeds back to the institutions that fight. It is the defence NATO's Chief Scientist report on cognitive warfare recommends: resilience built across the whole of society, around the OODA decision loop the adversary attacks. The report does not say how to build it.
The research it can run
A menu, not a programme. Each study carries its own hypothesis, informative null, and measurement. Recast any toward your own line.
DRDC research priority · Accelerate command, control and intelligence
Cognitive sovereignty
An instrument to test what outsourcing your thinking to AI does to judgment, especially under stress, on the slow decisions where a wrong call stays hidden for months. Missionloops uses AI the other way around: it teaches a strategic doctrine but never thinks for you. The training AI never sees the work, so leaning on it for answers is not even possible. People are paired into teams that keep learning together, their mission files are encrypted, and the pairs grow into trust networks.
The defence hypothesis
The bet is that leaning on an agreeable AI for the thinking is convenient in the moment and corrosive over time: the capacity to reason for yourself atrophies, fastest when stakes and stress are highest. It is the human-AI teaming gap DRDC's Cognition and Trust challenge named, and it is asserted, not measured. Open question: does graduated AI access sharpen or atrophy judgment, and does trained judgment hold when the AI is denied or turned against the operator?
DRDC research priority · Enable safety and security
Civil resilience to overmatch
An instrument to study whether a trained civilian minority becomes a national resilience layer: cohesion that resists social fracturing, and a distributed capability that persists when technology is denied.
The defence hypothesis
The proposal bets a small trained minority defends the wider population, and that the capability survives tech denial, but both are asserted. Open question: does a trained minority produce bystander spillover, durable cross-divide trust, and self-organizing resilience that holds when the platform is gone?
The research it can run
DRDC research priority · Defend North America
Arctic & the Northern build-out
An instrument to study whether trained operators speed the Northern build-out on its civil side. There the bottleneck is rarely money or engineering; it is coordination. The work is two jobs: clear the friction a local authority can resolve, and surface the rule-stuck friction, the kind no one in the field can fix, to the authority that can change it.
The defence hypothesis
Two recent Arctic projects of similar cost split on exactly this line: one cleared on schedule because the authority to clear it sat in the field, the other has stalled for years across five departments. Open question: how much of the Northern friction can a trained network resolve in the field, how much is genuinely rule-stuck, and do seeded communities self-organize the coordination top-down programmes do not produce?
The research it can run
DRDC research priority · People
Personnel: veterans & leadership
An instrument to study the network across the personnel lifecycle. It reveals what a recruiter's hour cannot reach: who has aptitude and who will stay, read from months of real-stakes work. And it produces what the institution is short of: leaders, built by having people guide others, and a network that veterans harden against intrusion from the inside.
The defence hypothesis
The bet is that the network multiplies personnel value the institution cannot otherwise reach: months of real-stakes work signal aptitude and retention better than an interview, guiding others builds leaders, and veterans harden the network from the inside. All of it is asserted. The sharpest test, and the one that does not depend on scale, is the last: does a veteran-integrated network resist cognitive intrusion better than a civilian-only one, and which mechanism does the work?
The research it can run
DRDC research priority · The institution
Innovation as a national multiplier
An instrument to study whether the civilian network produces field innovation the existing programs structurally miss, multiplying the defence spend rather than renting capability from foreign suppliers. A Taleb-style free option: bounded research spend, uncapped innovation upside.
The defence hypothesis
A fifth to a quarter of SR&ED credits flow to consultants rather than research, and the program excludes the gritty field adaptation most innovation needs. Open question: does a self-spreading, trust-organized network surface novel, field-validated doctrine the institution cannot produce top-down?
The research it can run
What makes it an instrument, not an app
Falsifiable by construction
Every question carries an informative null. A null is a finding, not a failure.
Content-blind measurement
The AI never sees the mission, so it cannot shape what it is meant to measure.
Population-scale sample
A real civilian operator pool under genuine stress, a substrate no managed cohort supplies.
Open to your hypotheses
The portfolio is a menu. Recast any question toward your line, or use it for something we did not foresee.
The researchers
Meredith Franklin (measurement) is confirmed prime. Construct seats (strategic reasoning, metacognition, judgment, volition, innovation, anthropology, institutional receptivity) are being recruited as a standby consortium. The full team →
The build
This is an instrument we are building, TRL 4 to 6, in stages; each study comes online as the build reaches its stage. The full build plan is in the proposal.
Discuss the instrument.
Contact us for the full proposal: the study breakdown, the budget, and the build plan. Sent person-to-person, no email gate.
scott (at) missionloops (dot) ca