Intensity Ladder Trial¶
Bounded assessment — instantiates Dose–Response Calibration
Climbs a predeclared ladder of intensity rungs from the bottom, stopping at the first rung that reliably produces the wanted effect.
Intensity Ladder Trial is a bottom-up assessment: it fixes a discrete ladder of intensity levels in advance, starts at the lowest rung, and steps up one rung at a time only until the effect it is looking for reliably appears. Its defining move is to stop climbing at the first rung that works — it is a search for the least sufficient level, not a mapping of the whole response range. Because the rungs and the ceiling are declared before any testing begins, the trial can never wander into open-ended escalation; it either reaches a rung that clears the target or exhausts a predeclared ladder and reports failure. Where a curve-fitting pilot wants data at every level, the ladder wants exactly one answer: the lowest rung that is enough.
Example¶
An online coding-practice platform is choosing how hard to make the first exercise a new learner sees. Too easy and nobody learns anything; too hard and beginners quit. The team sets a predeclared ladder of five difficulty tiers, from "fill in one missing line" up to "write the function from a spec," and fixes the ceiling at tier five — no learner in the trial is ever shown anything harder. New users are stepped up the ladder: start at tier one, and a learner only advances a rung once they clear the current one within a couple of attempts. The trial watches for the first tier at which learners are both challenged (they don't solve it instantly) and successful (they don't bail). That turns out to be tier two for most beginners — the minimum effective challenge. Tiers three through five would also "work" in the sense of being solvable eventually, but the trial stops at two because it is the lowest rung that produces engaged learning, and a harder start would only shed frustrated users.
How it works¶
Everything hinges on committing to the rungs and the ceiling before testing. The ladder is a short, ordered set of discrete levels with a hard top; the trial always begins at the bottom and advances one rung at a time, checking at each rung whether the target effect has been met. The distinctive logic is the stopping condition: it halts at the first success rather than continuing to gather the full curve, which means it spends the least intensity and the least exposure needed to answer "how low can we go and still work?" If the top rung is reached without success, that is itself a result — the effect is not achievable within safe bounds and a different mechanism is needed.
Tuning parameters¶
- Rung spacing — how far apart consecutive levels sit. Coarse rungs reach an answer in few steps but overshoot the true minimum; fine rungs pinpoint it but take many steps.
- Advancement criterion — what counts as "cleared" at a rung before stepping up. A strict criterion avoids false advancement but slows the climb; a loose one is fast but may promote on noise.
- Ceiling — the predeclared top rung. A higher ceiling can find effects that need real intensity but widens the exposure the trial permits.
- Confirmation before stopping — whether one success ends the climb or several are required. Requiring repeats guards against declaring a fluke rung effective, at the cost of extra trials.
- Start rung — whether to begin at the true floor or a plausible low guess. Starting higher saves steps but risks stepping past the minimum before ever testing it.
When it helps, and when it misleads¶
Its strength is safety and economy of intensity: by declaring the rungs and ceiling up front and climbing from the bottom, it makes reckless escalation structurally impossible and returns the smallest level that does the job — valuable exactly when higher intensity carries cost, side effects, or dropout. It suits an optimal-challenge setting, where the goal is the productive edge just beyond current ability — the zone of proximal development[n1] — rather than the maximum a learner can survive.
Its failure mode is coarseness and premature stopping. If the rungs are spaced too widely, the reported minimum is really "the lowest tested rung above the true minimum," and a single lucky success can halt the climb one rung too early. The classic misuse is treating the ladder's stop rung as the best level rather than the least sufficient one — the two coincide only when higher intensity has no upside. The discipline is to require confirmation before stopping and to remember the ladder answers "how little is enough," not "what is optimal."
How it implements the components¶
input_intensity— the predeclared, ordered rungs are the intensity dimension, tested discretely from the bottom up.safe_exploration_bounds— the fixed floor and ceiling, committed before testing, are what keep the climb from becoming open-ended escalation.minimum_effective_input— its whole output: the lowest rung that reliably produces the target effect.
It does not fit a continuous response curve (calibration_curve) or its confidence band (uncertainty_band) — that is Stimulus–Response Pilot; and it does not locate the operating band around the plateau (target_range), which Staffing Level Experiment supplies.
Related¶
- Instantiates: Dose–Response Calibration — the ladder is the archetype's minimum-finding instance: it climbs bounded rungs to the least sufficient level.
- Sibling mechanisms: Stimulus–Response Pilot · Advertising Spend Calibration · Policy Intensity Pilot · Staffing Level Experiment · Training Load Calibration · Alert Threshold Tuning · Medication Dose Calibration
Editorial Notes¶
Form Classification¶
Form family: Experiment, Test & Rehearsal
Rationale: Intensity Ladder Trial operates as a bounded trial, probe, simulation, or rehearsal that generates evidence from performance because it climbs a predeclared ladder of intensity rungs from the bottom, stopping at the first rung that reliably produces the wanted effect
Independent corroboration: The frozen evidence defines Intensity Ladder Trial as 'Climbs a predeclared ladder of intensity rungs from the bottom, stopping at the first rung that reliably produces the wanted effect', so its operative form is Experiment, Test & Rehearsal.
Review outcome: Independent reviewer agreement; high confidence.
Origin Attribution¶
Primary origin: Statistics & Experimental Design
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: Predeclared dose escalation with stopping criteria is a formal dose-finding trial design. Pharmacology supplies dose-response constraints, while education adapts graduated challenge and mastery to seek the lowest productive intensity.
Related originating lineages:
- Education & Pedagogy — Scaffolding and the zone of proximal development materially shape the search for the lowest challenging but achievable level.
- Medicine & Healthcare — Dose escalation and minimum-effective-dose trials materially supply monitored rung advancement.
- Pharmacology & Toxicology — Dose-escalation and minimum-effective-dose practice supply the bounded ladder and stop-at-first-effect structure.
Review resolution: Predeclared dose escalation with stopping criteria is a formal dose-finding trial design. Pharmacology supplies dose-response constraints, while education adapts graduated challenge and mastery to seek the lowest productive intensity. The retained alternate domains identify documented formative or independently established origins, not downstream applicability alone. domain_reach=multi_domain because the operating pattern has established use in several fields. The final marks encyclopedia_synthesis=true because the entry deliberately composes those lineages.
Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.
Review outcome: Researched adjudication after independent review; high confidence.
Sources consulted:
- https://www.fda.gov/media/164555/download — FDA guidance describing stepwise dose-escalation and dose-finding trial designs.
Notes¶
[n1] The zone of proximal development, from Lev Vygotsky, is the band of difficulty just beyond what a learner can do unaided but within reach with support — the productive target an intensity ladder climbs toward, and the reason the lowest challenging-yet-achievable rung, not the hardest survivable one, is the goal. ↩