Guides
Inoculation theory persuasion: how to read a prebunk's resistance gap
Inoculation theory persuasion is measured by the resistance gap. This article explains the metric, its limits, and when to act on prebunk test results.
What to take away
- The metric is the resistance gapthe percentage point difference in agreement with one named false claim between a prebunked group and a comparable control group.
- Read it at two weeks, not at the end of the session. The immediate reading shows the dose landed; the two-week reading shows resistance held.
- Act when the two-week gap reaches 8 points and less than a third of the effect has decayed. Below that, rebuild the prebunk instead of repeating it.
- The gap cannot show whether people met the claim in the wild, whether they acted differently, or whether the prebunk was honest.
The resistance gap, defined
A prebunk warns an audience about a misleading argument and offers a weak version of it to argue with, before the real claim arrives.
Resistance Gap Readings
Immediate gap
- Timing
- 24-48 hours
- Purpose
- Dose landed
- Threshold
- —
- Claim wording
- Identical
Two-week gap
- Timing
- 14 days
- Purpose
- Act on this
- Threshold
- —
- Claim wording
- Identical
Decay ratio
- Timing
- Two-week ÷ immediate
- Purpose
- Persistence check
- Threshold
- 0.67 or higher
- Claim wording
- Same false claim
Inoculation theory persuasion treats that dose the way immunology treats a vaccine: small, controlled exposure builds resistance to a larger one. The mechanism overview traces the idea from early social psychology work to current media literacy programs.
The metric that follows is the resistance gap. Define it as the percentage point difference in agreement with one specific false claim between an inoculated group and a comparable control group.
A clean number invites overreach, and the social proof mistakes file shows how quickly one figure becomes a claim about everyone.
Example Gap Calculation
12-point gap
Control group agreement: 34%
Prebunked group agreement: 22%
Difference: 12 percentage points
| Component | Definition | Reading that matters |
|---|---|---|
| Immediate gap | Difference at 24 to 48 hours | Confirms the dose landed |
| Two-week gap | Difference at 14 days | The number you act on |
| Decay ratio | Two-week gap divided by immediate gap | 0.67 or higher |
| Claim wording | Whether both readings use the same false claim | Must be identical |
How to read the gap
A single gap figure means little on its own.
Read it against the control group you actually ran, not against a number remembered from another campaign. The prebunking literature shows effects that move with topic, delivery and audience, so the only comparison that counts is the one inside your own study.
Illustrative ranges, not guarantees: a prebunk on a familiar claim often produces an immediate gap of 10 to 20 points, a two-week gap of 8 to 15 points, and a decay ratio near 0.7. An unfamiliar technical claim usually produces less.
The two-week gap is the reading. Everything before it describes the dose, not the resistance.
What the metric cannot show
The gap cannot show whether anyone met the claim outside the test. It measures stated agreement with one survey item, at one moment, under conditions you chose.
It cannot separate a fair prebunk from an unfair one either. A message that discredits a true claim produces a clean gap just as easily as one that discredits a false one.
Ethics sits outside the number, which is why informational influence ethics treats consent and accuracy as questions that measurement does not answer.
Attribution and its limits
Real campaigns have no clean control group. Public health messaging from the CDC shares the air with news coverage, clinics and peer conversation. A prebunk timed to a Washington, D.C. debate cycle competes with the debate itself.
Marketing teams run a version of the same test when they pre-empt a rival's comparison claim, which makes inoculation theory marketing an applied case of this design.
Attribution becomes a judgment call. You compare a matched audience that likely saw the prebunk with one that likely did not, then report the difference as an estimate rather than a fact.
A prebunk that misstates evidence to discredit a rival can also cross into deception, and the FTC holds statutory authority over unfair or deceptive acts under 15 U.S.C. 45.
When to stop measuring and decide
Three readings are enough: before launch, at two weeks, at ninety days.
Stop when the ninety-day gap sits inside your margin of error of the two-week gap, or when two consecutive readings move fewer than 3 points. Then decide once: keep, rebuild or retire.
A prebunk that holds for ninety days usually earned it through repetition and public commitment, and the commitment and consistency file explains why that pattern outlasts clever wording.
