A transitional analogy between the wrong model and the correct one reduces the cognitive leap required to correct it

What it is

A falsifying experience gets the user to accept that "something's wrong with my understanding," but there's still a technique for deciding which direction to correct toward: rather than having the user jump straight from the wrong model to the full, correct one, offer a bridging analogy first — an intermediate account that sits between the two, keeps part of the old understanding intact, and changes only the piece that was actually wrong. Correction happens in two steps instead of one.

Why it happens

Belief revision follows the same logic as learning new knowledge when it comes to whether it can attach to existing structure: if the new model can be reached by modifying or extending part of the old one, the user only has to adjust something they already have a handle on. If the new model requires tearing down the old one entirely and starting from zero, the cognitive burden is considerably larger, and easier to amplify through the defensive resistance described elsewhere — being told "your whole understanding was wrong, start over" is far more likely to trigger self-protective pushback than being told "this one part needs adjusting."

A bridging analogy's job is exactly to turn "wholesale demolition" into "local adjustment": it preserves whatever part of the old model wasn't actually broken, and introduces new content only at the specific point the falsifying experience exposed as wrong. The user doesn't have to abandon the entire framework they've already built — they just move one piece within it, swap out one part of the content — which shortens the cognitive path to correction and makes it more likely to be completed in full, rather than abandoned halfway because the leap felt too large, with a relapse back into the old model.

Studying it

This corresponds to the bridging analogy method in science education research: faced with a stubborn misconception, rather than presenting the target concept directly, present an intermediate analogy that's close enough to the misconception to be readily accepted, while already carrying some features of the target concept. The learner moves from the misconception to the intermediate analogy, and then from the intermediate analogy to the target concept — two small steps replacing one large one.

Common independent variables: whether the correction path includes a bridging analogy (presenting the target concept directly versus presenting the bridging analogy first). Common dependent variables: acceptance rate of the target concept, recurrence rate of the misconception after correction, participants' self-reported difficulty of understanding.

Methodology note: how well a bridging analogy works depends heavily on how well the analogy itself is chosen. In research, selecting one usually requires detailed interviews into the learner's existing understanding, to find that intermediate point which is close enough to the old understanding while already being partly correct. Doing the same in an interface context requires first finding out what the user's actual wrong model really looks like, rather than starting from a designer's own assumption about "probably what the user thinks."

Where it stops holding

• A bridging analogy requires that a suitable intermediate concept genuinely exists. If there's no natural transitional ground between the wrong model and the correct one, forcing together a makeshift bridging account risks introducing a third model the user now has to separately learn — and later have replaced too — which just adds an extra correction task rather than removing one.
• The bridging analogy itself must be simple enough that it isn't more complex than the wrong model it's replacing. If understanding the bridging analogy takes more effort than understanding the target concept directly would, this method loses its point — it adds a step for the user instead of removing one.
• This entry addresses the size of the cognitive span correction requires — it doesn't address whether the user is willing to accept the correction in the first place. Willingness depends heavily on whether the delivery triggers resistance, which is a separate layer that has to be handled at the same time for correction to actually land.

Applying it

• For each wrong model confirmed to be entrenched, work out specifically which part it shares with the correct model and which part needs to change, and look for a transitional account only for the part that needs changing — rather than designing an entirely new explanation from scratch.
• Present the bridging analogy at the same interaction moment as the falsifying experience, rather than as a separate lesson given afterward — right as the user experiences the failed prediction, they should also see "here's how the old understanding can be adjusted," instead of being left to guess at a direction during the gap that follows.
• How to check: find a few users who still hold the wrong model and test directly whether the bridging analogy itself needs extra explanation to be understood. If the bridging analogy requires a lengthy explanation of its own, it wasn't chosen close enough to the user's existing understanding, and a simpler intermediate point needs to be found.

Related

• Same group: A7.14.1 Telling users the correct procedure rarely overrides an entrenched wrong model; it takes a first-hand experience that falsifies the old one · A7.14.2 Correction should start with a small-scale, controlled-consequence falsifying experience, not a one-shot fix · A7.14.3 A correction delivered too bluntly triggers resistance and can strengthen the original wrong model
• Nearby: A7.03 Metaphor · A7.13 Cross-product transfer of mental models
• Search terms: bridging analogy · transitional model · conceptual change · belief revision