Mismatched visual and interaction fidelity pulls feedback onto the wrong dimension

What it is

The problem is the mix across dimensions, not only that the whole thing is too pretty. A prototype can look release-ready while interaction is still hotspots on stills; or a gesture can track while the UI is still a wireframe. When visual and interaction fidelity mismatch, feedback lands on the raised dimension—a finished skin draws color notes, a clickable but incomplete gesture draws “it stutters”—and the dimension the question actually lives on goes cold. Unlike over-polish as such: even with little total effort, if one dimension is high and the other low, the high one hijacks the agenda.

Why it happens

Attention and comment follow whichever cue most resembles “the real thing.” High visual supplies an illusion of finish; people start speaking product aesthetics and stop asking whether the flow connects. High interaction and low visual make every failure to track feel like a defect, even when the question was copy. Mismatch also misattributes failure: it is blamed on the dimension that looks unfinished (“I didn’t get it because it was a wireframe”) or on the one that looks too finished (“the color distracted me”). Mixed-fidelity helps only when the high dimension is the question dimension; mix the wrong way and mixed-fidelity misleads. People do not split experience along the researcher’s table of dimensions.

Studying it

Build two-factor materials: visual high/low × interaction high/low, and code where comments land on the same task. Test whether high-visual/low-interaction suppresses structural problems, and whether high-interaction/low-visual turns visual noise into “usability.” McCurdy and colleagues argued for intentional mix on the condition that the mix points at the question; the experiment is what feedback does when that pointing fails. Record whether the facilitator said “please ignore color” at the start, and check the effect—usually weak.

Where it stops holding

Brand and motion walkthroughs need both dimensions high; the mismatch warning does not apply. Physical and speech interfaces need the visual/interaction cut rewritten. Experts can be held to one dimension by an outline, so mismatch hurts less than with ordinary participants. Guerrilla prototypes under time pressure are often mismatched; the harm is manageable if that session’s comments are not written as conclusions about the other dimension.

Applying it

• Before making the material, write “the only dimension allowed to be high this round,” and deliberately keep the other low, marked undecided on the surface.
• Do not “put visuals on for the demo and wire interaction next week” and then ask about flow.
• When coding, drop or separately file comments that miss this round’s target dimension; they do not write the conclusion.
• If comments have already run to the wrong dimension, stop, retune the mix, and do not rely on the facilitator reminding people again.

Related

• Same group: Q5.10.2 Stakeholders easily read prototype finish as engineering finish · Q5.10.3 Participants fill in missing pieces and hide genuine confusion · Q5.10.4 Polish spend should match the cost of the current question, not completeness
• Adjacent: Q5.01 Fidelity levels · Q5.07 How prototypes mislead
• Search terms: mismatched fidelity dimensions · mixed-fidelity mismatch · feedback displacement