The Door Test
Applying affordances to real objects
"Let's audit the studio. Every object in here either helps you or confuses you. Grab your notebook — three categories: 1) the affordance matches the action, 2) it doesn't, 3) there's no signifier at all."
"The front door is the obvious one — the bar handle on the push side. That's category two."
Sable flipped her notebook open and wrote "AUDIT" at the top. She liked how Tomoko organized things — numbered, specific, no room for doubt.
"What about the kitchen faucet? I've been watching people all week. Everyone pulls the handle toward them for hot, but hot is actually the other direction."
"I timed it yesterday — 4 seconds before the temperature changes. That's long enough for someone to assume they chose correctly and walk away with cold water."
"So the handle direction is misleading, and the feedback is too slow to correct the mistake?"
"Two failures stacked. The mapping between handle direction and temperature is backwards, and the delayed feedback means no one ever learns the correct action."
"Could we fix it with a constraint? If the handle only turned one way, you'd control temperature by how far you turned — one motion, one variable."
"Single-axis constraint. The user can't get the direction wrong because there's only one direction. Write that down — it's a pattern worth remembering."
The principle of leveraging visible cues so users can identify options rather than retrieve them from memory. Designs that favor recognition make choices, actions, and information visible at the moment they're needed — reducing the burden on the user's memory.
- Recognition presents options visibly — a labeled menu is easier than a blank command line
- Recall demands users retrieve information from memory without prompts or cues
- Good design minimizes recall by showing what's possible at each step
In the studio: The Meridian kiosk asks riders to select a fare zone by number — pure recall. Displaying zone names on each button would shift the interaction to recognition, letting riders find their zone without memorizing the numbering system.
A user's internal understanding of how a system works — the assumptions, expectations, and beliefs they carry about what will happen when they interact with it. Mental models are built from prior experience and can be accurate or completely wrong.
- Users form mental models from experience with similar objects and systems
- When design matches the user's model, it feels intuitive; when it contradicts, confusion follows
- Designers can shape mental models over time by making system behavior consistent and predictable
In the studio: Everyone at Indent pulls the kitchen faucet handle toward them expecting hot water — a mental model built from years of using other faucets. This one works backwards, but the slow feedback never corrects the assumption.
Sable pressed the elevator call button and waited. The panel was a flat touchscreen designed to look like physical buttons — no click, no give beneath her fingertip. She pressed again, harder.
Without the tactile affordance of a real button, even a familiar action becomes uncertain — the same kind of mismatch Sable felt when the studio door's bar handle said "pull" but the door needed pushing.
The studio supply cabinet had two identical unmarked drawers. Sable opened both before finding pens in the second. By the next morning, a new visitor had made the same mistake.
A missing signifier forces trial and error. As Rafa described in his forum talk, good design makes the right action visible — not something you have to memorize.