A mode is a state where the same action produces different outcomes
Aliases: mode · moded interface · state-dependent command
What it is
A mode is a design arrangement in which the same physical action — the same key, the same click, the same gesture — is interpreted as a different command depending on the system's current state. This is different from a "setting": a setting changes a parameter's value, a mode changes the mapping from action to meaning itself. The classic example is the letter keys of an early text editor — pressed while in insert state they type characters, pressed while in command state they're read as editing instructions, so the identical keystroke does completely different things depending on state. A mode is not inherently a design flaw; it's a way of reusing a limited set of physical input channels — the number of keys on a keyboard, the number of gestures a touchscreen can hold — against a command set that's far larger.
Why it happens
The number of commands a system needs to express routinely exceeds the number of physical controls available, so reuse is close to unavoidable: assigning the same key or gesture to several functions is only possible if its meaning is made to depend on the current state. That's why modes exist — not designer oversight, but a structural response to scarce controls. The cost follows directly: once a mode exists, the user has to keep a mental record synchronized with the system's actual state, answering "which state am I in right now" on demand, and that record drifts out of sync with reality as time passes and other tasks intervene. The moment that record drifts is the opening where a mode error becomes possible; exactly how the error gets triggered is a separate, more specific mechanism. What matters here is that a mode's mere existence opens a running ledger the user's cognition has to keep maintaining.
Studying it
A common way to find out how many mode points a system actually contains is to draw its behavior as a state-transition diagram: list every input event (key, click, gesture) and check whether it leads to different action outcomes from different state nodes. Any input event that points to different outcomes from two different states marks a mode point. This systematic approach comes from formal specification and verification work on human-machine systems, and it catches mode points buried deep in an interface — ones outside the core workflow that reviewers rarely reopen during a walkthrough — far more reliably than eyeballing screens one at a time.
Where it stops holding
Not every state-dependent behavior counts as a mode worth worrying about. If the behavioral difference is made visible right at the moment of use — a menu that visibly swaps its option labels on a different page, entered by a deliberate choice the user just made — the error risk is low, because there's no "current state" memory that has to survive across time. A mode becomes genuinely dangerous when its state persists through other actions, other tasks, and elapsed time, which is exactly when a user's memory of the state has room to diverge from what the system actually holds.
Applying it
When reviewing an existing or in-progress interface, list every place that responds to the same key, gesture, or control position, and check whether it points to a different outcome depending on state; wherever it does, log it as a mode point and carry it into the indicator, exit, and reversibility checks that follow. Verification: have a test participant complete two tasks back to back, inserting a state-switching action between them, then watch whether — on returning to the first task — they still operate the shared control according to the pre-switch state's meaning. If they do, that mode point hasn't been made visible enough.
Related
- Same group: A10.02.2 mode errors are caused by a mismatch between the user's and the system's judgment of the current mode · A10.02.8 common sources of hidden modes: Caps Lock, IME state, edit vs. browse state
- Nearby: A10.01 distinguishing slips, mistakes, and violations (a mode error is a mistake in that scheme, not a slip)
- Search terms:
mode·mode error·moded interface
Cards in the same group
- A10.02.2Mode errors come from a mismatch between the user's and the system's judgment of the current mode
- A10.02.3The mode indicator belongs at the point of visual focus, not a status bar or corner
- A10.02.4A quasimode is sustained by continuous physical actuation and exits on release
- A10.02.5Modeless design assigns meaning through different actions instead of relying on state
- A10.02.6A mode can auto-revert after an idle timeout
- A10.02.7How explicit a mode switch needs to be should match its consequences
- A10.02.8Common sources of hidden modes — Caps Lock, IME state, edit vs. browse mode
- A10.02.9The consequences of a mode error must be recoverable