One screen carries one focus

What it is

A watch screen answers one question currently being asked. A navigation screen is “what is the next turn,” not remaining time, speed, and the whole route at once. A notification screen is “who, and what do they need,” not three unread threads side by side. Focus is the object of attention, not a pixel budget: drawing two questions smaller is still two foci. This entry counts how many questions one screen asks. It does not count how short the words are, and does not count whether the next screen is reached by paging or scrolling.

Why it happens

A microdisplay cannot host a two-step visual strategy of survey-then-read. One fixation covers about one primary object on that glass; a second object either lands in an unreadable periphery or needs a second saccade, and a wrist raise often will not wait for that saccade. Two foci also fight for weight: two equally bold numerals are read as “both matter,” and the decision becomes a comparison instead of a readout. Comparison asks working memory to hold two values—the cost a glance will not pay. One focus lets contrast, type size, and position all bet on that one thing, so it can be taken on the first fixation. A multi-focus screen looks information-rich; what actually leaves with the wearer is often the wrong one, or none.

Studying it

Use brief presentation: light the first screen for a second or two, then blank it, and ask the question that screen was supposed to answer. Contrast one-question screens with three-question screens (type shrunk to fit). Add a condition in which one of the three is visually promoted to primary.

Independent variables: number of questions on one screen, relative area of the primary object, whether secondary objects share color and size. Dependent variables: accuracy on the target question, rate of reporting the wrong object, reports of “don’t know where to look.”

A lab that allows unlimited looking is measuring reading, not whether the focus was taken on the first fixation. Walking or conversation as a secondary task is what makes a second saccade a luxury. Do not replace “how many questions” with a composite “information density” score—density can come from decoration; focus count must be counted in user questions.

Where it stops holding

A watch face is a special structure: time is the legal primary, complications are slots the wearer already chose. Their coexistence depends on “on this raise I already know which slot I want,” not on stuffing three ad-hoc questions onto one screen. Dual workout metrics (pace and heart rate) count as one focus only when the two values already form one judgment (“pace is fine but HR is high”); two unexplained numerals side by side are still two foci. Settings and consent screens were never glance objects and may hold several controls, but each screen should still hold one current decision.

Applying it

• Write one user question per screen; if you cannot, or the sentence contains “and,” split the screen.
• If secondary information must be present, demote it to a modifier that cannot be asked on its own (color, a small mark), not a second readable numeral.
• Give the primary object the clearly largest patch; everything else is at least one step weaker in area and contrast.
• Verify by lighting the screen for a second or two, covering it, and asking only that user question. A wrong answer, or the neighboring numeral, means more than one focus. Fix until only that one can be answered, then decide whether the squeezed-out question belongs on the next screen.

Related

• Within the group: K4.05.2 Text must be extremely compressed without becoming ambiguous · K4.05.3 Paging costs more than scrolling
• Adjacent: K4.01 Wrist Raise and Glance · K1.02 Screen Size and Density Differences · A1.09 Visual Search
• Search terms: one thing per screen · visual clutter · watch information density