A hit region can exceed its visual region

What it is

An expanded hit target makes the interactive region larger than its visible graphic. A compact icon can accept nearby clicks or touches, so visual and actionable boundaries need not coincide.

Why it happens

Movement endpoints vary, and a finger covers many pixels. A larger hit region absorbs edge error and reduces retries without enlarging the image. It changes event judgement, not the need for a legible action cue.

Fitts' law makes the mechanism precise: movement time grows with the index of difficulty ID = log2(D/W+1), where W is a target's effective width — the range hit testing actually accepts, not the size of the rendered graphic. Enlarge that accepted range and ID drops, taking movement time and error rate down with it, without a single pixel of the graphic changing. That is why expanding the hit region is not a shortcut: it moves the variable the formula is actually built on, and visual size was never that variable itself — only its usual stand-in.

Studying it

One extreme implementation of this idea is the bubble cursor (also called an area cursor): rather than a fixed margin, the hit region tracks whichever target is nearest and the cursor's effective activation area grows in real time to uniquely enclose it. Comparing a point cursor against this kind of dynamically expanded hit region with a reciprocal tapping task (ISO 9241-9) measures throughput (bits/s), error rate, and movement time; because the hit region's shape and size are asymmetric and position-dependent, the comparison needs layouts ranging from sparse to dense targets rather than extrapolating from a single index of difficulty measured on an isolated target.

Where it stops holding

Expanded regions compete among nearby controls; excessive expansion can activate an unseen object. For dangerous operations, canvas objects, and dense data marks, a large visual–interactive mismatch also harms predictability.

Dynamically expanding techniques degrade naturally in dense layouts: the more neighbours compete for space, the less room any one target has to grow, so the hit region gets squeezed back toward its visual size and the benefit fades as density rises. That is an inherent physical limit of the technique, not a flawed implementation.

Applying it

• Add transparent hit margin to small, isolated, frequent controls while retaining visual alignment.
• Do not extend across a neighbour's clear boundary; resolve conflicts consistently.
• In edge-click and touch tests, record which wrong object was activated, not only whether any object hit.
• Measure movement time and error rate under both the original visual size and the expanded hit region using a reciprocal tapping task; if expansion improves neither, the direction or amount of expansion is wrong, not the underlying idea.

Related

• Same group: C1.05.1 A hotspot is the single reference location within a cursor graphic · C1.05.2 A cursor's visual centroid and hotspot can create systematic offset · C1.05.4 Overlapping hit regions need explicit priority rules
• Nearby: C1.08 Drag and drop · J1 Motor and interaction accessibility
• Search terms: expanded hit target · hit slop · bubble cursor