Replacing physical keys with a touchscreen removes eyes-free capability

What it is

When volume, shutter, and radio presets move from raised keys onto glass hit regions, a finger sliding the housing no longer finds a stop, and a press no longer cages the pad in a cap. Loss of eyes-free operation after that replacement is not a sensitivity problem. Localization and “I am settled on it” are removed together. A chassis vibrator can rumble after a press; it does not tell the finger where the hit region is, and it cannot find the control before contact.

Why it happens

Capacitive glass is tangentially uniform: no step, no side wall, no slide-to-stop. Software hit regions also move and resize with each skin, so a motor program is dismantled by the next layout. A whole-device actuator confirms that a press happened in time; spatially the surface is still a plate, so the pad cannot check “am I inside the control” by contour. Hitting the right control then requires tearing gaze off the primary task and aiming it at the glass. Driving, carrying, framing a shot: that gaze tax becomes a primary-task error. Physical keys hand space to housing geometry; a touchscreen hands it back to the eyes.

Studying it

Implement the same function as a raised key and as a screen hit region, then compare them under a visually occupied primary task.

Independent variables: control type (raised / flush capacitive / on-screen), whether the hit region is spatially fixed, presence of chassis haptics, primary-task load. Dependent measures: glance count and duration off the primary scene, wrong-function triggers, primary-task outcome (lane deviation, missed focus).

Watching the device in a lab produces “the touchscreen is fast too”—vision is doing the tactile job. The informative signal is the head or eyes: when the hand reaches for volume during the primary task, does gaze follow. Pinning the hit region to one place for the whole study overestimates products whose controls migrate with each software release.

Where it stops holding

Physical volume and camera keys left on the bezel locally preserve eyes-free use; what is lost are the functions the glass absorbed. In-display haptics (piezo, ultrasonic texture) can say “you are on a button” after contact; they help little before contact, and gloves or thick films often kill them. Paddle shifters behind a wheel and a shutter on a camera body show that not every function must go on-screen. In a still setting where both eyes may stare at the device, the loss is easy to miss; it concentrates in walking, driving, and one-handed viewfinding.

Applying it

• Do not ship high-frequency, must-fire-while-looking-elsewhere functions as screen hit regions only; keep a physical key on a spine the grip already reaches.
• If the function must live on glass, pin the hit region to one corner and keep it there across generations; do not let a reskin move it.
• Treat chassis vibration as a receipt that a press occurred, not as a way to find the control.
• Verify: adjust the same function while walking or in a driving mock-up; count glances off the primary scene with eye tracking or a head-mounted camera. If the physical-key version needs clearly fewer looks, do not replace that key with a hit region. Repeat in gloves and with the film users actually apply.

Related

• Same group: C10.01.1 Physical keys can be located and confirmed without looking · C10.01.2 Key shape and spacing carry discrimination for eyes-free use
• Adjacent: C10.07 Tradeoffs Between Physical and On-Screen Controls · C6.02 Key Travel and Tactile Confirmation
• Search: eyes-free loss · touchscreen replacement · physical keys