C4.18.1Hand-role bindingdesignresearch

Some systems bind functions to the left or right hand, giving the other hand a different role

Aliases: hand-role binding · left-right division of labor · non-dominant hand

What it is

Hand-role binding pins a function to the left or the right hand: the right pinches to drag, the left summons a menu; or the non-dominant hand frames a reference while the dominant hand works inside it. The other hand is not a spare copy of the same commands; it holds a different role. That is the opposite of “either hand may perform this gesture,” and it is not finger-count encoding—the code is laterality, not how many digits are extended.

Why it happens

Both hands can move at once. If they share one input stream, a menu pose interrupts an in-flight drag. Binding roles to hands spends the body's left-right asymmetry on a stable mode channel so a button is not required to switch. Guiard's account of bimanual labor is that the non-dominant hand first supplies a reference frame and the dominant hand does high-frequency fine work on that frame. In mid-air this becomes “left = palette, right = tool,” which only holds if the system keeps knowing which hand is which. Once the binding sticks, muscle memory lives on one side; skill does not automatically transfer when the opposite hand is suddenly asked to do the same job.

Studying it

Compare three mappings: either hand may fire the function, the function is fixed to a nominated hand, and non-dominant frame plus dominant tool. Tasks include menu-plus-drag or “left hand holds, right hand rotates.” Dependent measures are time, mode errors (wrong hand), bimanual interference, and workload. Record handedness; do not publish “right-hand tool is faster” from a right-handed sample as a general result. Allow enough practice: the benefit of binding shows up after learning, and the first minutes of clumsiness hide the long-term division of labor.

Where it stops holding

Someone holding a cup, in a cast, pushing a stroller, or missing a limb cannot keep a contract that says “the left hand must open the menu.” Binding is meaningful when a headset sees both hands; if only one hand enters the camera volume, a hard left-right pin hides the function. Brief one-shot commands do not need role division; forcing a binding raises discovery cost. If the tool already encodes mode with color or shape, a laterality binding is a second code, and users will not know which one wins on conflict.

Applying it

  • Use hand-role binding only on tasks where both hands work simultaneously and continuously; keep one-shot commands as “either hand.”
  • Express the split with stable left-right marks (a left-side panel, a right-hand tool cue), not only once in an onboarding animation.
  • Include scripts where one hand is occupied (holding a cup, gripping a rail) and confirm that critical functions still have a path that does not need the bound hand.

Related

  • Same group: C4.18.2 Hand-role binding requires continuous correct left-right identification; errors swap functions · C4.18.3 Independent bimanual operation requires separate onset, offset, and engagement state machines · C4.18.4 Handedness differences mean default left-right bindings may not fit left-handers
  • Adjacent: C4.09 Finger count as input · C4.13 Full-body pose and skeleton tracking
  • Search: hand-role binding · bimanual · Guiard kinematic chain

Cards in the same group

Quick Actions

Share

Share this page

ios_share

https://hci.top/en/handbook/C4.18.1