Physiological input latency usually exceeds pointing devices and does not fit operations that need immediate feedback

What it is

Mouse and touch close a loop in tens of milliseconds; people confirm they are still in control from immediate displacement. Most physiological channels’ stable detection is an order of magnitude slower. Operations that must “follow as I move”—drag, ink, aim, a virtual hand—mismatch: the person has already altered the action on missing feedback before the label arrives.

Why it happens

A pointing device turns mechanical displacement into a cursor almost immediately; the visuo-motor loop experiences lags much above about 100 ms as stickiness or loss of control. Physiological channels already spend most of the budget on effector and window; add classification and radio and they easily enter “the action is over when feedback arrives.” Immediate feedback also carries discoverability: without a following cursor, users do not know whether the command was heard. Sticking physiology on a drag is opening, with a slow sensor, a servo designed for hand speed. What fits physiology is open-loop or long-period: pick and wait, adjust a second-scale continuous quantity, adapt to a state that has already finished.

Studying it

Run the same task with mouse, touch, and the target physiological channel; plot time against index of difficulty (Fitts) or writing speed. Factor: delay compensation on or off. Outcomes: errors, rated loss of control, whether people switch to pause-and-click. If users spontaneously recode a continuous task as discrete taps, the loop is already broken. A control that artificially delays a mouse to physiological magnitude tests whether collapse is “physiological noise” or “pure lateness.”

Where it stops holding

An EMG click can approach a key under careful engineering, and still rarely approaches precise dragging. Predictive rendering works for touch inking because kinematic inertia is predictable; intent prediction from physiology is much weaker. A deliberately slow energy bar in a game is not loss of control, because it does not pretend to follow the hand. Eye pointing has its own delay structure and should not be lumped with EDA and heart rate as “physiology, therefore unfit.”

Applying it

• Leave drag, ink, and aim to hand and eye; bind physiology to discrete confirm or second-scale adjustment.
• If a continuous object must be driven physiologically, quantize and slow its motion so delay is small relative to the object’s time constant.
• State in the spec “not for operations that need immediate following feedback,” and do not speed-edit demo videos.
• Verify by driving a drag or tracking task with the target channel; stop–go paths or a switch to tapping mean it does not fit immediate-feedback work.

Related

• Same group: C9.11.1 There is inherent delay from motor or affective intent to stable physiological detection · C9.11.2 Latency magnitude differs by signal type; EMG is faster than heart-rate change · C9.11.4 Hiding latency with smoothing or prediction can introduce misclassification
• Adjacent: C1.15 Pointing Device Metrics and Throughput · C5.08 Inking Latency
• Search: control-display lag · closed-loop pointing · physiological input delay