The correction phase closes in on the target with small submovements

What it is

Once the initial impulse has cleared most of the distance, the small stretch that remains isn't handled by one more burst of force — it's completed in relay by a series of much slower, much smaller submovements, each nudging a small step toward the target and then checking the remaining gap again, repeating until the position falls within the target's bounds. Together these make up the correction phase of target acquisition, and it operates completely differently from the initial impulse: the impulse is a single ballistic burst fired once and never revised, while the correction phase is completed by checking and adjusting one step at a time.

Why it happens

The correction phase consists of several small-amplitude submovements because, at this point, the remaining distance is already small, and continuing to move at high speed risks overshooting for little benefit given the risk. The more reliable approach is to keep each step's amplitude low, move a short distance at reduced speed, and — while speed is low enough to still make out the current position — check the gap to the target and decide the direction and distance of the next step. Each submovement is itself one feedback-driven round of adjustment, and its amplitude keeps shrinking as the position gets closer to the target, until the landing point is judged to be within the target's bounds.

Studying it

A common way to identify submovements is to find the secondary velocity peaks that appear after the initial impulse ends on the velocity-time curve, with each secondary peak corresponding to one distinct submovement; counting these secondary peaks and measuring the displacement amplitude and peak velocity for each one describes exactly how the correction phase unfolds. Researchers also typically compare submovement amplitude and count across task conditions (whether visual monitoring of current position is allowed, for instance) to confirm that this phase genuinely depends on being able to check the current state.

Where it stops holding

Identifying the boundaries of submovements also depends on being able to distinguish clear secondary peaks in the velocity curve; in recordings with low speed and substantial jitter, the boundary between adjacent submovements can be ambiguous, and different detection algorithms can produce somewhat different submovement counts. How the correction phase actually unfolds is also jointly shaped by target size, allowed time, and feedback delay; this fact by itself only describes the structural point that correction is completed through small submovements, not how those factors specifically change their number or amplitude.

Applying it

• Controls involving fine positioning (a resize handle's edge, an alignment snap point, a text-insertion caret) should support and expect that a user's movement speed will drop sharply and amplitude will shrink noticeably when approaching the target; a control's response sensitivity shouldn't hold a constant movement-to-displacement ratio across the entire drag — it should lower that ratio near the target, matching the naturally slower movement rhythm users adopt there.
• Fine-adjustment interactions (a precision slider, a fine-tune dial) can take advantage of this structure by automatically switching to a finer sensitivity tier once a sharp drop in movement speed is detected, so the small submovements users naturally make correspond to smaller actual value changes, reducing repeated micro-adjustments caused by mismatched sensitivity.
• How to check: record the displacement trajectory over the final short stretch of a drag or pointing task, and check whether it shows repeated small back-and-forth or stop-and-move patterns. If it does, the control's current sensitivity setting may not be differentiating this correction phase, and it's worth testing whether a tiered sensitivity scheme reduces the total number of adjustments needed to complete the operation.

Related

• Same group: A8.08.1 The initial impulse phase covers most of the distance quickly · A8.08.3 The number of submovements determines total movement time · A8.08.4 Smaller targets require more submovements · A8.08.5 Undershoot and overshoot of the initial impulse carry different costs
• Nearby: A8.07 Closed-Loop Control · A8.11 Graded Force Control
• Search terms: homing submovement · current control · target acquisition kinematics