Once an object boundary forms, it automatically steers attention to follow the whole object rather than staying fixed in place

What it is

Once attention has locked onto a formed perceptual object, when that object moves, attention automatically moves with it rather than staying behind at the spatial location where the object used to be, waiting for whatever new content shows up there. This automatic-following property is another key piece of evidence that object-based attention is distinct from pure space-based attention — if attention were allocated purely by location, new content appearing at the object's old position should get the same boost. What actually happens is that attention tends to follow the object that moved away, and the vacated spot does not automatically inherit the previous attentional advantage.

Why it happens

The perceptual system keeps continuous track of an object's identity — even while it is moving or briefly occluded — as long as its trajectory is continuous and consistent with the expectation that objects in the physical world don't vanish and reappear out of nowhere. The system keeps treating it as the same persisting thing; this ongoing tracking capacity is part of what's called the object file mechanism. Attentional allocation is bound to this identity-tracking process rather than to a fixed screen coordinate, so when the object moves, the anchor for attention updates to the object's new position along with it. This is also why a rapidly moving object, or one that briefly disappears and reappears, can often still be kept track of as long as its motion path fits continuity expectations — without the user having to re-search the whole scene.

Studying it

A common paradigm for studying this has participants track one or more target objects moving on screen, sometimes briefly occluded, while manipulating movement speed, occlusion duration, and whether the trajectory is physically continuous, then measuring tracking success rate and how long it takes to relocate a target after losing track of it. Another paradigm directly compares the detection performance predicted by two hypotheses — "attention follows the moving object" versus "attention stays at the object's original location" — to pin down whether the anchor for attention is really the object or the coordinate.

Common independent variables: target speed, occlusion duration, and the number of targets tracked simultaneously. Common dependent variables: tracking accuracy, and reaction time to re-identify a target after occlusion.

Methodological caveat: there is a clear upper limit on how many objects can be tracked at once, and tracking performance drops sharply beyond that limit. The number of targets chosen in a study directly determines the tracking-capacity ceiling that gets measured, so comparisons across studies need to check whether the target counts actually match.

Where it stops holding

• The automatic-following effect depends on the motion trajectory itself being predictable or at least continuity-consistent. When a trajectory jumps randomly and lacks a predictable pattern, the following effect degrades noticeably, and users are more likely to briefly lose the target once it moves.
• There is a capacity limit on how many objects can be tracked simultaneously; once that limit is exceeded, additional objects can no longer be tracked reliably, and attention effectively degrades back to needing a fresh search.
• This entry only covers the phenomenon of object boundaries steering attention to follow motion once formed. How an object boundary itself gets identified against a cluttered background is a separate question in perceptual grouping.

Applying it

• When interface elements transition through animation (a list item expanding into a detail view, a card flying from one position to another), keep the motion path continuous and avoid abrupt, jump-cut position changes. This lets a user's attention naturally follow the element, cutting down the search time needed to relocate the target once the animation ends.
• Avoid having multiple elements animate at once with crossing, easily-confusable paths. When the number of targets that need tracking exceeds what a user can reliably track simultaneously, consider staggering the presentation or reducing how many elements move at the same time.
• When an element briefly disappears behind an occlusion and reappears, have it reappear at a position and timing consistent with motion continuity (continuing along its original path from where it was occluded), rather than having it appear out of nowhere at a new position that breaks the expected motion logic.
• Verification: record the interface animation at real device frame rates and check whether key elements' motion paths show position jumps caused by dropped frames. Those jump points are exactly where users are most likely to lose track of the target.

Related

• Same group: A5.15.1 Attention can be allocated to a region of space or to a whole perceptual object · A5.15.2 Two properties within the same object both benefit from attention, but properties across objects do not · A5.15.3 The spread of spatial attention can narrow or widen like a spotlight
• Adjacent: A5.07 Attentional capture
• Search terms: object file · multiple object tracking · object-based attention · motion continuity