Increasing cursor count dilutes the state information a single cursor can carry

What it is

One cursor can concentrate affordance, busy, selection, or drag state in shape, colour, animation, and position. As count rises, this state information is diluted: people cannot read every subtle change or correctly pair state, identity, and object relationship across several moving marks.

Why it happens

Visual attention itself has a limit on how many independent moving targets it can track reliably at once — a capacity known as multiple object tracking, and beyond it, the extra targets are either ignored or tracked with mistaken identity. A multi-cursor scene runs straight into that limit: every cursor is moving and can change state at any moment, so attention has to keep switching allocation among them instead of concentrating fully as it can with a single cursor. Colour makes this worse: it is asked to encode both "who this is" and "what state it's in" at once, cramming two kinds of information into a single visual dimension, and beyond a very small number of cursors, the two inevitably interfere with each other.

Studying it

Increase cursor count, motion frequency, and state types stepwise; measure state recognition, identity attribution, task performance, occlusion, and workload. Compare highlighting only active cursors, aggregated summaries, local detail, and constant full display to identify when per-cursor encoding fails. Concretely, use multiple object tracking as the explanatory frame and measure specifically where the ceiling on accurately tracked cursor count sits, then use that number to decide at what headcount the system should switch to an aggregated display.

Where it stops holding

Not every viewer needs all detail: an observer may need only presence while an editor needs exact position. Infinite cursor enlargement does not solve dilution; it adds occlusion and distraction. Layered disclosure is usually better than piled encoding.

Applying it

• Layer identity, activity, and operation state: keep default simple and expand detail when active or selected.
• Carry global state in object highlights, side summaries, or notifications instead of placing everything on cursors.
• How to check: measure the ceiling on how many cursors people can accurately track at once under the current design, and once simultaneous participants exceed that number, automatically collapse the excess into an aggregated cue such as "N more" instead of continuing to lay them all out individually without limit.

Related

• Same group: C1.24.1 In multi-cursor settings, each cursor needs a visually distinguishable identity marker · C1.24.2 Simultaneous multiuser operation in one area needs arbitration over whose action takes effect · C1.24.3 One-person multi-cursor and real-time multiuser cursors carry different semantics
• Nearby: D1 Output and feedback channels · A1 Human capabilities and human-factors foundations
• Search terms: visual attention · cursor state · information density