I suspect this is because human perception of brightness (like human perception of sound volume) isn't linear, so the same amount of light spread over more of the visual field may appear brighter. Of course, you're not trying to maximise brightness so much as contrast, which complicates things.

The problem I see with spinning a photon light (as per Susan's suggestion) is that, unlike a lightstick, most of the light will go directly away from the centre of motion. It might still be better to spin it so that the plane of motion is parallel, rather than perpendicular to, the direction of the observer, but it'd be nice to have some empirical evidence.

eeph