Perception demos · MSense Lab

A blind spot for blue, right where you look

The centre of your fovea, the spot you point at whatever you look at, is blind to two kinds of things. It has no rods, so very dim objects vanish when you look straight at them (that is why astronomers look slightly beside a faint star). And a yellow macular pigment plus very few "blue" cones make it almost blind to small blue objects. These demos let you see that hole and what your brain does with it.

You need a deep blue filter, for example a "Tokyo Blue" lighting gel, held close to one eye; close the other eye. Dim the room. Without a filter you can still try demo 1 with the blue–yellow setting.

1 · See your own blue spot

Fixate the cross in the middle of the flickering field. After a few seconds, many people see a dark spot, like an ink drop, right where they are looking. It is about one degree wide: Maxwell's spot, the shadow of your macular pigment.

+

2 · Motion extrapolation into the blue spot

With the filter on, fixate the vertical line. Don't follow the dot. Watch where it disappears before the line and where it reappears after it.

Dot:

What to notice. The hole is symmetric around the line, so the dot should vanish and reappear at the same distance on both sides. It doesn't: it seems to travel into the hole before vanishing and reappears further away. While the dot is invisible, your visual system keeps predicting its path, compensating for the tens of milliseconds that signals need to travel from eye to brain.

Why only here? Normally, when a moving object stops, the retina fires a strong "stop" signal that overrides the prediction, so we don't see an overshoot (try "Bar stops at flash" in the flash-lag demo). Inside the blue-blind spot, the retina cannot send that signal, so the prediction becomes visible. In the lab, the overshoot corresponded to roughly a tenth of a second. Switch the dot to green: the fovea sees green well, and the overshoot should shrink.

Shi, Z., & Nijhawan, R. (2012). Motion extrapolation in the central fovea. PLoS ONE, 7(3), e33651. Screen colours and filters vary, so the effect is a demonstration, not a measurement.