Why a moving object is remembered worse than a still one

Viewers follow a labelled shape across a screen for eight seconds and are then asked what the label said. Accuracy is low. Viewers shown the same shape stationary for the same duration report the label reliably. Both groups looked at it for eight seconds and both saw it clearly.

What everyone sees

Sustained looking feels like sustained recording. Tracking a moving object is effortful and the effort feels like attention being paid, so failure to recall afterwards is read as a memory problem separate from the looking.

What is actually happening

Multiple-object tracking research, beginning with Pylyshyn and Storm, established that maintaining spatial indexes on moving items consumes a limited resource. Later work found that tracking impairs encoding of the tracked items’ features: position is maintained while identity is not. Alvarez and Franconeri showed the resource trades off flexibly, so a demanding tracking task leaves little capacity for the feature binding that recall depends on.

Why it stays hidden

The dissociation hides because vision reports its results as complete. There is no experience of position being kept while identity is dropped; the object simply looks like an object, labelled and whole. The deficit only appears when recall is requested, by which point the encoding opportunity has passed and the viewer has a vivid memory of having looked.