Clark’s nutcrackers hide seeds, but remembering food is not simply remembering a nearby object. Laboratory studies ask which relationships and features of a place can guide their search.

Learning a relationship, not one distance
In a 1997 experiment, Alan Kamil and Juli Jones trained nutcrackers to find a point midway between two landmarks. The spacing between those landmarks varied. The birds also located the midpoint at new spacings, showing transfer beyond the training distances.
“Halfway between these two objects” is different from “a fixed distance from that object.” Further tests distinguished locating the imaginary line joining the landmarks from locating the right position along it. This establishes an experimental ability, not a demonstration that every natural cache is recovered using a midpoint rule.
More objects need not mean better guidance
A 2014 study by Peter Bednekoff and Russell Balda examined actual cache recovery in rooms. Adding objects to a large room did not improve accuracy, nor did subdividing it with panels. The birds nevertheless recovered caches more accurately in a smaller room than in the larger one.

The authors inferred that large structural features helped guide recovery. Their suggestion that stable cues are useful where snow or disturbance obscures smaller landmarks is an ecological interpretation, not a field experiment proving that explanation.
Together, these studies show why the arrangement of a search environment matters. For a different experimental approach, see chickadee cache memory and neural recordings.

Leave a Reply