Pointing to an allocentric and egocentric remembered target

Pointing to a visual target that disappears prior to movement requires the maintenance of a memory representation about the location of the target. It has been shown that a target can be stored egocentrically, allocentrically, or in both frames of reference simultaneously. The main goal of the present study was to compare the accuracy and kinematics of a pointing movement to a remembered target when egocentric, allocentric, or combined egocentric and allocentric coding was possible. The task was to localize, memorize, and reach to a remembered target. Condition 1 was the "no-context" condition and involved presenting the target in a completely dark environment (egocentric condition). For 2 other conditions, the target was presented within a visual context provided by an illuminated square. Condition 2 was the "stationary-context" condition and involved keeping the context at the same position during the whole trial (egocentric and/or allocentric coding). Condition 3 was a "moved-context" condition that involved shifting the context to a different location during the recall delay (allocentric coding). Movement accuracy and kinematics results were strikingly similar for the moved-context and stationary-context conditions. These results suggest that when both allocentric and egocentric coding are possible, an allocentric strategy is used.