Item

Abstract

Motion contributes to object recognition, particularly if shape is visually similar or if an object moves the same way on different encounters. Here we examined how
shape diagnosticity and motion variability may interact
and contribute to recognition. Observers learned novel
objects that rotated in depth in a constant or variable way
on each trial. For shape-nondiagnostic objects, constant
motion lead to faster learning and variable motion
resulted in more robust representations to changes to
object dynamics. By comparison, for shape-diagnostic
objects, performance was similar with both types of
motion and remained sensitive to changes to object
dynamics.