Children's use of spatial reference frames and spatial relations in a scene recognition task (DATASET CHILDREN)

Abstract

In order to represent spatial relations between objects in a scene we may adopt different reference frames: a visual snapshot (‘integrated picture’), an egocentric (body-referenced) and/or an allocentric (external-referenced) reference frame. These reference frames rely upon different neural structures. Children use different frames of reference more effectively with age, possibly showing the functional maturation of these neural structures. In the present study, children 5 to 10 years were tested on an object-location memory scene recognition task in which, on some trials, the viewpoint between presentation and test was changed, either over a small angle or over a large angle to dissociate the use of different frames of reference. Besides children’s ability to make a correct ‘coarse’ judgment of an object’s location in the scene under different viewpoint conditions, we were interested in children’s ability to reproduce categorical and coordinate spatial relations after a viewpoint change. Although participants performed worst on the large viewpoint-change condition, which measured participant’ use of an allocentric reference frame, in all task conditions mean performance was above chance level. This suggested that the first five years reflect an important time-period for maturational changes in the neural systems related to reference frame use, and the age differences in overall performance might show that small maturational changes take place afterwards. In addition, participants performed better on the categorical trials compared to the coordinate trials, showing that children use these relations to memorize object locations.