Abstract
The purpose of the present study was to investigate the effects of category similarity and spatial contiguity between central and incidental stimuli on children's incidental learning from the viewpoint of reexamination of a selective attention hypothesis.
There were three variables in the present experiment category similarity (similar, nonsimilar), spatial contiguity (contiguous, spaced), and age levels (kindergarten, second grade, and fifth grade). Thirty two Ss in each grade were assigned to one of the four experimental conditions, contiguous-similar, contiguous-nonsimilar, spaced-similar, and spacednonsimilar.
The task consistod of a display of eight cards, consisting of two line drawings, central and incidental stimuli. The same two stimuli were always paired together. Following orienting instructions which forced the S to focus his attention on central stimulus in each card, the eight cards were arranged in a row and were presented simultaneously on a display panel. After the stimulus presentation, each S was required to recall as many central and incidental stimuli as he could. Then an incidental cued recall test was given.
The followings were the main results.
(a) The central learing scores for the fifth graders were significantly higher than those for the kindergarteners and for the second graders.
(b) The incidental learning scores for the fifth graders were significantly higher than those for the kindergarteners and for the second graders in the contiguous-similar condition, but not in the other conditions.
(c) For the category similarity, the incidental learning scores in the similar condition were significantly higher than those in the nonsimilar condition.
(d) For the spatial contiguity, the incidental learning scores in thecontiguous condition were significantly higher than those in the spaced condition.
(e) The amounts of incidental cued recall in the contiguous-similar condition were significantly higher than those in the other conditions.
The main findings that the scores of the incidental learning for the fifth graders were significantly higher than those for the kindergarteners and for the second graders in the contiguous-similar condition is a contrast to the previous studies showing a consistency among age levels.
