Attention is considered to be one of the main processes supporting human cognitive activity. Moreover, the ability to select critical or relevant information and ignore others is considered to be a condition for successful task performance, and is usually thought to be related to the executivefunctioning deterted after “the five to seven shift”. In Exp. 1, the attentional processes in Hagen's central-incidental learning tasks were analyzed by using eye-tracking data, which seemed to reflect attentional behavior more directly than memory performance data. The eye-movements of 72 children (kindergarteners: 21, 2nd graders: 25, 6th graders: 26) were recorded while working on the tasks. Results: (1) the number of fixations on the central atimuli in 2nd graders was significantly greater than the kindergarteners; (2) a significant decrease in the number of fixation shifts between the central and incidental stimuli as the age increased. Such results seemed to imply that sufficient attention allocation did not appear below 2nd grade. It was also observed that even the 6th graders who fixated least on incidental stimuli, still fixated at least twice on incidental stimuli. Therefore, to assume that some information processing entering the memory process had occurred during the fixation period was relevant. Susk observance implied that the recall performance measure alone did not completely reflect the attentional processes in the entering period. In Exp. 2, the recognition scores of the central and incidental stimuli of another group of subjects were compared with the obtained scores in Exp. 1. In the recognition test, all subjects in three age groups showed more than 50% of correct responses to the incidental stimuli. This result supported the assumption made in Exp. 1. Through the above experiments, it was suggested that information selection was carried out in two processing stages: first, attention was devoted to the input stimuli, then followed by selective memorization.