Abstract
Two experiments were conducted to examine the following three factors determining the effects of verbal labelling on the concept identification: Exp. I for the factor of familiarity to label, Exp. II for the factors of the variety of stimulus dimension to be labelled and the compatibility between sensory impressions of the label and stimulus features.
In Exp. I, 90 second-grade children were divided into three groups. The Ss in Group 1 learned to associate familiar words (e. g. IE, MICHI) with the corresponding stimulus figures. These were nonsense figures which did not contain any dimensional features. The Ss in Group 2 learned to associate unfamiliar words (e. g. KAOKU, GAIRO) with the same stimulus as used in Group 1. The Ss in Group 3 (control group) were required to count the number every time when the nonsense figures were presented. Immediately after the task mentioned above, all Ss were required to identify the nonsense figures on the basis of the verbal cues acquired in the labelling task, and then reported verbal introspectations.
In Exp. II, 80 third-grade children took part and each S was assigned to one of the eight conditions. The experimental method in Exp. II was the same as in Exp. I, except the stimulus materials. The stimulus materials consisted of non sense figures varied with two colors (red and blue) and with two forms (round and angular). The labelling words consisted of familiar common nouns which were classified in terms of sensory impressions, i. e,“post” for red,“sky” for blue,“egg” for round and “horn” for angular. The 2×2×2 factorial design was used: two tasks (color or form identification task), two kinds of the stimulus dimension to be labelled (relevant or irrelevant dimension of the identification task) and two compatibilities between the sensory impressions of the label and attributs of stimulus figure (compatible or unrelated).
The main results were as follows: (a) No significant differences in the identification performance were found between familiar label and unfamiliar label conditions (Exp. I), and between compatible label and unrelated label conditions (Exp. II).(b) It was shown, however, that the labelling procedures for the relevant dimension improved the identification performance more than for irrelevant dimension (Exp. II).(c) Further, the familiarity to labels increased the introspective reports of labels learned (Exp. I). The compatibility between labels and figures had effects on the speed of labelling learning (Exp. II).
These results were discussed with reference to the three stages of conceptual behavior.
