日本語近似系列の知覚と再生

抄録

The purpose of this study was to investigate psychological characteristics of the series of approximation to Japanese. Two kinds of the series of approximation were experimentally made by the method which was quite similar to that of Shannon-Miller's guessing technique. The orders of both syllable and word-approximations were 0th, 2nd, 3rd, 4th, 5th, 6th, 7th, and text series. The text series means a series of meaningful syllables or words quoted from magazines and novels.
Experiment 1. Syllable-approximation and intelligibility.
(A) Procedure: 12 undergraduate students were asked to listen to the auditory stimuli which were given under the masking white noise at the rate of one syllable per second. Five different orders of approximation which consisted of 20 syllables were used as the stimuli. The level of noise was controlled. The S-N ratios were -8, -4, 0, +4, +8, in db.
(B) Results: The intelligibility of syllables under the condition of the masking tends to increase with the ascending order of approximation (p<0.01; See Fig. 1).
Experiment 2. Syllable-approximation and span of attention.
(A): 8 students responsed to the visual stimuli which were presented tachistoscopically. The stimuli were 8 typewritten Japanese letters which were placed in a line vertically or horizontally.
(B): The span of attention tends to extend with the ascending order of approximation (p<0.01; see Table 2).
Experiment 3. word-approximation and immediate recall.
(A): 20 students were asked to recall the word material which had been presented in audition. The material consisted of groups of 6, 10, 20 and 50 words.
(B): The rate of recall of words increases in accordance with the order of approximation (p<0.01) and this tendency is almost equal to Miller-Selfridge's data (see Fig. 2).
Experiment 4. Syllable-approximation and immediate recall.
(A): The same condition as that of Exp. 3 except the material being syllables.
(B): The rate of recall of syllables increases in accordance with the order of approximation (p<0.01; see Fig. 3).
Experiment 5. Syllable-approximation visually presented and immediate recall.
(A): 17 students were asked to recall the letters which had been presented by a film-strip projector at the speed of 2 letters per second.
(B): So far as this experimental condition is concerned there is no significant difference between the auditory and the visual channel through which the messages were conveyed (p>0.99; see Fig. 4).
Experiment 6. Syllable-approximation and eye-voice span.
(A): 10 students were asked to read the material consisting of 24 letters. While the subject was reading aloud, the light was suddenly extinguished; he was required to continue saying as many which he had seen as possible.
(B): The eye-voice span expands in accordance with the order of approximation (p<0.01; see Fig. 5).