Early research indicated that much of the natural speech signal was redundant and that the duration of certain phonemes in normal speech was longer that required for reliable recognition. So, especially on the visually handicapped, speech compression was studied in order to transmit much informations effectively through the auditory system. Speech compression was one of the communication methods in which brief segments of the speech signal was removed without significant distortion in vocal pitch or quality. But there were some problems awaiting solution. First, usually speech compression was produced by a speech compressor. This apparatus, however, was not popularized in Japan. So we were not able to produce speech compression with ease. Second, the speech compressor removed segments of speech from original materials at regular intervals (indiscriminate compression). But selective compression was required to compress certain parts of materials. The purpose of this study was to establish the system which produced time compression of speech by means of a microcomputer and to test the effects of speech compression on the comprehension of speech signals in the visually handicapped. Speech compression was produced in the following manner. Recorded natural speech signal was converted from analog to digital values. These were sampled at 20kHz, digitized to 8 bit (one sign bit) samples. Then these samples were processed on the CPU of a microcomputer and data processed was passed through digital-to-analog converter to subjects who heard the compressed signals with a headphone. Sampling period was 100msec. and compression ratios used in this experiment were 40, 50 and 60%. Materials presented to subjects were 63 questions which were prepared and recorded by the experimenter. Each question was presented within about 3.3 msec. The subjects were 15 visually handicapped students and were required to answer the questions with yes or no or names which were presented in the questions. The results obtained were as follows: 1) mean correct responses reduced as the ratios of speech compression increased. Especially when the compression was 60%, the score dropped off sharply. 2) with the 60% compression, correct responses increased as the number of presentations increased. Based on the results mentioned above, the application of the speech compression by means of the microcomputer was discussed in relation to the ability of the information processing of the auditory system.

View full abstract

The present study was conducted to estimate the size of the deaf pupils' recognition vocabulary by means of new vocabulary test for deaf pupils. The test consisted of 50 items, each of which included one meaningful sentence and three meaningless sentences in their random ordering. Subjects were 288 pupils with the severe hearing loss and were instructed to find out the meaningful sentence in each of 50 items. The obtained results are summarized as follows: (1) Scores of 43% of the deaf pupils were lower than the minimum score, i, e., the chance level. (2) Except for these pupils, mean percentages of correct response were 47.03% in junior group, 58.04% in junior high group, and 62.60% in senior high group. (3) When the percentages of correct response for each item were compared among the three subgroups, the patterns of item difficulty were similar for all groups; the coefficient of concordance for these groups was .874. (4) Within each group the percentages of correct response for each item were related to the rating system of the basic vocabulary, the form classes (parts of speech), and the contents. (5) The reliability coefficient was .983, using Kuder - Richardson (Formula 20) internal consistency coefficient. The structure of the deaf pupils' recognition vocabulary and the usefulness of our test were discussed.

View full abstract

Gait studies were performed on 13 children with mental retardation between the ages of 4 and 18 years. Walking speed; step length, and step length/shin length ratio; duration of stance and double-limb stance (as percentage of the gait cycle); and rotations of the lower-extremity joints in the sagittal plane were analyzed throuout a walking cycle, and were evaluated. 8 subjects were diagnosed as a group of the abnormal gait, and their patterns of the sagittal-plane angular rotations were similar to those of normal infants. Most of the group remained whithin the 2nd year in mental development, and the rest, who were over the 2nd year, had overt motor disorders.

View full abstract

The purpose of this study was to establish the acquisition phase and discuss the acquisition process of "GA" and "O". The subjects were 80 retarded children and 80 normal children with a vocabulary age between 4:0 and 7:11. In order to test the subjects' ability to use JOSHI, 33 incomplete sentences without JOSHI were presented to them using a taperecorder and picture cards. Then, they were asked to repeat while completing the sentence after listening to each incomplete sentence. The results of this experiment were summarized as follows: 1) As to the acquisition of "GA", same sentence types were mistaken in both subject groups, although the acquisition phases were different respectively. 2) The mentally retarded had great difficulty in especially one kind of these sentences. It was suggested that the factor related to this difficulty was the strength of connection between noun and JOSHI in the mentally retarded. 3) As to the acquisition of "O", there was no difference between the two groups. Therefore, it seemed that the two groups manifested same acquisition process.

View full abstract

The experimenter administered the Handedness Inventory, Finger Tapping Test, Finger Identification Test and Tactile Form Recognition Test to 20 learning disabled children, 10 males and 10 females, and to 38 normal control children, 19 males and 19 females. Their age ranged from 9 to 15 years old. The results of the Handedness Inventory showed that 8 (40%) of the learning disabled children were right handers, 8 (40%) were left handers. On the other hand, 32 (84%) of the control children were right handers, 4 (11%) were mixed handers, and 2 (5%) were left handers. The distribution of handedness differed significantly between the learnig disabled and normal children (x^2=119.34, df=2, p<.001). That is, 60% of the learning disabled were non-right handers whereas 16% of the normal were non-right handers. The scores of the left and right hand on the Finger Tapping Test were as follows: 82.6 and 88.1 (male learning disabled children), 78.4 and 98.0 (female learning disabled children). 106.2 and 113.8 (male normal children), 103.4 and 118.6 (female normal children). These results suggest that the performance of the learning disabled children were significantly inferior to those of the normal children (F (1, 48)=16.46, p<.001), and the right hand performance was superior to the left hand in both groups of the children (F (1, 48)=24.44, p<.001). In the Finger Identification Test, the mean percentages of correct responses for the left and right hand were as follows: 46% and 41% (male learning disabled children), 48% and 36% (female learning disabled children), 69% and 77% (male normal children), 78% and 82% (female normal children). These results indicate that the accuracy of the Finger Identification was significanly different between the learning disabled and the normal control children (F (1, 48)=52.95, p<.001). The right hand performance was superior to that of the left hand in the normal children, whereas the left hand performance was superior to that of the right hand in the learning disabled children. In the Tactile Form Recognition Test, the normal control children's performance was superior to that of the learning disabled children (F (1, 48)=65.63, p<.001). The results of the above four measures suggest that the learning disabled children may have intra-hemispheric dysfunction and/or deficits of inter-hemispheric integration, whereas the normal children have established the cerebral specialization and may use both hemisphere's functions synergistically.

View full abstract

The purpose of present study was to investigate the understanding of vocabulary in the word classification task and the word comprehension task in hearing impaired children. Subjects were 123 hearing impaired children, aged from 9 to 16 and 119 hearing children, aged from 6 to 12. Subjects were asked to choice a correct item among items. This experiment consisted of two tasks; word classification and word comprehension. In the word classification task, the hearing impaired were inferior to the hearing in correct response. Also, in the word comprehension task, the hearing impaired were inferior to the hearing. In examining the performance of the word classification basing on the perfomance of the word comprehension, both the hearing impaired and the hearing decreased the response based on the form and the number of HIRAGANA and increased the response based on the meaningful selection among the items. These results were suggested that the poor in vocabulary did not understand the correct meaning of the word but tend to depend on the form of the word. In the further teaching language, the teacher should enrich the foundation of the language, at the sam time, expand the vocabulary noticing the meaningful function among the words.

View full abstract