This paper reviews the classical theory and new findings on the properties of language networks at the representational level of linguistic information (semantic, lexical, and phonological representations), as revealed by cognitive psychology, psycholinguistics, and cognitive neuropsychology. There are several speech therapy methods based on these theoretical backgrounds for aphasia with word finding difficulties. We report on some of our cases using Semantic Feature Analysis (SFA) and Phonological Components Analysis (PCA), and we describe a treatment method that makes active use of language networks. Our findings suggest that SFA treatment allows smooth word production because explicit recall of semantic features facilitates lexical access from the semantic network. We speculate that PCA treatment may be able to improve phonological encoding, which is important for lexical access, by taking advantage of lexical-phonological networks. However, there is disagreement as to whether, and which, factors lead to a generalizing effect of the two treatments in improving the production of untrained words. It is also unclear in which kinds of cases these treatments will be effective. We need to continue our studies to clarify these issues in the future.

It has been suggested that the demographic background, clinical characteristics, physical motor function, and cognitive function at admission of patients with dysphagia due to cerebrovascular disease affect the availability of oral intake at discharge. However, the factors predicting the availability of oral intake at discharge are still unknown.
In the present study, patients were allocated to two groups according to the presence or absence of tube feeding at discharge. Factors related to oral intake in patients with dysphagia were then compared by using the following factors: number of days from disease onset to hospitalization, length of hospital stay, outcome, functional independence measure (FIM) motor scores at admission and discharge, FIM cognitive scores at admission and discharge, and FIM gain scores.
Univariate analysis showed significant between-group differences in the number of days from disease onset to hospitalization, length of hospital stay, FIM score at admission, FIM score at discharge, and FIM motor gain score. Logistic regression analysis was performed by using the presence or absence of tube feeding at discharge as the dependent variable and age, gender, number of days from onset to admission, FIM motor score at admission, and FIM cognitive score at admission as independent variables.
FIM motor score and FIM cognitive score at admission were extracted as significant factors predicting the availability of oral intake. The results of this study suggest that FIM motor score and FIM cognitive score at admission are predictors of the likelihood of oral intake at discharge.

By using the same series of pictures, we examined the development of narrative speech in the preschool and school years in five children with ASD without intellectual disability who had received language instruction in early childhood. We compared the 1) linguistic form, 2) content, and 3) story structure of utterances, and we examined the development of the children during school age and any remaining issues. The results showed that 1) language forms such as MLU (mean length of utterance) corresponded to typical development in early childhood; 2) in the content of utterances, the number of causal expressions increased and developmental changes were observed in narratives related to emotions; 3) in terms of story structure, although there was development from enumerative reaction sequences to a simple episodic structure, there remained issues such as composition into a cohesive structure using narrative grammar and cases where the narrative remained enumerative. Our findings suggest that, in supporting the language development of children with ASD, it is important to provide continuous guidance from early childhood to school age with regard to teaching to achieve a rich narrative structure such as the structure of narrative grammar.

The purpose of this study was to observe laryngeal movement during various vocalizations in healthy subjects by using a stretch strain sensor (B4S^TM), which can be used to visualize laryngeal elevation movements in real time.
The subjects were 35 students at the Prefectural University of Hiroshima who had no audible hoarseness. There were 15 male and 20 female participants. The B4S^TM was pressed against the front of the neck and the waveform of the stretch strain sensor was measured during vocalization. The vocalizations were: 1) easy, natural voice; 2) loud/soft voice; 3) high-pitched/low-pitched vocalization; and 4) falsetto vocalization. The results showed that men had a greater degree of laryngeal elevation than women, and there were gender differences in laryngeal movement during vocalization. With changes in voice pitch, laryngeal elevation ranged from highest to lowest in both men and women in the following order: falsetto, high voice, grounded voice, and low voice. With changes in voice volume, the observed changes in laryngeal elevation were less marked than those with changes in the pitch. The B4S^TM was a useful device for confirming laryngeal movement during vocalization.

This study investigated the mechanics of the human pharynx during swallowing and speech production in two healthy volunteers by using lateral X-ray fluoroscopy. Upward movement of the pharyngeal wall by contraction of the stylopharyngeal muscle was observed by chasing radiologic contrast medium injected submucosally into the posterior pharyngeal wall. Upon constrictive movement of the pharynx by contraction of the pharyngeal constrictor muscles, we measured the distance from the posterior surface of the pharyngeal wall to the anterior surface of the 2nd to 4th cervical vertebrae, as well as the width of the orifice of the cervical esophagus. During swallowing, upward movement of the pharyngeal wall was seen initially. Just before the wall reached its highest position, the orifice of the cervical esophagus began to open, and just after the wall had reached its highest position the pharyngeal cavity was constricted in the region from the 2nd to the 4th cervical vertebrae. The orifice of the cervical esophagus then closed. These movement patterns of the pharynx during swallowing were highly reproducible in each subject, suggesting that this steady mechanical system was controlled by the central pattern generator of the medulla. Upward movement of the pharyngeal wall was also apparent during speech production. During the production of high-front vowels, the distance of upward movement of the pharyngeal wall was greater than that during the production of low-back vowels. In contrast, during the production of high-front vowels, the distance from the posterior surface of the pharyngeal wall to the anterior surface of the 2nd cervical vertebra was smaller than that during the production of low-back vowels. During the production of voiced consonants, the distance of upward movement of the pharyngeal wall was greater than that during the production of vowels. Therefore, it can be considered that, in speech production, the pharynx contributes to adjust the shape of the pharyngeal cavity and acoustical features and to maintain the continuity of running speech.

Aphonia following cerebral hemorrhage is commonly seen in stroke units. Its origin has only been speculated to be functional, and some patients remain aphonic in the recovery phase of stroke. We present a case of aphonia after stroke, and we show a novel technique of percutaneous tracheal compression that led to the effective gaining of voice production. A 42-year-old right-handed man with no previous history of dysphonia was admitted to our hospital upon the onset of subcortical hemorrhage in the area of the left frontal lobe. He was mute more than 1 month after craniotomy for hematoma removal. The patient's attempts at vocalization and speaking were observed in the stroke ward, but his aphonia persisted. He was referred to a speech clinician for speech-language evaluation, and steady airflow for vocalization attempts and free of the speech organs from paralysis were confirmed. Speech-language disorders such as dysarthria, aphasia, and apraxia were denied. Endoscopic examination of the larynx revealed that vocal fold adduction coordinated with airflow was not achieved despite the absence of organic abnormalities such as vocal fold paralysis and edema. Production of sustained vowels was requested under the speech clinician's percutaneous tracheal compression. After 4 days of voice training sessions, the patient consistently produced voiced sounds without tracheal compression. Aphonia after stroke may result from reduced respiratory-laryngeal coordination and lack of voluntary voicing control. The tracheal compression technique resembles Kayser-Gutzmann manual laryngeal compression for optimizing glottal adjustment, and it can be used in dysphonic patients with aphonia or hypophonia due to functional glottal closure insufficiency.