Exploration And Applied Study On The Cognitive Mechanism Of Visual Semantic Understanding Of Chinese Children With Autism

Autism is a neurodevelopmental disorder characterized by lifelong socio-communicative impairments and a restricted range of behaviors, which can be diagnosed in infants and young children. Although there are many differences in autism, semantics-pragmatics is prominently impaired in these children. At present, most of researches on autism are focused on the autistic adults whose first language is English. However, little attention is paid to impaired language function in Chinese children with autism. Therefore, the experimental methods of cognitive psychology to systematically explore the mechanism of visual semantic processing in Chinese children with autism were executed and an effective way of early language intervention to autistic children is established in the paper.The study was divided into six parts.The Literature Review was given in the part I and part II, analyzing the principles, methods and conclusions of relevant researches at home and abroad. On this basis, the whole scheme of the study was designed. In the study, the degree of semantic understanding defects, the mode and characteristic of semantic understanding in Chinese children with autism by a combination of series of cognitive experiments with intervention experiment were all investigated. Furthermore, "visual semantic understanding", which is possibly the cognitive mechanism, was also to be validated and explored.The experiments were respectively designed and shown in the part III and part IV, The two parts were central parts of the whole study. Five experiments were designed step by step to explore the characteristic and cognitive mechanism of semantic understanding of Chinese children with autism.In order to investigate word understanding ability in Chinese children with autism, Experiment I adopted "Category Verification" procedure and then examined the effect of exemplar typicality on reaction time and accuracy of categorization in Chinese children with autism and matched controls. In compared with the matched controls, it is found for the children with autism that the response to typical exemplars was the same. However, the response to atypical exemplars was much worse. The results showed a pattern of intact and impaired semantic functions of word level in autism. In Experiment II, Chinese children with autism and matched controls were tested with a homograph reading task, in which pronunciation of target words was determined by integration of whole sentence context. At the same time, they were also tested with one-order and second-order false belief tasks. Then the relationship between theory of mind deficits and weak coherence was studied in the part. It was revealed that a relative failure in processing information for context-dependent meaning in the homograph reading task was represented in autistic children at all level theory of mind performance. As a result, there were the defects of sentence understanding in autistic children and the coexistence of the two types of defects (theory of mind deficits and weak coherence) in the same group. Experiment III was used to test paragraphs understanding ability in autism. The experiment presented scenarios which had a situation and outcome which only cohered if a bridging inference was drawn. It is found that the autistic children were less likely to select the most coherent (bridging) inference from competing alternatives. The results indicated that the autistic children are impaired in achieving coherence of context information. In Experiment IV, Chinese children with autism and matched controls were tested with "word fragment completion" task. The aim of the experiment was to compare the effect of pictures and words with respect to semantic priming effect in subjects. It is found for both groups that there were semantic priming effects in both conditions, which shows the unimpaired semantic function in autism to some extent in word level. The control children performed similarly on both priming tasks. However, the autistic children performed significantly better on a picture-word task than on a word-word task. So it is clear that there is the possible advantage of pictures over words in access to semantics in autism.In Experiment V, the aim of three sub-experiments was to test the levels-of-processing effect in different subjects, which suggests that semantic processing of verbal materials facilitates long term memory better than shallow (phonological or perceptual) processing. The levels-of-processing task was conducted in normal children, autistic children, mild mental retardation and ability-matched controls with different ages. Two main results were found. The levels-of-processing effect was not found in the two young groups (autistic group and controls group), which showed the perceptual processing, especially visuospatial operation, may be more effective in facilitating long-term memory than ordinary semantic processing in young children. In old groups, however, the situation was different. The levels-of-processing effect was confirmed in the old control group, which indicated that semantic processing was better to facilitate long-term memory than perceptual processing with the increase of age. But in old autistic group, it was still lack of the levels-of-processing effect. This result confirmed again the important role of superior perceptual processing in old autistic children.An interventional study was carried out in part V. On the basis of previous experimental researches, a multiple baseline case study was conducted in experiment VI in order to investigate the effectiveness of visual supporting language instruction on increasing social language skills of a Chinese child with high function autism. The results showed visual supporting instruction was effective in increasing the child's total verbalizations, mean length of utterance ( MLU ) , scripted/unscripted verbalizations, initiations and responses.In the part VI, synthetically analysis and summary of the basic and behavioral intervention study were carried out. Some important and interesting results were found.1. Chinese children with autism have relatively linguistic processing ability of simple lexicosemantics. However, their linguistic ability in processing complex semantics are severely impaired, such as lexical level, sentence level and paragraph level etc.2. In semantic processing, Chinese children with autism show prominent pictorial coding superiority. It seems to suggest a possibility that semantic processing in autistic children involved more visuospatial operations. Nevertheless, this strategy of semantic processing reflects a mechanical or immature pattern of semantic processing.3. The visual supporting instruction designed with the inspiration of basic researches shows high applicability. It is benefit to the imitative language and responsive language. In addition, the non-imitative language and initiative language of the autistic subject are also improved.4. From the point of view of cognitive channel, there are also bottom-up information processing deficits besides top-down in the linguistic functional impairments for autism.5. Both the "weak central coherence" (WCC) theory and "complex information processing defects" theory resulted from the research with English language system are further validated with the study of exploring the mechanism of semantic processing and the interferential research on the effectiveness of visual supporting instruction.