Language technologies applied to document simplification for helping autistic people

People with autism spectrum disorder (ASD) experience reading comprehension difficulties, often misinterpreting complex texts, metaphors, and idioms. We have developed and tested a new assistive technology tool for adaptive, personalized text simplification, called Open Book. This tool is an open-sourced, online platform that uses Natural Language Processing with the specific aim of assisting reading and aiding understanding of written text for people with ASD. The accessibility and effectiveness of Open Book was tested by examining the differences in text comprehension scores between the original texts and texts that were simplified by Open Book tool, randomly allocated to study participants. Two hundred forty-three participants (153 adults and 90 adolescents) with high-functioning ASD were recruited in the UK, Spain, and Bulgaria. Regarding the primary outcome, results showed that both adults and adolescents with ASD gave more correct answers for the simplified (M = 11.2, SD = 4.1) than original texts (M = 10, SD = 4.1; p < 0.001). This finding was consistent across age groups and countries. Regarding the secondary outcome, when participants were asked to blindly rate how easy was to understand each text, simplified texts were rated as easier (M = 7.6, SD = 2.4) to understand than the original texts (M = 8.7, SD = 2.6; p < 0.001). The Open Book software seems to have the potential to be a useful tool in assisting reading among people with ASD. Our findings support our primary hypothesis that texts simplified through Open Book were easier to comprehend compared to original texts.

Universidade Federal de Sao Carlos (UFSCar), Dept Letras, BR-13565905 Sao Carlos, SP, Brazil

Reading comprehension is one of the most important skills learned in school and it has an important contribution to the academic success of children with Autism Spectrum Disorder (ASD). Though previous studies have investigated reading comprehension difficulties in ASD and highlighted factors that contribute to these difficulties, this evidence has mainly stemmed from children with ASD and intact cognitive skills. Also, much emphasis has been placed on the relation between reading comprehension and word recognition skills, while the role of other skills, including fluency and morphosyntax, remains underexplored. This study addresses these gaps by investigating reading comprehension in two groups of school-aged children with ASD, one with intact and one with low cognitive abilities, also exploring the roles of word decoding, fluency and morphosyntax in each group's reading comprehension performance. The study recruited 16 children with ASD and low cognitive abilities, and 22 age-matched children with ASD and intact cognitive skills. The children were assessed on four reading subdomains, namely, decoding, fluency, morphosyntax, and reading comprehension. The children with ASD and low cognitive abilities scored significantly lower than their peers with intact cognitive abilities in all reading subdomains, except for decoding, verb production and compound word formation. Regression analyses showed that reading comprehension in the group with ASD and intact cognitive abilities was independently driven by their decoding and fluency skills, and to a lesser extent, by morphosyntax. On the other hand, the children with ASD and low cognitive abilities mainly drew on their decoding, and to a lesser extent, their morphosyntactic skills to perform in reading comprehension. The results suggest that reading comprehension was more strongly affected in the children with ASD and low cognitive abilities as compared to those with intact cognitive skills. About half of the children with ASD and intact cognitive skills also exhibited mild-to-moderate reading comprehension difficulties, further implying that ASD may influence reading comprehension regardless of cognitive functioning. Finally, strengths in decoding seemed to predominantly drive cognitively-impaired children's reading performance, while the group with ASD and intact cognitive skills mainly recruited fluency and metalinguistic lexical skills to cope with reading comprehension demands, further suggesting that metalinguistic awareness may be a viable way to enhance reading comprehension in ASD.

Machine learning techniques have long been the foundations of speech processing. Bayesian classiflcation, decision trees, unsupervised clustering, the EM algorithm, maximum entropy, etc. are all part of existing speech recognition systems. The success of statistical speech recognition has led to the rise of statistical and empirical methods in natural language processing. Indeed, many of the machine learning techniques used in language processing, from statistical part-of-speech tagging to the noisy channel model for machine translation have roots in work conducted in the speech fleld. However, advances in learning theory and algorithmic machine learning approaches in recent years have led to signiflcant changes in the direction and emphasis of the statistical and learning centered research in natural language processing and made a mark on natural language and speech processing. Approaches such as memory based learning, a range of linear classiflers such as Boosting, SVMs and SNoW and others have been successfully applied to a broad range of natural language problems, and these now inspire new research in speech retrieval and recognition. We have seen an increasingly close collaboration between voice and language processing researchers in some of the shared tasks such as spontaneous speech recognition and understanding, voice data information extraction, and machine translation. The purpose of this special issue was to invite speech and language researchers to communicate with each other, and with the machine learning community on the latest machine learning advances in their work. The call for papers was met with great enthusiasm from the speech and natural language community. Thirty six submissions were received; each paper was reviewed by at least three reviewers. Only ten papers were selected re∞ecting not only some of the best work on machine learning in the areas of natural language and spoken language processing but also what we view as a collection of papers that represent current trends in these areas of research both from the perspective of

Introduction Autism spectrum disorders (ASD) are a group of neurodevelopmental conditions characterized by impairments in social interaction, communication (both verbal and non-verbal), and repetitive behaviors. The global prevalence of ASD has increased significantly, with an estimated 28.3 million cases worldwide. Although many individuals with ASD have normal cognitive and language skills, difficulties with social interactions and understanding nonverbal cues can interfere with their ability to form romantic and sexual relationships, potentially leading to inappropriate behaviors and a distorted experience of sexuality. Objectives This paper aims to review the literature on sexuality in individuals with ASD, focusing on typical sexual behaviors, sexual preferences, as well as hypersexuality and paraphilic fantasies and behaviors within this population. Methods A non-systematic literature review was conducted, with article selection from PubMed using the keywords: “autism spectrum disorders”, “sexuality”, “hypersexuality” and “paraphilia”. Results Studies have shown that adolescents with ASD experience higher rates of inappropriate sexual behaviors and gender dysphoria compared to neurotypical peers. Variants in sexual orientation, including homosexuality, asexuality, and bisexuality, are more prevalent in this population. Recent research indicates that sexual experiences, both alone and with others, are common among individuals with high-functioning ASD, with one study revealing that 47% expressed interest in having a romantic partner. Despite this, adults with ASD, especially men, are generally less likely to be in romantic relationships. They also exhibit more hypersexual and paraphilic fantasies and behaviors than neurotypical individuals. Hypersexual behaviors are predominantly observed among male ASD individuals, while paraphilias, such as voyeurism and fetishism, are frequently reported among both ASD men and women. Sadistic and masochistic fantasies and behaviors are also common in this group. Conclusions The results indicate that individuals with ASD have a higher prevalence of sexual orientation variants and inappropriate behaviors compared to their peers. Although they have sexual interests and desires for relationships, their ability to express sexuality in a healthy way is hindered by communication deficits, social interaction challenges, and an unsupportive environment, exacerbated by inadequate sexual education. These challenges prevent many from fully embracing their sexuality. Specialized, evidence-based sexual education addressing the unique needs of individuals with ASD is essential for promoting healthier sexual behaviors and improving psychosocial well-being. Disclosure of Interest None Declared

Through the eyes of autism: Eye care for these special patients

Graph theory and the fields of natural language processing and information retrieval are well-studied disciplines. Traditionally, these areas have been perceived as distinct, with different algorithms, different applications and different potential end-users. However, recent research has shown that these disciplines are intimately connected, with a large variety of natural language processing and information retrieval applications finding efficient solutions within graph-theoretical frameworks. This book extensively covers the use of graph-based algorithms for natural language processing and information retrieval. It brings together topics as diverse as lexical semantics, text summarization, text mining, ontology construction, text classification and information retrieval, which are connected by the common underlying theme of the use of graph-theoretical methods for text and information processing tasks. Readers will come away with a firm understanding of the major methods and applications in natural language processing and information retrieval that rely on graph-based representations and algorithms.

Background Abstract The Centers for Disease Control and Prevention (CDC) in the United States has found that more children than ever before are being classified as having autism spectrum disorders (ASDs) (Chiri and Warfield 2011: 1081). According to Matenge (2014: 1), ASD is a pervasive developmental disorder characterised by two essential features: a combination of impairments in social interaction and social communication, and restricted, repetitive, and stereotyped patterns of behaviour, interests and activities. Malcolm-Smith et al. (2013: 4) report that data on the incidence, prevalence, and impact of ASD in South Africa are almost entirely lacking. No epidemiological studies of ASD have been conducted in the country (Malcolm- Smith et al. 2013: 4). Diagnostic and intervention services, particularly at state level, are scarce (Malcolm-Smith et al. 2013: 4). Hence, those that are in place are heavily overburdened. Of particular concern for South Africa are research findings indicating that ASD is under-identified in low socioeconomic status communities, so in these contexts ASD often goes undiagnosed, or is diagnosed late (Malcolm-Smith et al. 2013: 4). Matenge (2014: 19) argues that there is, therefore, a need to assess the level of knowledge, understanding and the experiences of South African nurses regarding autism, as they are usually the first professionals that families of children with autism contact. Research purpose The purpose of this research was to determine the knowledge, understanding and perceptions of professional nurses who work in primary health clinics in eThekwini Health District, KwaZulu-Natal, regarding ASD. Research design A quantitative, non-experimental, descriptive research survey design was utilised. A self-reporting questionnaire was used to collect data. Sampling and sample size Simple random sampling was used to determine which clinics in the eThekwini health district would be part of the study. The sample size for the clinics was thirty four, of those thirty were eThekwini municipality clinics and four KwaZulu-Natal Department of Health clinics. These clinics were utilised and provided the 275 professional nurses needed. The sample was calculated with a margin of error and an alpha value of 0.05; the researcher needed a minimum of 275 respondents. The researcher selected the respondents utilising non-probability, convenience sampling. Data analysis Data was analysed with the assistance of the statistician, using SPSS version 22. Descriptive statistics were used to explore differences between or among groups (Grove, Burns and Gray 2013: 217). Inferential statistics are used to draw conclusions regarding the differences between groups in the population from which the samples are drawn and the relationship between the variables in the population (Morgan et al. 2015: 84). Both descriptive and inferential statistics were used to analyse data in this study. Conclusion and findings It was found that the average professional nurse working in a primary healthcare clinic has 65.8% knowledge and understanding of ASD. Of particular concern is that the majority of respondents 51% (n=141) did not know or understand that the child with autism loves routine which is a key symptom in making a diagnosis of ASD, which may lead to late, or missed diagnosis. Eighty eight percent (n = 242) of respondents felt that they would benefit from further training on the identification and diagnosis of ASD. It was also found that respondents that have completed the Psychiatric Nursing Science course were more knowledgeable about ASD and confident in diagnosing ASD and counselling parents on ASD and available ASD services in the community. Recommendations The researcher recommended that: Autism spectrum disorder to be included in the curriculum for nurses training and that regular in-service training be done to keep the professional nurses up to date with on new information related to ASD. It was also recommended that all children should be screened for ASD at 18 and 24 months at all PHCs in eThekwini district and that posters on the signs and symptoms of ASD should be displayed at primary healthcare clinics in full view of patients, parents and staff

Intranasal Oxytocin Improves Emotion Recognition for Youth with Autism Spectrum Disorders

Objective: Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by impairments in social interactions and behavioral problems. Various genetic and environmental factors, including oxidative stress, are claimed to play a role in the etiopathogenesis of ASD. In this study, we aimed to examine the status of oxidative metabolism in ASD and the association between oxidative parameters and ASD symptom severity and subtype of ASD. Method: Thirty-three children and adolescents diagnosed with ASD (16 children diagnosed with autistic disorder, 13 children with pervasive developmental disorder not otherwise specified, and 4 children with Asperger syndrome) according to Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR) and 28 healthy controls, aged 2-17 years, were recruited in this study. Total oxidant status (TOS) and total antioxidant status (TAS) were evaluated using Rel Assay Kit in children and adolescents. The oxidative stress index (OSI) was calculated by dividing the TOS values by the TAS values. Autistic symptoms for these patients were scored on the Childhood Autism Rating Scale (CARS). Results: In patients with ASD, TAS was statistically significantly lower and OSI statistically significantly higher than in healthy controls. There were no statistically significant differences in TOS between the ASD and control groups. There were no statistically significant differences between the subtypes of ASD in terms of oxidative stress parameters. In addition, TAS, TOS, and OSI values did not differ statistically significantly between the patients’ CARS groups, and were not associated with the CARS scores of the patients. Conclusion: Our findings suggest that oxidative imbalance is present in ASD and that oxidative stress may play a role in the etiopathogenesis of ASD. Therefore, it is suggested that antioxidants may have beneficial effects on ASD and may be a new therapeutic target in treating ASD.

Abstractive single document summarization is considered as a challenging problem in the field of artificial intelligence and natural language processing. Meanwhile and specifically in the last two years, several deep learning summarization approaches were proposed that once again attracted the attention of researchers to this field.It is a well-known issue that deep learning approaches do not work well with small amounts of data. With some exceptions, this is, unfortunately, the case for most of the datasets available for the summarization task. Besides this problem, it should be considered that phonetic, morphological, semantic and syntactic features of the language are constantly changing over the time and unfortunately most of the summarization corpora are constructed from old resources. Another problem is the language of the corpora. Not only in the summarization field, but also in other fields of natural language processing, most of the corpora are only available in English. In addition to the above problems, license terms, and fees of the corpora are obstacles that prevent many academics and specifically non-academics from accessing these data.This work describes an open source framework to create an extendable multilingual corpus for abstractive single document summarization that addresses the above-mentioned problems. We describe a tool consisted of a scalable crawler and a centralized key-value store database to construct a corpus of an arbitrary size using a news aggregator service.

Abstract. Large language models (LLMs) have revolutionized the field of natural language processing (NLP), demonstrating remarkable capabilities in understanding, generating, and manipulating human language. This comprehensive review explores the development, applications, optimizations, and challenges of LLMs. This paper begin by tracing the evolution of these models and their foundational architectures, such as the Transformer, GPT, and BERT. We then delve into the applications of LLMs in natural language understanding tasks, including sentiment analysis, named entity recognition, question answering, and text summarization, highlighting real-world use cases. Next, we examine the role of LLMs in natural language generation, covering areas such as content creation, language translation, personalized recommendations, and automated responses. We further discuss LLM applications in other NLP tasks like text style transfer, text correction, and language model pre-training. Subsequently, we explore techniques for optimizing and improving LLMs, including model compression, explainability, robustness, and security. Finally, we address the challenges posed by the significant computational requirements, sample inefficiency, and ethical considerations surrounding LLMs. We conclude by discussing potential future research directions, such as efficient architectures, few-shot learning, bias mitigation, and privacy-preserving techniques, which will shape the ongoing development and responsible deployment of LLMs in NLP.

Mitochondrial dysfunction and mitochondrial unfolded protein response (UPRmt): unravelling their roles in autism spectrum disorder pathogenesis.

Background and AimsChildren with autism spectrum disorder (ASD) present with impairments in social interaction and stereotypic behaviors. About a third may exhibit delays in verbal expression beyond preschool age, potentially categorizing them as nonverbal/minimally verbal (NV/MV), a condition that can persist into adulthood and affect their quality of life. The risk and prognostic factors associated with this reduced verbal outcome remain uncertain. This study aims to identify such factors within children diagnosed with ASD in Cyprus.MethodsIn this case-control study, 56 children aged 3–12 years, with an ASD diagnosis, participated. Among them, cases were 22 children classified as ASD-NV/MV, and controls were 34 children classified as verbal (ASD-V), matched by age group and gender. Retrospective information on familial, perinatal, and developmental risk and prognostic factors were collected to calculate the familial risk score (FRS), perinatal risk score (PRS), and developmental risk score (DRS). Early development information was collected for the Early Development Score (EDS) and Early Gesture Score (EGS), to measure the children's skill level as toddlers across milestones. A low EDS and/or low EGS reflected general developmental delays and decreased frequency of early gestures and were considered in the DRS. A parent report questionnaire was utilized to determine the current overall linguistic level and status of participants, distinguishing cases from controls.ResultsAge group and gender-matched cases and controls were similar in socioeconomic status and demographic characteristics (p > .05). Among the various familial (e.g., sibling with ASD), perinatal (e.g., prematurity), and developmental (e.g., ASD regression) factors examined individually, as well as collectively as scores in a conditional logistic regression (CLR) model, only a high DRS (p = .03), due to low EDS (p = .04) was significantly associated with linguistic status. When considering all risk scores in a multivariate CLR model, children with a high DRS were more likely to belong to the cases than to the control group (p = .02). In a subsequent model with low EDS and low EGS, only the low EDS was significantly associated with the case group. Results showed that children with ASD and a low EDS, reflecting general delays in early development, were 4.5 times more likely to belong to the cases group than those with a high EDS (p = .02).ConclusionsEarly developmental delays in developmental milestones across various domains like gesture, motor, play, linguistic, cognition, and joint attention, in toddlerhood, were associated with later decreased verbal outcomes. Children in the sample with such early delays (low EDS), had a higher likelihood of persistent language delays (ASD-NV/MV) even at late school age. Future studies are needed to duplicate findings and explore possible contributing factors affecting linguistic outcome in ASD through prospective studies exploring within ASD differences.ImplicationsThese findings emphasized the importance of monitoring early development closely for children at risk for ASD, pre-diagnosis, to provide additional early support for those more likely to be ASD-NV/MV. Parents and specialists like pediatricians, educators, speech-language pathologists, among others, can track the EDS score of children at risk for ASD and refer to the appropriate specialists for early stimulation, intervention, and parent consultation promptly.

Autism Spectrum Disorder (ASD) is a developmental condition that causes lifelong impairments in social interaction and communication skills.Besides early interventions and ASD services, technology has become essential in assisting ASD individuals in building and improving their communication and social skills.However, most technological innovations have been developed for Western countries.When considering Saudi Arabia as a developing country, people with ASD may have to deal with other difficulties, such as culture, privacy, and societal integration.To address this, we developed and evaluated a framework to investigate the factors that most influence the adoption of ASD Learning Applications (LA).Additionally, 16 hypotheses were defined to test specific relationships between the factors.A quantitative method was utilised to design our experimental study.This paper presents an evaluation of the Autism Spectrum Disorder Learning Application Framework (ASDLAF) from the perspective of individuals with ASD in Saudi Arabia.Utilising a quantitative approach, data was gathered through a survey with a total of 395 participants.Results of hypotheses testing revealed significant support for most factors except social rules (SR), religion (RE), and age (AG).These findings underscore the importance of cultural, technological, and pedagogical considerations in technology adoption for ASD individuals.The results also suggest that while the ASDLAF may be effective in addressing certain factors of ASD learning applications, there are still areas that need to be further developed to meet the specific cultural and religious needs of individuals in Saudi Arabia.Furthermore, it will be essential for developers and stakeholders to consider these factors in order to create more beneficial and effective learning tools for individuals with ASD in this region.