Deep learning-based school attendance prediction for autistic students.

Autism Spectrum Disorder is a neurodevelopmental disorder characterized by deficits in social communication and interaction as well as the presence of repetitive, restricted patterns of behavior, interests, or activities. Many autistic students experience difficulty with daily functioning at school and home. Given these difficulties, regular school attendance is a primary source for autistic students to receive an appropriate range of needed educational and therapeutic interventions. Moreover, school absenteeism (SA) is associated with negative consequences such as school drop-out. Therefore, early SA prediction would help school districts to intervene properly to ameliorate this issue. Due to its heterogeneity, autistic students show within-group differences concerning their SA. A comprehensive statistical analysis performed by the authors shows that the individual and demographic characteristics of the targeted population are not predictive factors of SA. So, we used the students' recent previous attendance to predict their future attendance. We introduce a deep learning-based framework for predicting short-and long-term SA of autistic students using the Long Short-Term Memory (LSTM) and Multilayer Perceptron (MLP) algorithms. The adopted algorithms outperform other machine learning algorithms. In detail, LSTM increased the accuracy and recall of short-term SA prediction by 20% and 13%, while the same scores of long-term SA prediction increased by 5% using MLP.

Development and Validation of the Assessment of Bullying Experiences Questionnaire for Neurodivergent Youth.

Bullying victimization is a prevalent concern for neurodivergent (e.g., autistic, ADHD) youth. Bullying assessment methods vary widely and there is currently no questionnaire specific to neurodivergent youth. The Assessment of Bullying Experiences (ABE) was created to fill this gap. The ABE questionnaire was completed by 335 parents of school-age youth characterized as autistic, having ADHD, or community comparison. Exploratory and Confirmatory Factor Analysis identified a four-factor solution, aligning with verbal, physical, relational, and cyber victimization. Construct validity analyses indicate the ABE converges with an existing bullying questionnaire and diverges from disruptive behavior or internalizing symptoms. The ABE questionnaire is a valid measure of bullying that furthers understanding of nuance in peer victimization for neurodivergent youth and informs group-specific intervention.

Brief Report: Predictors of School Refusal Due to Bullying in Children with Autism Spectrum Disorder and Attention-Deficit/Hyperactivity Disorder.

Children with Autism Spectrum Disorder (ASD) or Attention-Deficit/Hyperactivity Disorder (ADHD) are at increased risk for bullying victimization. School refusal is a 'red flag' for identification of bullying in children with ASD and/or ADHD. This study examined the impact of diagnoses, demographics, and school variables on school refusal due to bullying. Participants were 97 parents of 154 children with ASD, ADHD, ASD + ADHD, other diagnoses, or no diagnosis. Children with ASD + ADHD were most likely to refuse school due to bullying. Classroom aides and behavior problems were protective and risk factors, respectively. In the final regression model, child diagnosis no longer predicted school refusal. School refusal and problem behavior warrant consideration as a marker of distress for victimized children.

Conceptualizing bullying in children with autism spectrum disorder: Using a mixed model to differentiate behavior types and identify predictors.

Children with autism spectrum disorder experience bullying more frequently than their typical peers. Inconsistent definitions for and imprecise measurement of bullying in the literature impede a better understanding of this difference, and multiple types of bullying topographies create additional dimensions for analysis. In this study, participants rated the severity of bullying depicted in written vignettes of child-dyadic interactions. The vignettes varied across child age (4-15 years old) and described either one of four different types of bullying or non-bullying behavior. Participants included teachers and parents of children with autism spectrum disorder and community members without an autism spectrum disorder child. Participants' severity ratings of vignettes that described bullying differed by bullying type (i.e. verbal, physical, cyber, and interpersonal). Multilevel modeling revealed that bullying severity ratings are impacted by the age of children in the vignette, being a community member without children, and other demographic variables. These findings have implications for research methodology, assessment, and conceptualization of bullying in typical children as well as those with autism spectrum disorder.

Using the PDD Behavior Inventory as a Level 2 Screener: A Classification and Regression Trees Analysis.

In order to improve discrimination accuracy between Autism Spectrum Disorder (ASD) and similar neurodevelopmental disorders, a data mining procedure, Classification and Regression Trees (CART), was used on a large multi-site sample of PDD Behavior Inventory (PDDBI) forms on children with and without ASD. Discrimination accuracy exceeded 80 %, generalized to an independent validation set, and generalized across age groups and sites, and agreed well with ADOS classifications. Parent PDDBIs yielded better results than teacher PDDBIs but, when CART predictions agreed across informants, sensitivity increased. Results also revealed three subtypes of ASD: minimally verbal, verbal, and atypical; and two, relatively common subtypes of non-ASD children: social pragmatic problems and good social skills. These subgroups corresponded to differences in behavior profiles and associated bio-medical findings.

Digital Communication and Records in Service Provision and Supervision: Regulation and Practice.

While the use of computer-based communication, video recordings, and other "electronic" records is commonplace in clinical service settings and research, management of digital records can become a great burden from both practical and regulatory perspectives. Three types of challenges commonly present themselves: regulatory requirements; storage, transmission, and access; and analysis for clinical and research decision-making. Unfortunately, few practitioners and organizations are well enough informed to set necessary policies and procedures in an effective, comprehensive manner. The three challenges are addressed using a demonstrative example of policies and procedural guidelines from an applied perspective, maintaining the unique emphasis behavior analysts place upon quantitative analysis. Specifically, we provide a brief review of federal requirements relevant to the use of video and electronic records in the USA; non-jargon pragmatic solutions to managing and storing video and electronic records; and last, specific methodologies to facilitate extraction of quantitative information in a cost-effective manner.

Caregiver perspectives on unintentional injury risk in children with an autism spectrum disorder.

Unintentional injury risk research for children with an autism spectrum disorder (ASD) is currently limited. This article presents findings from a two-phase investigation of caregiver perspectives regarding unintentional injury risk in children with an ASD. Results indicate that children with an ASD exhibit elevated rates of risk-taking behaviors compared with peers, which increases the likelihood of more frequent and severe injuries. In addition, although ASD symptom severity positively correlated with risk-taking behavior and frequency of injury, children with an ASD were rarely rated as high risks for injury by caregivers. Implications are discussed in the context of pediatric health service provision.

Commentary on Drash and Tudor: An analysis of autism as a contingency-shaped disorder of verbal behavior.

A critique is presented of the etiological model for autism as presented by Drash and Tudor. The model is rejected based upon conceptual and methodological grounds. The major points raised concern overgeneralization of research findings, limitations of single subject methodology to answer population based questions, current neurobiological research, and the danger of uni-dimensional models. Specific examples of cases that do not fit Drash and Tudor model are presented.

The PDD Behavior Inventory: a rating scale for assessing response to intervention in children with pervasive developmental disorder.

The PDD Behavior Inventory (PDDBI) is a rating scale filled out by caregivers or teachers that was designed to assess children having a Pervasive Developmental Disorder (PDD; autism, Asperger disorder, PDD-NOS, or childhood disintegrative disorder). Both adaptive and maladaptive behaviors are assessed in the scale, making it useful for treatment studies in which decreases in maladaptive behaviors and improvements in adaptive social and language skills relevant to PDD are expected. The adaptive behaviors assessed include core features of the disorder such as joint attention skills, pretend play, and referential gesture. The maladaptive behaviors sample a wide variety of behaviors observed in both lower- and higher-functioning individuals and include stereotyped behaviors, fears, aggression, social interaction deficits, and aberrant language. The inventory was found to have a high degree of internal consistency. Inter-rater reliability was better for adaptive behaviors than for maladaptive behaviors. Factor analyses confirmed the structure of the PDDBI and indicated good construct validity. In a subsample of children between 3 and 6 years of age, raw scores for adaptive behaviors increased with age in the parent and teacher versions, as did measures of social pragmatic problems. It was concluded that the PDDBI is both reliable and valid and is useful in providing information not typically available in most instruments used to assess children with PDD.