Skip to content

STAART Network Centers: University of Washington

Project Descriptions

University of Washington Autism Research Center of Excellence

 

Primary Site: University of Washington
Geraldine Dawson, Ph.D., Director
Elizabeth Aylward, Ph.D., Co-Director

 

Early Characteristics and Intervention in Autism
Principle Investigator: Dawson, site: University of Washington

This project has three broad goals: (1) to improve early identification of autism, (2) to assess the efficacy of early intensive behavioral intervention for treatment of autism, and (3) to better understand individual child factors that account for variability in response to early intervention. The specific aims are:
1) To identify cognitive, social, neuropsychological, and electrical brain activity differences between 18-24 month old toddlers with autism versus 18-24 month olds with developmental (cognitive) delay (DD) and typical development. 2) To conduct a randomized study of early intervention to evaluate the efficacy of early intervention for improving outcomes of young children with autism, based on measures of cognitive, language, and social behavior. 3) To also evaluate the efficacy of early intervention for improving outcomes based on measures of brain activity. Our previous studies have shown that children with autism show atypical patterns of event-related brain potentials in response to social stimuli. Given the emphasis on improving social behavior (e.g., eye contact) in early intervention, it is possible that early intervention may result in improved outcome on brain measures related to social processing. We will determine whether early intervention results in changes in brain activity such that, after treatment, children who have received early intervention will show more normal patterns of brain activity than those who do not receive such intervention. 4) To identify individual child factors that account for variation in response to early intervention in young children with autism. We hypothesize that three child factors will be important predictors of response to intervention. These are (1) IQ, (2) severity of autism symptoms, and (3) degree of early brain impairment, specifically, degree of medial temporal lobe dysfunction. A better understanding of factors related to response to early intervention would inform decisions regarding appropriate, individualized intervention methods and elucidate brain mechanisms involved in autism.

Early Language Characteristics in Autism
Principal Investigator: Kuhl; site: University of Washington

Language and communication impairments are key components of autism. Our laboratory has been conducting studies on preschool age children with autism examining early aspects of language processing. These studies have revealed critical differences in phonetic discrimination, social communication, and cross-modal processing between preschool age children with autism and mental-age and chronological-age matched groups of developmentally delayed (DD) and typically developing children. As these measures reflect abilities that emerge during infancy, these results signal the potential of early speech measures for identifying children with autism at a very young age. Furthermore, it is possible that these early measures of language and communication ability may prove to be very sensitive predictors of language outcome for children with autism. In the current proposal, we plan to examine these early speech measures - namely, (1) event-related brain potential measures of phonetic perception, (2) listening preference for speech versus mechanical-sounding auditory signals, and (3) vocal imitation abilities - in 18-24 month children with autism, and comparison groups of children with DD and typical development to determine whether such measures discriminate children with autism at an early age. Furthermore, we will assess their value in explaining individual differences and predicting language outcome at age 4 for children with autism.

Specific Aims:

  1. Basic Language Processing: In our recent studies, event-related brain potential (ERP) tests of phonetic perception demonstrated that 3-4 year old children with autism exhibit deficits in the basic capacity to differentiate the building blocks of language, the phonetic units that make up words. An inability to differentiate basic speech units would make language learning very difficult. We propose to use ERP to assess the capacity of the brain to differentiate basic speech units in 18-24 month-old children with autism, and comparison groups of children with DD and typical development. We also will examine the predictive relation between the presence of an early deficit in phonetic processing and language outcome at 4 years of age in children with autism.
  2. Social Communication Skills That Underpin Language: Research on typically developing infants indicates that listening to ambient language results in a sophisticated "mapping" of the properties of that language; this process is now considered critical to language acquisition. We found that listening preferences in children with autism differ dramatically from those demonstrated by typically developing and DD children (Kuhl & Padden, in preparation). Given a choice, preschool aged children with autism preferred listening to mechanical-sounding auditory signals (ones acoustically matched to speech) rather than speech. We found that preference for this "sine-wave analog" was related to level of language ability, as assessed by standardized measures. In the current proposal, we will employ this speech preference test and assess whether 18-24 month old children with autism exhibit abnormal listening preferences. We will also examine its potential for predicting language outcome in children with autism at age 4.
  3. Speech Imitation Our previous results suggest that preschool age children with autism are much more likely to imitate simple syllables that they both see and hear than typically developing children. Moreover, we observed a new phenomenon, eye-covering, in preschool age children with autism during speech imitation tasks. Among the children with autism, eye-covering in response to speech was not seen in those who vocally imitated, and was associated with lower receptive language skills. In ongoing studies, we are investigating the possibility that children with autism have difficulty simultaneously processing auditory and visual signals. In the current proposal, we plan to examine vocal imitation in 18-24 month old children with autism, assessing both imitation and eye-covering, and their association with language outcomes.

 

Electrophysiological and fMRI Studies of Face Processing in Autism
Principal Investigator: Aylward; site: University of Washington

We propose to study the brain bases of one of the most basic aspects of social cognition, face processing. Parallel studies utilizing event-related potentials (ERPs) and functional magnetic resonance imaging (fMRI) will provide information about abnormalities in the timing and regional distribution of brain response to faces in autism. Twenty-five adolescents and adults with idiopathic autism and 25 IQ- and age-matched typical individuals will participate in each study. These studies will examine ERP and fMRI in response to face vs. nonface visual stimuli, to specific face parts (eyes vs. mouths), to moving vs. static face parts, and to familiar vs. unfamiliar faces. Separate eye-tracking studies on the same individuals will examine eye movements during viewing of different types of face stimuli. We will test a novel hypothesis that abnormalities in face processing in autism are related to atypical attentional strategies when viewing faces. Results of the ERP, fMRI, and eye-tracking studies will also be correlated with performance on neuropsychological tests of face perception and memory, and behavioral measures of social impairment.

These studies will shed light on the nature and neural bases of face processing impairments in autism. Such information is clinically important for early identification, development of targeted interventions, and investigation of the genetic basis of autism. There is strong evidence of a genetic component in autism, and we expect that this research will lead to more refined measures of quantitative traits that can be used in genetic studies.

Specific Aims:

  1. To elucidate the nature and brain bases of impairments in early stage processing of face features in autism. We hypothesize that ERP and fMRI brain activation will show differences in both timing and regional distribution of brain function for individuals with autism, as compared with typical participants. It is hypothesized that group differences will be found during viewing of human faces and isolated face components (eyes, mouths) but not during viewing of nonface visual stimuli, suggesting a selective impairment of social information processing.
  2. To determine whether face processing impairments in individuals with autism include abnormalities in the processing of face movements. This study addresses the origin of a core impairment in autism, namely "joint attention," which requires sensitivity to another person's eye gaze patterns. Based on recent evidence suggesting that individuals with autism show attentional preference for the mouth over eyes when viewing faces, we hypothesize that fMRI and ERP activations will differ for individuals with autism, as compared with typical participants, during viewing of eye movements, but not during viewing of mouth movements.
  3. To examine the neural patterns associated with face recognition in autism. Based on recent data from young children with autism showing memory impairments for faces but not objects, we hypothesize that fMRI and ERP activations will differ for individuals with autism, as compared with typical participants, during viewing of familiar faces (e.g., those of family members), but will not differ during viewing of familiar objects.
  4. To determine whether abnormal neural processing of faces in individuals with autism is related to atypical attentional strategies. We hypothesize that patterns of eye tracking will differ for individuals with autism, as compared with typical participants, when they view static and moving faces, and that individual differences in attentional strategies, as measured by eye-tracking during viewing of faces, will be associated with differences in patterns of brain activity, as measured by ERP and fMRI during face perception tasks.
  5. To determine the relation between neural indices of brain function and behavioral tests of neuropsychological function, social cognition and behavior. We will examine whether abnormalities in ERP, fMRI, and eye tracking are related to performance on neuropsychological tests of face perception/memory and behavioral measures of social functioning.
  6. To determine the effect of extensive training in face processing on fusiform activation. Following an initial fMRI scan involving viewing of faces vs. houses, we will provide multi-session training to familiarize subjects with either a set of faces or a set of houses. Following training, we will perform fMRI scans that will allow us to determine for each training group whether fusiform gyrus activation increases from baseline for comparison of trained vs. untrained stimuli and faces vs. houses. We hypothesize that subjects who received face training will show more activation for trained faces vs. houses, and perhaps for untrained faces vs. houses (which would indicate that training generalized face processing skills to unfamiliar faces). We will compare changes in activation for subjects in the face-trained group with subjects in the house-trained group to determine whether the face training had more of an effect on fusiform activation than training not designed to improve face processing skills.

Brain Development in Autism
Principal Investigator: Dager; site: University of Washington

Previous work in our laboratory demonstrated that 3-4 year-old children with autism spectrum disorder (ASD) have cerebral enlargement, as well as proportional enlargement of subcortical regions (cerebellum, hippocampi and amygdalae), compared to age-matched children with developmental delay (DD) and typical development (TD). Children with strictly defined autistic disorder (AD) showed enlargement of the amygdalae, in excess of overall cerebral enlargement. From the same groups of children, proton echo-planar spectroscopic imaging (PEPSI) revealed widespread regional decreases in brain neurochemical concentrations and abnormal metabolite relaxation in the ASD children relative to controls, indicative of abnormal cellular composition. These findings demonstrate abnormal brain developmental processes in autism by 3-4 years of age.

We aim to increase our understanding of abnormalities in early brain development by conducting brain imaging research with children with AD at the earliest age it is currently possible to diagnose this disorder. We propose to characterize brain structure and tissue-based neurochemistry in forty-eight 18-24 month-old children with AD and matched comparison groups of 25 children with DD and TD. In the AD group, we will relate brain structural and chemical findings taken at 18-24 months to neurocognitive and symptom measures taken at 18-24 months and later at 4 years of age. Furthermore, based on repeated measures of head circumference obtained retrospectively from birth to 18-24 months, and prospectively from 18-24 month to 4 years, the early developmental course of head growth in autism will be examined.

Specific Aims:

  1. Two-dimensional (2-D) and three-dimensional (3-D) magnetic resonance imaging (MRI) will be acquired from 18- to 24-month-old children with idiopathic AD (N=48), idiopathic DD (N=25) and TD (N=25). Morphometric analyses, employing well-established segmentation techniques, will be used to characterize differences in brain morphology among the three groups. We will (1) determine volumes of cerebrum, cerebellum, hippocampi, amygdalae, ventricles, thalami, and basal ganglia, (2) measure brain gray and whitematter composition and characterize the patterns of distribution and (3) evaluate for variations in gyral patterns. We will investigate whether brain structural abnormalities, identified in our work studying 3- to 4-year-old children with AD (Sparks et al., 2002), specifically enlarged cerebrum and subcortical regions, are present as early as 18 to 24 months in children with AD.
  2. From the same samples of children with AD, DD and TD, 2-D proton echo-planar spectroscopic imaging (PEPSI) chemical images will be used to quantitate regional brain chemistry and T2 relaxation time constants. For specific brain regions exhibiting structural abnormalities (e.g., larger than normal), we hypothesize that brain chemical patterns ofN-acetyl aspartate, inositol, choline, creatine and lactate levels will help determine if such structural abnormalities reflect underlying alterations of brain cellular composition.
  3. Relations will be examined between individual differences in regional-specific measures of brain anatomy/chemistry and concurrent measures of behavior, including symptom severity, intellectual and language function, social abilities, and neuropsychological function. We hypothesize that regions observed to exhibit structural and/or chemical abnormalities in older children with autism, which include the cerebral hemispheres, medial temporal lobes and midline nuclei, will be abnormal in 18- to 24-month-old children with AD.
  4. We will assess whether brain structural and chemical measures taken at 18 to 24 months are predictive of clinical course, response to early intervention, and outcome at age 4. We will test hypotheses that severity of autistic symptom presentation and longitudinal course will be related to the degree of structural and chemical abnormalities found in discrete brain regions. Specifically, we hypothesize that amygdalar enlargement and lower NAA will be predictive of more severe autistic symptom expression and a slower rate of neurocognitive and social growth during the preschool years, and will serve as a moderator of response to early intervention.
  5. In order to better understand abnormalities in brain development from birth to 4 years of age, we will examine patterns of head growth from birth to 4 years in children with AD, DD and TD. Postpartum medical records, including birth records and well-baby checks, will be used to identify postnatal patterns of head circumference growth, accounting for factors including height, weight, ethnicity and gender. We will examine differences among patterns of head growth for the three groups of children (AD, DD and TD) at 18 to 24 months of age, and examine the relation between individual differences in head growth pattern and clinical course and outcome during the preschool years.

    Back to STAART Network Centers Page and Other Centers