Research Abstracts from CPEA Investigators

Overall description: Language and communicative impairments are among the core features that define autism. Universal symptoms of this impairment include deficits in pragmatic aspects of language, which are closely tied to theory of mind deficits, and social-communicative deficits that involve non-verbal channels of communication, including the ability to process faces and their communicative signals. There is also a subgroup of individuals with autism who have additional linguistic impairments, which include deficits in phonology, lexical and grammatical knowledge. In the current award period we identified this subgroup of language-impaired children with autism and proposed that they have language deficits, and associated structural brain abnormalities that are identical to those found in specific language impairment (SLI). In the next award period we plan to pursue these language and social-communicative deficits in three projects, all involving the same groups of children with autism, SLI and normal controls. Project I will investigate abnormalities in the perception of faces and their communicative signals, exploring the relationship between these deficits and social impairment in autism. Project II will investigate deficits in language processing in autism, to further characterize similarities between the autism subgroup and SLI. Project III will use structural and functional magnetic resonance imaging (MRI) to investigate the relationships between the cognitive/linguistic deficits in Projects I and II with underlying brain pathology in autism. The projects will be served by a unified Clinical and Administrative Core that will recruit and assess the children participating in these projects, maintain the program project database, provide statistical consultation, and coordinate all the administrative activities for the program project including our relationship with the CPEA.

Project I: Faces and Their Communicative Signals in Autism
Principal Investigator: Robert M Joseph, Ph.D.

The goal of the proposed research is to investigate the social-communicative abilities of children with autism as revealed through their capacity to process information from people's faces. In recent research, we found that children with autism engage in atypical face recognition strategies that involve an unusual reliance on the mouth, and deficient processing of the eyes. We plan to extend this work by investigating the entire range of face perception abilities in autism, including perception of face identity, facial expressions of emotion, eye gaze direction, and facial speech. A series of experiments will be conducted with autistic children aged 10-16 and with NVIQ>70; an age- and NVIQ-matched, language-impaired control group, and an age-matched normal control group. The main predictions are that:

1. In face identification, children with autism will exhibit impaired eye recognition even when they are cued to attend to the eye region, and will exhibit impaired processing of eyes but intact processing of mouths across measures of holistic face recognition.
2. They will show a similar pattern of deficiency in the perception of facial emotions, with intact recognition of emotions expressed via the mouth, but impaired recognition of emotions expressed primarily through the eyes.
3. They will be impaired in following eye gaze and in judging eye gaze direction.
4. They will exhibit a heightened perception of speech-related movements of the mouth.

Across experiments, our goal is to assess how impairments in different aspects of face perception may be related in autism. In addition, we will assess the relationship of all experimental face perception variables to visual scanning patterns; lower-level visual perception abilities; and concurrent social-communicative functioning. In its broad focus on all aspects of face processing, this research will provide a systematic delineation of autistic impairments in a vital channel of communication, and will thus contribute to a better understanding of autism's core symptomatology. Further, in investigating a range of face processing skills that have been intensively studied on the neuroanatomical level, this research can contribute to the elucidation of the brain bases of autism and its genetic etiology.

Project II: Language Impairments in Children with Autism
Principal Investigator: Helen Tager-Flusberg, Ph.D.

Delays and deficits in language are among the defining features of autism. Deficits in the domain of pragmatics, which involves the ability to use language in appropriate ways to communicate with others, are universally found in autism. Our ongoing research has shown that these pragmatic deficits are closely related to theory of mind impairments in children with autism. We have also found that among relatively high-functioning children with autism, about one-half demonstrate significant impairments in linguistic knowledge, and that these impairments mirror the profile of deficits that define specific language impairment (SLI). We hypothesize that this group of language-impaired children with autism represents an important subtype that suggests overlap between these two disorders: autism and SLI. We plan to follow up these findings in experiments on language processing skills in children with autism, children with SLI and normal controls. Four series of experiments are proposed that investigate: (1) Phonological representations—deficits in non-word repetition and discrimination; (2) Morpho-syntax—deficits in grammatical knowledge related to marking tense using natural language samples, judgment and reaction-time tasks; (3) Prosody—the use of prosodic information in the speech signal to disambiguate syntactic information, or to convey a speaker's emotional state; and (4) Word learning and naming—use of phonological or speaker's intention cues to words presented in different contexts. We hypothesize that the subgroup of language-impaired children with autism will perform like the SLI children in the experiments that tap phonological and grammatical knowledge. In contrast all the children with autism will perform worse that the SLI and normal controls on those experiments that entail inferring or using knowledge about speakers emotions or intentions. The findings from these studies will provide important new information about the language impairments in both autism and SLI and help to refine the language-impaired subtype in autism for future genetic and neuroimaging studies.

Project III: Neuroimaging of Language and Communication in Autism
Principal Investigator: Gordon Harris, Ph.D.

Deficits in language and social-communication functioning represent one of the primary manifestations of autism. However, to date there has been little research into the underlying neural mechanisms responsible for language and social-communication dysfunction in autism. Project III of the program project, Language in Autism: Clinical and Basic Studies, will use structural MRI and functional MRI (fMRI) techniques to evaluate the anatomy and physiology of brain regions and tasks related to language and social-communication in autism. Subjects will include adolescent (age 10-16) autistic, specific language impaired (SLI), and matched control groups. fMRI is ideally suited for the functional studies in this project because it is non-invasive, and does not use ionizing radiation, allowing multiple experimental conditions to be evaluated within individual subjects. The structural imaging studies will examine gray matter volumes in language and social-communication-related regions within the cerebral cortex using segmentation, parcellation, and voxel-based morphometry (VBM) methods. White matter and cerebellum have also been implicated in autism, and so structural analyses will also include cerebellar and white matter parcellation and diffusion-tensor imaging (DTI) of white matter organization. Measures of brain structure will be evaluated for correlation with language performance determined through the Core evaluations. The fMRI experiments will evaluate high-functioning autistic subjects and matched control groups under five experimental conditions: lexical-semantic encoding, past tense processing, phonological processing, processing facial expressions linked to eyes vs. mouths, and activation related to direct vs. averted eye gaze. The three language fMRI tasks relate to experiments and findings from Project II (language), while the two social-communication tasks relate directly to Project I experiments and findings (social-communication). There is evidence that each of these domains demonstrates abnormality in autism, and we will evaluate the neural activation responses to these activation paradigms through a series of five experimental conditions.