Skip to content

STAART Network Centers: Yale University

Project Descriptions

The Social Neuroscience of Autism and Related Disorders

Primary Site: Yale University
Fred Volkmar, M.D., Director
Ami Klin, Ph.D., Co-director

Eye-Tracking Studies of Social Engagement
Principal Investigator: Klin; site: Yale

This project focuses on two recent hypotheses of early-emerging social cognitive mechanisms shown to be impaired in individuals with autism. The first hypothesis states that when viewing naturalistic social situations, individuals with autism exhibit aberrant patterns of visual fixation and that these patterns predict the participants' relative levels of social competence. This hypothesis originated from our research using eye-tracking technology to measure fixation times on different aspects of social scenes. The second hypothesis states that individuals with autism have a decreased ability to impose social meaning on ambiguous visual stimuli. This hypothesis originated form our research in which participants with autism had difficulty making social attributions to geometric animations depicting typical social interactions. We requested 5 years of support to examine these hypotheses in 60 infants suspected of having autism spectrum disorders (ASD) seen longitudinally at age 24 to 36 months (Time 1) and again at age 48 to 60 months (Time 2). This group will be matched on nonverbal MA to 30 children with non-autistic developmental delays (DD) also followed longitudinally. ASK and DD groups will be matched to typical controls cross-sectionally (N1=40 in Time 1 and N2=40 in Time 2). In Specific Aim 1 we plan to use eye-tracking technology to study visual scanning patterns during viewing of naturalistic social interactions presented as scenes of children at play and as facial close-ups of an engaging adult. In Specific Aim 2 we plan to use eye-tracking technology to study the capacity for imposing social meaning upon ambiguous representations of people (people rendered as moving point-light displays). For both Specific Aims, data will be analyzed cross-sectionally at Times 1 and 2. At both times we plan to study the relationship of visual fixation patterns to diagnostic group membership and to standardized outcome measures of social and communicative competence. In addition, the longitudinal component of the study creates and opportunity to test the predictive power of aberrant visual fixation patterns at Time 1 relative to confirmatory diagnostic group membership and outcome measures at Time 2. Apart from providing a unique window into the ways in which young children with autism search for meaning when confronted with social situations, our longer-term goal is to develop our eye-tracking paradigms into performance-based measures capable of (1) identifying children with autism in the first two years of life (an important goal give the proven benefits of early intervention), and (2) defining a continuum of social competence (a critical component in genetic research of varying manifestations of autism). This project builds on funded work using similar paradigms both with older individuals with autism and with an animal model of autism (non-human primates with mesiofrontal-limbic ablations).

Specific Aims:

  1. Aim 1 is to study the relationship between visual scanning of naturalistic social scenes relative to group membership and standardized measures of social and communicative competence.
  2. Aim 2 is to study the relationship between the capacity to attribute social meaning to visually ambiguous displays of human interaction relative to group membership and standardized measures of social and communicative competence. The capacity to attribute social meaning will be quantified by visual scanning measurements of preferential attention to the upright social animation.

Gaze Processing in Young Children with Autism
Principal Investigator: Volkmar; site: Yale

The deficits in spontaneous gaze monitoring of others are well documented in ASD and constitute a part of a complex social-cognitive deficit in joint attention skills. Little is known, however, about the extent and the underlying mechanisms of these deficits. This research will investigate gaze perception in children under the age of 3 years diagnosed with an autism spectrum disorder (ASD). It will target skills that emerge in typical development in the first months of life and remain functional throughout adulthood, namely covert attentional cueing by perceived direction of gaze (i.e. reflexive joint attention). These skills will be measured using an experimental procedure in which latency of eye movement to peripheral targets following presentation of directional cues is assessed. Two preliminary studies suggest that this method is highly effective in studying young children with ASD and that there are important similarities as well as differences in gaze processing between ASD and TC groups. The extent of these differences and their underlying mechanisms remain to be explored.

Specific Aims:

  1. to test the hypotheses that children with ASD under 3 years of age process information about eye gaze direction differently than children with developmental delays (DD) and typically developing children;
  2. examine the relationship between the performance on perceptual cueing tasks and severity deficits in social functioning;
  3. to examine the relationship between the attention work we will have elucidated some of the mechanisms underlying deficits in spontaneous gaze monitoring in autism. Our future research will focus on applying this knowledge to devising simple behavioral methods of identifying infants at risk for autism as well as intervention procedures aimed at the specific social-cognitive deficits observed in early onset autism.

Roots of Social Communication: Auditory Preferences
Principal Investigator: Paul; site: Yale

This project focuses on four hypotheses related to early-emerging social communicative mechanisms known to be impaired in individuals with autism: (1) Toddlers (24-36 mo.) with symptoms of autism will show less than typical preference for listening to speech under controlled experimental conditions, as opposed to nonsocial auditory stimuli with analogous acoustic properties. (2) Toddlers with symptoms of autism will show less than typical preference for three prosodic patterns characteristic of their native language: stress patterns in words, pause patterns in sentences, and intonation contours in connected speech. (3) Listening preferences will be correlated with concurrent degree of communicative and social competence and will predict later language and social competence. (4) Listening preferences at 24-36 mo. will discriminate children who meet diagnostic criteria for autism spectrum disorders from those who do not, both concurrently and when diagnosis is confirmed at 48-60 mo. We requested 5 years of support to examine these hypotheses in children with autism-related conditions seen at 24-36 mo (Time 1) and again at 48-60 mo (Time 2). This group will be matched to a group of children with non-autistic developmental delays and to a group with typical development. For Hypothesis 1 and 2, the preferential head-turn paradigm, a method used for over 20 years in the study of infant auditory perception, will be used to examine the group differences in listening preference. For Hypothesis 3, the relationship between these listening preferences and standardized measures of communicative and social competence will be examined at both Time 1 and Time 2. For Hypothesis 4, the predictive power of listening patterns relative to concurrent diagnostic assignment and confirmatory diagnostic group membership two years later will be investigated. Our longer-term goals are to develop these preferential liste4ning paradigms into performance based measures capable of identifying children at risk of having autism in the first year of life, which is an important goal given the proven benefits of early intervention. In addition, the information gained in this study will be used to generate hypotheses regarding appropriate treatment procedures to optimize early listening experience in children with these disorders.

Specific Aims:

  1. to determine whether significant differences exist in the auditory preferences for speech and native language prosodic patterns shown by young children with autistic symptoms, as compared to the listening preferences of children with typical and delayed development;
  2. to determine whether there are significant relationships between these preferences and concurrent indicators of socialization, communication and of diagnostic assignment; and
  3. to determine whether these preferences can predict indicators of socialization, communication and diagnostic assignment two years later.

Behavioral and Neural Plasticity in Face Perception
Principal Investigator: Schultz; site: Yale

Persons with autism spectrum conditions (ASDs) have skill deficits in facial identity and facial emotion recognition. For example, we recently showed that the time spent examining the eye region of the face by persons with an ASD distinguished them from a matched control group with %100 accuracy. Moreover, using fMRl we have shown that the function of 'fusiform face area' is abnormally hypo-active in persons with an ASD. The hypoactivation of the fusiform gyrus is, in fact, the most well replicated finding in the neuroimaging literature on autism, with at least 4 independent studies confirming it. Importantly, the degree of hypoactivation is also strongly predictive of the degree of social deficit (this relationship is not mediated by subject task performance during the fMRI). In addition, we now have fMRl data showing that persons with an ASD exhibit less activity in a cortical region specialized for decoding facial expressions (the superior temporal sulcus), and in the amygdala. A next logical step in this line of work is to test the plasticity of these brain areas by systematically training persons with autism and related conditions on a curriculum of face and facial expression computer games. Prior work with perceptual expertise training in typical persons has demonstrated that this approach can increase activation in a specialized region of the fusiform gyrus. This suggests that an intervention program for persons with an ASD may be able to normalize activations in these areas. We propose to use a computerized game platform, "Lets Face It!", in a randomized intervention trial (6 months in duration) for 63 children with an ASD in comparison to a matched group of 63 persons with an ASD in a delayed intervention group. This will be a multi-site trial at Yale University, Oberlin College and the Delaware Autism Program. Sixty of the participants will have pre- and post fMRl and eye tracking studies, in addition to a core battery of face and social skills assessments. We aim to test the effectiveness of a training curriculum based on "Lets Face It!" for improving (1) facial identity skills, (2) facial expression skills and (3) general social skills. In addition, we predict that a systematic training program will increases activation of the fusiform face area, the superior temporal sulcus region associated with facial expressions, and the amygdala.

Specific Aims:

  1. To assess the perceptual skill of 126 persons with an ASD on a visuoperceptual battery of face identity and facial expression recognition tests, including tests designed to measure configural processing and feature- based processing. Although there is good evidence for deficits in face and facial expression recognition in ASD, the nature of the processes underlying these deficits has not yet been carefully studied with measures that are grounded in contemporary visual-cognitive theory.
  2. To finish development of a computerized gaming platform, "Lets Face It!" (LFI), a rehabilitation tool specifically for persons with an ASD. LFI is designed to be an engaging and naturally rewarding set of games with activities targeting three skills areas:
    (a) attention and interest in faces and facial expressions,
    (b) perceptual and recognition skills for face identity and expression and
    (c) understanding the relationship between social contexts and facial expressions.
  3. To test in a randomized treatment design, the effectiveness of an intervention curriculum based on LFI to:
    (a) enhance the sa/ience/re/evance of faces and facial expressions for persons with an ASD.
    (b) improve visual perceptual abilities of persons with an ASD to discriminate and recognize facial identity and facial expressions.
    (c) improve social skills and functioning, broadly construed.
  4. Sixty-three persons with an ASD will be randomly assigned to the active intervention, and 63 to a delayed- intervention control group (a 6 month waiting list, followed by active intervention). The intervention will span 6 months and 125 to 150 hours of active training, and will be coordinated at three sites (Yale, Oberlin and the Delaware Autism Program). Pre- and post intervention measurements will include the baseline characterization battery described as Aim 1, measures of social skill, and specialized measures described in Aims 4 and 5.

  5. To test whether the LFI curriculum can increase the percent time attending to the eye region of the face as measured with pre- and post- intervention eye tracking studies, in 60 persons with an ASD evaluated at the Yale site.
  6. To test whether the LFI curriculum can increase the blood oxygen level dependent (BOLD) signal level within cortical areas selective for facial identity and facial expressions measured with pre- and post intervention functional MRI (fMRI) evaluations in the same 60 persons described in Aim 4.

Citalopram Treatment in Children with Autism Spectrum Disorders and High Levels of Repetitive Behavior
Principal Investigator: Scahill; site: Yale
Multi-site study - see description under Social and Affective Processes in Autism
Primary Site: Boston University School of Medicine

Back to STAART Network Centers Page and Other Centers