Home > Funding > Grant opportunities > Abstracts > Canonical neural computation in autism

Canonical neural computation in autism

David Heeger, Ph.D.
New York University

Marlene Behrmann, Ph.D.
Carnegie Mellon University

Tony Norcia, Ph.D.
Stanford University

A pervasive aspect of autism is hypersensitivity to sensory stimuli. This problem causes significant disruption in the lives of those with autism, but is often considered secondary to cognitive and social differences. Based on their knowledge of sensory processing, David Heeger, Marlene Behrmann, Tony Norcia and their colleagues hypothesized that sensory problems are a manifestation of pervasive impairments in fundamental neural computations.

They hypothesized that malfunction in these neural computations may be the core deficit in autism, providing a foundation not only for sensory hypersensitivity, but also for cognitive and social differences. Understanding autism-related differences in sensory responses may elucidate the underlying psychological and neural differences in autism. Instead of focusing on deficits in a particular functionally specialized brain area or system, the researchers hypothesized that generic, brain-wide circuit malfunctions lie at the root of these disorders.

The researchers measured brain activity with functional magnetic resonance imaging in the visual, somatosensory and auditory cortices in response to sight, touch and sound. They found that sensory-evoked brain responses are less reliable and more variable in individuals with autism than in controls1.

This neural response unreliability is correlated across all three sensory modalities and with the severity of the autism profile based on Autism Diagnostic Observation Schedule scores1. This robust and atypical neural activity is not observed when there is no driving stimulus. The altered neural response reliability in autism may be associated with variability in perception. But there is no evidence of deficits in visual attention (the ability to attend to stimuli in the environment) for individuals with autism2,3.

In a complementary series of experiments, Norcia and his colleagues used electroencephalography (EEG) to measure brain activity in children with autism. They observed that visual contrast processing and sensitivity-regulation mechanisms are abnormal in autism. These abnormalities can be measured quantitatively and noninvasively using EEG in children and infants with only a few minutes of recording.

Together, this research addresses fundamental changes in neural computation, possible neurobiological mechanisms that give rise to these changes and the consequence of these changes for autism’s behavioral profile.


1. Dinstein I. et al. Neuron 75, 981-991 (2012) PubMed

2. Grubb M.A. et al. J. Vis13, 9 (2013) PubMed

3. Grubb M.A. et al. Autism Res. 6, 108-118 (2013) PubMed

proposalCENTRAL #177638