Surprising region of brain may contribute to autism, says neuroscientist

In March, Oxford neurologist Lady Susan Greenfield announced in the House of Lords, and in an interview with the United Kingdom's Daily Mail, that she thinks social networking sites like Facebook, Twitter and MySpace can "rewire" children's brains and cause autism. This led countless panicked parents to ask at what age they should unglue their kids from computers to keep them from "catching" autism.

Greg Allen
Dr. Greg AllenPhoto: Christina Murrey
Dr. Greg AllenPhoto: Christina Murrey

Baroness Greenfield's contribution rounds out an extensive list of suggested autism causes that includes vaccines, genetic errors, abnormal bowels, being an older mom when the child is conceived, being an older father when the child is conceived, air pollution, viral infections, too much TV watching, poor nutrition, food allergies, immune system deficiencies, "bad parenting," lead poisoning and mercury poisoning. In a forum on the Web site, one enterprising reader offered alien abduction as a precipitating factor.Dr. Greg Allen, a neuroscientist in The University of Texas at Austin's College of Education, can't announce just yet that he has an airtight answer, but he's on a promising track.Since current data indicate around 1 in 150 children is diagnosed with autism, there's a real sense of urgency for scientists in the quest for autism's roots. Even though there have been major scientific advances over the past several years, some of the answers only seem to highlight the disorder's complexity and reinforce the hypothesis that multiple factors contribute to it. From the angle of genetics alone, scientists already know that over a dozen genetic defects are associated with autism.According to Allen, the challenge is exacerbated by the fact that what people refer to as "autism" actually is one of a group of spectrum disorders known as pervasive development disorders, which means that it doesn't manifest itself identically in each individual. It's a much more complicated condition than that, with different symptoms appearing in different people and the symptoms varying in severity.What is known and what remains fairly consistent is that autism spectrum disorder can be detectable as early as one year of age, usually is diagnosed when a child is around three or four years old and almost always is spotted by the time a child is five. Experts agree that, by definition, a person with autism exhibits three core characteristics--abnormalities in social interactions, marked aberrant communication skills and atypical repetitive behaviors. These symptoms may be severe or mild, ranging from minimal social impairments on the mild end to extreme retardation and total withdrawal from the world.An abundance of inconclusive study findings and contradictory information has made it difficult for parents who want definitive answers from credible sources. Dr. Greg Allen "With autism, right now we're faced with many questions and relatively few answers," said Allen, who joined the Department of Educational Psychology last fall. "There are scientists trying to develop the best tools for diagnosing autism and there are others working on the best intervention methods, and then you have those, like me, who are trying to figure out what underlies the behaviors and symptoms of autism."An abundance of inconclusive study findings and contradictory information has made it difficult for parents who want definitive answers from credible sources. Over the past 20 years or so, the cerebellum has consistently been found to be a site of brain abnormality in people with autism. Because of that, my approach involves using magnetic resonance imaging (MRI) to investigate the ways in which the cerebellum of a person with autism differs from that of a non-autistic person, and then to examine how that pathology contributes to the symptoms and behaviors of autism."According to Allen, the traditional view in the scientific community has been that the sole function of the cerebellum is to coordinate movement. However, over the past 10 or 15 years, scientists have reevaluated the cerebellum's role and some have suggested that this neuron-rich region at the base of the brain that's connected to all major divisions of the central nervous system also seems to affect cognition, emotion and the sensory domain."I was very fortunate to be at UC San Diego as a graduate student in the '90s working with Dr. Eric Courchesne, one of the leading researchers on the neural basis of autism," said Allen. "At that time, he had published papers showing that the size of the cerebellum in autistic individuals is smaller than in a person who doesn't have autism. He also was able to show an interesting correlation between the size of the cerebellum and aspects of attention, particularly one's ability to rapidly and accurately orient or shift attention."

Brain scan showing cerebellum
Allen's brain scans of people with autism show that cerebellar activation is abnormally low during tasks requiring attention and abnormally high during simple motor tasks. 

Using the research foundation that Courchesne and his colleagues established, Allen began to do functional magnetic resonance imaging (fMRI) brain scans, looking closely at the role of the cerebellum in attention. During an fMRI scan, the MRI signal changes when there's a change in task, so Allen hoped that examining fMRI scans of autistic and non-autistic individuals' brains would offer some insight into the role the cerebellum plays in the execution of attention tasks, as well as the role abnormal cerebellar functioning might play in autism.

Although a good portion of the scientific community now acknowledges that the cerebellum does more than coordinate movement, a thorough definition of its function remains elusive. Despite the research findings of scholars like Allen and Courchesne, there are some who hesitate to accept that the cerebellum may play a part in everything from working memory to conditioned anxiety, complex reasoning and problem solving.

Even when designing his earliest fMRI studies, Allen anticipated naysayers' arguments and structured his studies accordingly.

"When you do MRI scans to examine cerebellar activity during non-motor tasks, skeptics are going to assert that there really must have been some motor activity involved after all, however minimal, and that's why we're seeing activity in the cerebellum" said Allen. "So, in my very first fMRI study, I anticipated this 'motor activity argument' and had research subjects press a button when they saw stimuli in one condition, and then in another condition, I had them mentally tabulate the stimuli that they saw, with no motor activity. By comparing these conditions and a third one involving motor activity alone, I was able to show that the parts of the cerebellum involved in motor activity and the parts involved in attention are anatomically distinct."

Child playing outside
One in 150 individuals is diagnosed with autism, usually by three years of age. Finding the cause of autism will help refine treatment techniques as well as unlock the key to prevention. 

That study laid the groundwork for follow-up investigations in which Allen showed that in autistic individuals cerebellar activation is abnormally low during tasks involving attention and abnormally high during simple motor tasks. This difference in activation patterns may be related to regional differences in the number of Purkinje cells.

Purkinje cells are neurons that provide the only source of output from the cerebellum's cortex and control output from the cerebellum to other parts of the brain. Scientists have discovered that there are fewer of these cells in autistic individuals than in those without autism.

"A significant reduction in Purkinje cell numbers, which is one of the most consistent neurobiological findings in autism, will have a major impact on how the cerebellum communicates with the rest of the nervous system," said Allen. "This is why the focus of my current research is examining connections within the cerebellum and between the cerebellum and the rest of the brain. Understanding these connections may be a key piece to the puzzle of how abnormality in the cerebellum affects behavior and the symptoms of autism. But of course, that's the tip of the iceberg. We have so many more pieces of the puzzle to address."

A recent three-year, $750,000 grant from the National Institutes of Health will allow Allen to keep working on the puzzle, doing MRI brain scans and investigating the neural basis of autism.

Allen, who is the only researcher in Austin doing MRI scans to study autism, will begin the brain scans in April and is actively recruiting high-functioning, autistic 18- to 26-year olds for the project.

"One of the most exciting things about autism research right now is the wide range of scientists from diverse disciplines who are tackling questions related to this spectrum of disorders," said Allen. "So, while it is true that there are many questions yet to be answered, the interest that has emerged among researchers in more recent years is very encouraging. With so many scientists investigating from so many different angles, we're making great strides toward a clearer understanding of autism, and I feel confident that major findings are on the horizon."