U Magazine
U Magazine
UCLA Health
David Geffen School of Medicine

Dr. Daniel Geschwind

UCLA Medicine Fall 2009-Dr Daniel GeschwindThe globally renowned UCLA scientist straddles the disciplines of neurology and genetics in his effort to reveal the hidden origins of autism.

INITIALLY TRAINED AS A NEUROLOGIST, Daniel Geschwind, M.D., Ph.D., began his career trying to understand how humans develop their higher cognitive specializations like language and advanced social abilities – those parts of us that make humans human. “I was asking the question, ‘How does our brain, the organ that underlies these behaviors, develop? Why is it different from chimpanzees or mice?’ And that led to genetics because genetics is an extremely powerful set of tools for unraveling those kinds of mysteries,” says Dr. Geschwind, who is director of the Neurogenetics Program and the UCLA Center for Autism Research and Treatment in the Semel Institute for Neuroscience & Human Behavior, and he holds the Gordon and Virginia MacDonald Distinguished Chair in Human Genetics.

Working in partnership with 30 research institutions across the country, Dr. Geschwind and his team recently identified a new gene variant that is common in the general population but occurs about 20 percent more often in children with autism, and which is most active in key regions of the brain that support language, speech and interpreting social behavior. The findings, published in the journal Nature, suggest that the gene plays a critical role in shaping the developing brain and may contribute to a prenatal risk for autism.

This interview is an edited transcription of a conversation between Dr. Geschwind and Susan Loewenberg, the producing director of L.A. Theatre Works. The interview was produced in connection with a radio-theater drama about a young girl with autism, Lucy, and was aired in August 2009 on the L.A. Theatre Works weekly public-radio broadcast. It is published with permission of L.A. Theatre Works.

When was autism first recognized?
Dr. Geschwind: Autism wasn’t recognized as a distinct condition until it was described by [Austrian psychiatrist and physician] Leo Kanner in the 1940s. In his seminal publication, Kanner described 11 children, mostly boys, who had an autistic disorder. If one goes back and looks at that, it’s extremely accurate, vivid and in-depth, and really does relate to a lot of what we see in autism today. The key aspect is a dysfunction in normal social interaction and engagement, but it also includes problems in language and difficulties in an area that can be thought of as mental flexibility, so that the children might have repetitive behaviors, or restrictive behaviors, or a resistance to changes in their normal routine that can cause them to become quite disruptive. That’s really the fundamental core three features of autism.

What about the inability to express emotion or reciprocate affection?
Dr. Geschwind: Emotion and affection and their reciprocation are not considered core features of autism, although they may often be abnormal. Many autistic children are very affectionate. It’s an issue of how they relate socially, and so there is quite a range. Now, that doesn’t mean that it’s absolutely 100 percent normal, but it’s not the absence of those things that defines autism; it’s just that they may be slightly or moderately abnormal. Many autistic children have difficulty properly perceiving emotions in others. We don’t know whether the core problem with social behavior and social cognition and what we call social reciprocity or engagement are due to a difficulty with emotional regulation or the ability to perceive emotions in the faces of others or in the gestures, or if that’s just a part of it.

Is there some thought that an appropriate reaction is going on in the brain, but the actual expression of it to the outside world is encumbered?
Dr. Geschwind:
Social interaction – the sharing and giving of ideas and mental states – requires an integration of an enormous amount of the brain and its systems. You can have pieces of things, but it can look very abnormal on the outside because of the really delicate orchestration that has to go on within the brain. The one thing about autism is that it’s very heterogeneous in what its causes are, its etiology, as well as what it clinically looks like, so that there are people with autism who have severe dyspraxia (difficulty with planning and executing complex movements) or difficulty actually speaking (dysphasia), and its variability is remarkable. Now, people have gotten some of those folks to type, and it’s clear that some non-verbal children have rich language and are very intelligent. There definitely is a subset of these children and adults who will have quite complex and meaningful inner worlds that aren’t being expressed to the outer world. We don’t know the extent to which that might be universal across autism, but I suspect that it’s more frequent than we thought before.

In the early 1970s, the influential psychologist Bruno Bettelheim promoted what he called the “refrigerator mother” theory, which held that cold, un-nurturing parents, especially mothers, were to blame for autism. How long did this view of the condition prevail, and how pervasive and infl uential was it?
Dr. Geschwind: The “refrigerator mother” theory was quite unfortunate in the sense that it really moved autism out of being a biological or biomedical disorder into the realm of poor parenting, which totally influenced the direction of research. If you think that something is not biological, that it’s just due to poor parenting, then you don’t look into its causes and etiology. Now we know that autism has an enormous genetic liability, that there are environmental things as well that can lead to autism. So, we know that poor parenting is not the cause of autism.

But, one also has to put Bruno Bettelheim’s work in perspective in that although his theory was, I think, very destructive for the field and very tough for families as well, it was based on research that actually had gone on in the 1940s and 1950s with monkeys. The psychologist Harry Harlow and his colleagues showed that if you don’t give monkeys the right mothering, they end up being socially awkward and looking somewhat autistic and showing autistic-like behaviors. I think that extreme examples of extraordinary depravation of normal parent/child bonding will cause all kinds of mental illness or contribute to them, but those extremes are seldom seen in any society.

One example of children under such deprivation is orphans in Romania who have an increased prevalence of autistic behaviors due to the fact that they basically had no proper parenting and were abused or mistreated. I don’t know the literature there perfectly, but that’s a really big outlier in terms of parenting, and that would not explain the vast majority of autism. But autism and brain development and brain disorders are also due to an interaction of genetic factors and things that happen in the environment. So, extreme forms of deprivation in the right genetic context could certainly lead to severe abnormalities of behavior that might be considered along the autistic spectrum. One thing to say, though, about the issue of the parent/child interaction, it’s clearly important in child development, but the point of the “refrigerator mother” was that it was causal, and it’s very clear that in the vast, vast majority of cases, the parents have nothing to do with the causality.

UCLA Medicine Fall 2009-Dr GeschwindAutism is genetic, yet you don’t have autistic parents of autistic children. How do you explain that?
Dr. Geschwind: Genetics works in many ways. If we think about autism like we would think about height or weight, often we see tall parents who tend to have tall kids, but not always. In fact, if tall parents have five or six children, the children may vary in height quite a bit. You may have thin parents and slightly heavy children, or vice versa. Th is is complex genetics. It’s not that there is one gene where there’s a relationship that causes obesity or tallness in most cases. These common conditions, or a condition like autism, have to be thought of as quantitative traits, parts of a spectrum where we all have a lot of variation.

So in autism, we see that in the parents of autistic kids, there is – especially in families that have more than one child – a clustering of very sub-threshold symptoms. They wouldn’t be called autism, but if you compared a group of parents with these traits to a group of parents who have non-autistic children, they might be characterized as more aloof or more rigid. They may have had a higher incidence of language delays as children. Those kinds of things are seen. So, the idea is it’s a kind of mixing of multiple genetic risk factors with the environment that ends up with the child with autism, but that’s not the case in every family. There clearly are different kinds of autisms. A common form of autism would be this kind of familial autism where you see very subtle signs or hints in the parents of specific aspects that relate to the condition, like the rigidity I mentioned, but you would never call the parents autistic. In fact, they’re not.

Let’s go back to the issue of psychology versus biology. Are there still people today who dispute the biological interpretation?
Dr. Geschwind:
Now that we know that the mind emanates from the brain, it’s hard to separate psychology from the biomedical because a lot of what we think and feel comes from this organ called the brain. And so, there are different ways to approach different levels and kinds of disorders. Some need medication. Some need medication plus therapy, and others are much better off without any medication and with in-depth therapy. Imagine if you have a bad experience, or a series of bad experiences, even bad parenting, that kind of experience is going to need therapy. Again, it really depends upon where the disorder really comes from, where its roots and etiology are. I think that very few people would now say that autism is not a biomedical disorder. I think the major debate at this point is how much of it is genetics and how much of it is environment.

If a percentage of the genetic mutations that are believed to be responsible for autism have been identified, are you hopeful that one day we will be able to interfere with these genetic abnormalities and effect a cure?
Dr. Geschwind:
I think it’s too early to know exactly where the genetics is going to lead us. One hope would be that we can identify biochemical pathways that are disturbed at least in a fraction of children who are amenable to treatment. We already know in mice models that if we introduce certain mutations, some of which are related to autism, that cause a severe neurodevelopmental disorder and let their brains develop, and if we then turn the gene on or off or give the adult mouse a drug that would treat the pathway, that we can have a major effect on the adult with the disease.

So, in some cases, it may not even be necessary to intervene at the developmental stage to change the brain development. We may be able to affect brain function in children or in adolescents or even adults, and that gives me a lot more hope because the difficulty of intervening in-utero with gene therapy or something like that is really off the table right now in terms of a viable therapy. So, the idea of being able to intervene later would be helpful.

On the other hand, we know that if we look at 100 children with autism, we can take one-third of the children who don’t speak, let’s say, at age 2½ or 3, and with the proper and most state-of-the-art cognitive and behavior intervention, we can get those children speaking. But that kind of therapy takes money, it takes very specialized expertise – there aren’t enough people trained to deliver it – and it’s going to take research to know how to best implement that on a large scale. But there’s enormous hope that if we took what we know now and actually could apply that in our educational and early-development systems, that we’d have an enormous impact on children.

That’s why it’s critical, even though we don’t have genetic interventions, to diagnose children with autism as early as possible, while the brain is still plastic and while these behavioral and cognitive interventions actually can do the most good. The earlier we intervene, the better chance we give a child with autism or another neurodevelopmental disability to eventually be mainstreamed or to have a successful and satisfying life.

To listen to the complete, unedited L.A. Theatre Works interview with Dr. Geschwind, go to:


Add a comment

Please note that we are unable to respond to medical questions. For information about health care, or if you need help in choosing a UCLA physician, please contact UCLA Physician Referral Service (PRS) at 1-800-UCLA-MD1 (1-800-825-2631) and ask to speak with a referral nurse. Thank you.