Thinking in pictures

The first sign that a baby may be autistic is that it stiffens up and resists being held and cuddled. It may be extremely sensitive to touch and respond by pulling away or screaming. More obvious symptoms of autism usually occur between twelve and twenty-four months of age. I was my mother's first child, and I was like a little wild animal. I struggled to get away when held, but if I was left alone in the big baby carriage I seldom fussed. Mother first realized that something was dreadfully wrong when I failed to start talking like the little girl next door, and it seemed that I might be deaf. Between nonstop tantrums and a penchant for smearing feces, I was a terrible two-year-old.

At that time, I showed the symptoms of classic autism: no speech, poor eye contact, tantrums, appearance of deafness, no interest in people, and constant staring off into space. I was taken to a neurologist, and when a hearing test revealed that I was not deaf, I was given the label «brain-damaged.» Most doctors over forty years ago had never heard of autism. A few years later, when more doctors learned about it, that label was applied.

I can remember the frustration of not being able to talk at age three. This caused me to throw many a tantrum. I could understand what people said to me, but I could not get my words out. It was like a big stutter, and starting words was difficult. My first few words were very difficult to produce and generally had only one syllable, such as «bah» for ball. I can remember logically thinking to myself that I would have to scream because I had no other way to communicate. Tantrums also occurred when I became tired or stressed by too much noise, such as horns going off at a birthday party. My behavior was like a tripping circuit breaker. One minute I was fine, and the next minute I was on the floor kicking and screaming like a crazed wildcat.

I can remember the day I bit my teacher's leg. It was late in the afternoon and I was getting tired. I just lost it. But it was only after I came out of it, when I saw her bleeding leg, that I realized I had bitten her. Tantrums occurred suddenly, like epileptic seizures. Mother figured out that like seizures, they had to run their course. Getting angry once a tantrum started just made it worse. She explained to my elementary school teachers that the best way to handle me if I had a tantrum was not to get angry or excited. She learned that tantrums could be prevented by getting me out of noisy places when I got tired. Privileges such as watching Howdy Doody on TV were withdrawn when I had a bad day at school. She even figured out that I'd sometimes throw a tantrum to avoid going to class.

When left alone, I would often space out and become hypnotized. I could sit for hours on the beach watching sand dribbling through my fingers. I'd study each individual grain of sand as it flowed between my fingers. Each grain was different, and I was like a scientist studying the grains under a microscope. As I scrutinized their shapes and contours, I went into a trance which cut me off from the sights and sounds around me.

Rocking and spinning were other ways to shut out the world when I became overloaded with too much noise. Rocking made me feel calm. It was like taking an addictive drug. The more I did it, the more I wanted to do it. My mother and my teachers would stop me so I would get back in touch with the rest of the world. I also loved to spin, and I seldom got dizzy. When I stopped spinning, I enjoyed the sensation of watching the room spin.

Today, autism is regarded as an early childhood disorder by definition, and it is three times more common in boys than girls. For the diagnosis to be made, autistic symptoms must appear before the age of three. The most common symptoms in young children are no speech or abnormal speech, lack of eye contact, frequent temper tantrums, oversensitivity to touch, the appearance of deafness, a preference for being alone, rocking or other rhythmic stereotypic behavior, aloofness, and lack of social contact with parents and siblings. Another sign is inappropriate play with toys. The child may spend long periods of time spinning the wheel of a toy car instead of driving it around on the floor.

Diagnosing autism is complicated by the fact that the behavioral criteria are constantly being changed. These criteria are listed in the Diagnostic and Statistical Manual published by the American Psychiatric Association. Using those in the third edition of the book, 91 percent of young children displaying autistic symptoms would be labeled autistic. However, using the newest edition of the book, the label would apply to only 59 percent of the cases, because the criteria have been narrowed.

Many parents with an autistic child will go to many different specialists looking for a precise diagnosis. Unfortunately, diagnosing autism is not like diagnosing measles or a specific chromosomal defect such as Down syndrome. Even though autism is a neurological disorder, it is still diagnosed by observing a child's behavior. There is no blood test or brain scan that can give an absolute diagnosis, though brain scans may partially replace observation in the future.

The new diagnostic categories are autism, pervasive developmental disorder (PDD), Asperger's syndrome, and disintegrative disorder, and there is much controversy among professionals about them. Some consider these categories to be true separate entities, and others believe that they lie on an autistic continuum and there is no definite distinction between them.

A three-year-old child would be labeled autistic if he or she lacked both social relatedness and speech or had abnormal speech. This diagnosis is also called classic Kanner's syndrome, after Leo Kanner, the physician who first described this form of autism, in 1943. These individuals usually learn to talk, but they remain very severely handicapped because of extremely rigid thinking, poor ability to generalize, and no common sense. Some of the Kanner people have savant skills, such as calendar calculation. The savant group comprises about 10 percent of the children and adults who are diagnosed.

A child with classic Kanner's syndrome has little or no flexibility of thinking or behavior. Charles Hart describes this rigidity in his autistic brother, Sumner, who had to be constantly coached by his mother. He had to be told each step of getting undressed and going to bed. Hart goes on to describe the behavior of his autistic son, Ted, during a birthday party when ice cream cones were served. The other children immediately began to lick them, but Ted just stared at his and appeared to be afraid of it. He didn't know what to do, because in the past he had eaten ice cream with a spoon.

Another serious problem for people with Kanner's syndrome is lack of common sense. They can easily learn how to get on a bus to go to school, but have no idea what to do if something interrupts the routine. Any disruption of routine causes a panic attack, anxiety, or a flight response, unless the person is taught what to do when something goes wrong. Rigid thinking makes it difficult to teach people with Kanner-type autism the subtleties of socially appropriate behavior. For example, at an autism meeting, a young man with Kanner's syndrome walked up to every person and asked, «Where are your earrings?» Kanner autistics need to be told in a clear simple way what is appropriate and inappropriate social behavior.

Uta Frith, a researcher at the MRC Cognitive Development Unit in London, has found that some people with Kanner's syndrome are unable to imagine what another person is thinking. She developed a «theory of mind» test to determine the extent of the problem. For example, Joe, Dick, and a person with autism are sitting at a table. Joe places a candy bar in a box and shuts the lid. The telephone rings, and Dick leaves the room to answer the phone. While Dick is gone, Joe eats the candy bar and puts a pen in the box. The autistic person who is watching is asked, «What does Dick think is in the box?» Many people with autism will give the wrong answer and say «a pen.» They are not able to figure out that Dick, who is now outside the room, thinks that the box still contains a candy bar.

People with Asperger's syndrome, who tend to be far less handicapped than people with Kanner-type autism, can usually pass this test and generally perform better on tests of flexible problem-solving than Kanner's syndrome autistics. In fact, many Asperger individuals never get formally diagnosed, and they often hold jobs and live independently. Children with Asperger's syndrome have more normal speech development and much better cognitive skills than those with classic Kanner's. Another label for Asperger's syndrome is «high-functioning autism.» One noticeable difference between Kanner's and Asperger's syndromes is that Asperger children are often clumsy. The diagnosis of Asperger's is often confused with PDD, a label that is applied to children with mild symptoms which are not quite serious enough to call for one of the other labels.

Children diagnosed as having disintegrative disorder start to develop normal speech and social behavior and then regress and lose their speech after age two. Many of them never regain their speech, and they have difficulty learning simple household chores. These individuals are also referred to as having low-functioning autism, and they require supervised living arrangements for their entire lives. Some children with disintegrative disorder improve and become high-functioning, but overall, children in this category are likely to remain low-functioning. There is a large group of children labeled autistic who start to develop normally and then regress and lose their speech before age two. These early regressives sometimes have a better prognosis than late regressives. Those who never learn to talk usually have severe neurological impairments that show up on routine tests. They are also more likely to have epilepsy than Kanner or Asperger children. Individuals who are low-functioning often have very poor ability to understand spoken words. Kanner, Asperger, and PDD children and adults usually have a much better ability to understand speech.

Children in all of the diagnostic categories benefit from placement in a good educational program. Prognosis is improved if intensive education is started before age three. I finally learned to speak at three and a half, after a year of intensive speech therapy. Children who regress at eighteen to twenty-four months of age respond to intensive educational programs when speech loss first occurs, but as they become older they may require calmer, quieter teaching methods to prevent sensory overload. If an educational program is successful, many autistic symptoms become less severe.

The only accurate way to diagnose autism in an adult is to interview the person about his or her early childhood and obtain descriptions of his or her behavior from parents or teachers. Other disorders with autistic symptoms, such as acquired aphasia (loss of speech), disintegrative disorder, and Landau-Kleffner syndrome, occur at an older age. A child may have normal or near-normal speech and then lose it between the ages of two and seven. In some cases disintegrative disorder and Landau-Kleffner syndrome may have similar underlying brain abnormalities. Landau-Kleffner syndrome is a type of epilepsy that often causes a child to lose speech. Small seizures scramble hearing and make it difficult or impossible for the child to understand spoken words. A proper diagnosis requires very sophisticated tests, because the seizures are difficult to detect. They will not show up on a simple brain-wave (EEG) test. These disorders can often be successfully treated with anticonvulsants (epilepsy drugs) or corticosteroids such as prednisone. Anticonvulsant medications may also be helpful to autistic children who have abnormal EEGs or sensory scrambling. Other neurological disorders that have symptoms of autism are Fragile X syndrome, Rhett's syndrome, and tuberous sclerosis. Educational and treatment programs that help autistic children are usually helpful for children with these disorders also.

There is still confusion in diagnosing between autism and schizophrenia. Some professionals claim that children with autism develop schizophrenic characteristics in adulthood. Like autism's, schizophrenia's current diagnostic criteria are purely behavioral, though both are neurological disorders. In the future, brain scans will be sophisticated enough to provide an accurate diagnosis. Thus far, brain research has shown that these conditions have different patterns of abnormalities. By definition, autism starts in early childhood, while the first symptoms of schizophrenia usually occur in adolescence or early adulthood. Schizophrenia has two major components, the positive symptoms, which include full-blown hallucinations and delusions accompanied by incoherent thinking, and the negative symptoms, such as flat, dull affect and monotone speech. These negative symptoms often resemble the lack of affect seen in adults with autism.

In the British Journal of Psychiatry, Dr. P. Liddle and Dr. T. Barnes wrote that schizophrenia may really be two or three separate conditions. The positive symptoms are entirely different from symptoms of autism, but the negative ones may partially overlap with autistic symptoms. Confusion of the two conditions is the reason that some doctors attempt to treat autism with neuroleptic drugs such as Haldol and Mellaril. But neuroleptics should not be the first-choice medications for autism, because other, safer drugs are often more effective. Neuroleptic drugs have very severe side effects and can damage the nervous system.

Over ten years ago, Dr. Peter Tanguay and Rose Mary Edwards, at UCLA, hypothesized that distortion of auditory input during a critical phase in early childhood development may be one cause of handicaps in language and thinking. The exact timing of the sensory processing problems may determine whether a child has Kanner's syndrome or is a nonverbal, low-functioning autistic. I hypothesize that oversensitivity to touch and auditory scrambling prior to age two may cause the rigidity of thinking and lack of emotional development found in Kanner-type autism. These children partially recover the ability to understand speech between the ages of two and a half and three. Disintegrative disorder children, who develop normally up to two years of age, may be more emotionally normal because emotional centers in the brain have had an opportunity to develop before the onset of sensory processing problems. It may be that a simple difference in timing determines which type of autism develops. Early sensory processing problems may prevent development of the emotional centers of the brain in Kanner-type autistics, while the acquisition of language is more disturbed when sensory processing difficulties occur slightly later.

Research has very clearly shown that autism is a neurological disorder that reveals distinct abnormalities in the brain. Brain autopsy research by Dr. Margaret Bauman has shown that those with both autism and disintegrative disorder have immature development of the cerebellum and the limbic system. Indications of a delay in brain maturation can also be seen in autistic children's brain waves. Dr. David Canter and his associates at the University of Maryland found that low-functioning children between the ages of four and twelve have EEG readings that resemble the brain-wave pattern of a two-year-old. The question is what causes these abnormalities. Studies by many researchers are showing that there may be a cluster of genes that can put a person at risk for many disorders, including autism, depression, anxiety, dyslexia, attention deficit disorder, and other problems.

There is no single autism gene, though most cases of autism have a strong genetic basis. If a person is autistic, his or her chances of having an autistic child are greatly increased. There is also a tendency for the siblings of autistic children to have a higher incidence of learning problems than other children. Studies by Susan Folstein and Mark Rutter in London showed that in 42 percent of the families surveyed, either a sibling or a parent of an autistic child had delayed speech or learning problems.

Genetics, however, does not completely control brain development. Studies of identical twins by Folstein and Rutter show that sometimes one twin is severely autistic and the other has only a few autistic traits. MRI (magnetic resonance imaging) brain scans of identical twin schizophrenics have shown that the more severely afflicted twin has greater brain abnormalities. The brain is so complex that genetics cannot tell every little developing neuron exactly where it should be connected. There is a 10 percent variation in brain anatomical structure that is not controlled by genetics. Brain scans of normal identical twins by Michael Gazzaniga, at the Dartmouth Medical School, showed an easily observable variation in brain structure, but twins' brains are more similar than the brains of unrelated people. Likewise, the personalities of identical twins are similar. Studies at the University of Minnesota by Thomas Bouchard and his colleagues of twins reared in different families show that basic traits such as mathematical ability, athletic ability, and temperament are highly inheritable. A summary of these studies concluded that roughly half of what a person becomes is determined by genetics and the other half is determined by environment and upbringing.

Other theories suggest that if a fetus is exposed to certain toxins and viruses, these may interact with genes to cause the abnormal brain development typical of autism. If either parent is exposed to chemical toxins that slightly damage his or her genetic material, that could increase the likelihood of autism or some other developmental disorder. Some parents suspect that an allergic reaction to early childhood vaccinations triggers autistic regression. If this is true it is likely that the vaccine interacts with genetic factors. Another possibility is immune system abnormalities which interfere with brain development. However, there is still too much that is not known, and neither parent should be held responsible for an autistic child. Scientific studies and interviews with families indicate that both the father's and the mother's side contribute genetically to autism.

The Autistic Continuum

Countless researchers have attempted to figure out what factors determine the difference between high- and low-functioning autism. High-functioning children with Kanner's or Asperger's syndrome usually develop good speech and often do well academically Low-functioning children are often unable to speak or can say only a few words. They also have trouble learning simple skills such as buttoning a shirt. At age three, both types have similar behaviors, but as they grow older the difference becomes more and more apparent.

When my speech therapist held my chin and directed me to look at her, it jerked me out of my private world, but for others forcing eye contact can cause the opposite reaction — brain overload and shutdown. For instance, Donna Williams, the author of Nobody Nowhere, explained that she could use only one sensory channel at a time. If a teacher had grabbed her chin and forced eye contact, she would have turned off her ears. Her descriptions of sensory jumbling provide an important bridge to understanding the difference between high-functioning and low-functioning autism, which I would describe as a sensory processing continuum. At one end of the continuum is a person with Asperger's or Kanner's autism who has mild sensory oversensitivity problems, and at the other end of the spectrum is the low-functioning person who receives jumbled, inaccurate information, both visually and aurally.

I was able to learn to speak because I could understand speech, but low-functioning autistics may never learn to speak because their brains cannot discriminate among speech sounds. Many of these people are mentally retarded, but a few individuals may have a near-normal brain trapped inside a sensory system that does not work. Those who escape the prison of low-functioning autism probably do so because just enough undistorted information gets through. They do not totally lose contact with the world around them.

Twenty years ago, Carl Delacato, a therapist who worked with autistic children, speculated that low-functioning individuals may have «white noise» in their sensory channels. In his book The Ultimate Stranger, he described three kinds of sensory processing problems: hyper, hypo, and white noise. Hyper means oversensitive, hypo means undersensitive, and white noise means internal interference.

In questioning many people with autism, I soon found that there was a continuum of sensory abnormalities that would provide insight into the world of nonverbal people with autism. I imagine that the extent of sensory jumbling they experience would be equivalent to taking Donna's sensory problems and multiplying them tenfold. I am lucky in that I responded well when my mother, teachers, and governess kept encouraging social interaction and play. I was seldom allowed to retreat into the soothing world of rocking or spinning objects. When I daydreamed, my teachers yanked me back to reality.

Almost half of all very young children with autism respond well to gently intrusive programs in which they are constantly encouraged to look at the teacher and interact. Brightly colored wall decorations made learning fun for me, but they may be too distracting for a child with sensory jumbling. The popular Lovaas program, developed at UCLA, is being used successfully there to mainstream nearly half of young autistic children into a normal kindergarten or first grade. The Lovaas method pairs words with objects, and the children are rewarded with praise and food when they correctly match a word with an object. While this program is wonderful for some kids, it is certain to be confusing and possibly painful for children with severe sensory jumbling and mixing problems.

These children require a different approach. Touch is often their most reliable sense, and they learn best if teachers use a tactile system. One mother taught her nonverbal daughter to draw a circle by holding her hand and guiding it to make a circle. Plastic letters that can be felt are often useful for teaching words. The more protected these children are from distracting sights and sounds, the more likely it is that their dysfunctional nervous system will be able to perceive speech accurately To help them hear better, teachers must protect them from visual stimuli that will cause sensory overload. They may hear best in a quiet, dimly illuminated room that is free of fluorescent lights and bright wall decorations. Sometimes hearing is enhanced if the teacher whispers or sings softly. Teachers need to speak slowly to accommodate a nervous system that processes information slowly. And sudden movements that will cause sensory confusion should also be avoided.

Children who are echolalic — who repeat what they hear— may be at a midpoint on the sensory processing continuum. Enough recognizable speech gets through for them to be able to repeat the words. Dr. Doris Allen, at the Albert Einstein Hospital in New York, emphasizes that echolalia should not be discouraged, so as not to inhibit speech. The child repeats what has been said to verify that he heard it correctly. Research by Laura Berk, at Illinois State University, has shown that normal children talk to themselves to help them control their behavior and learn new skills. Since autism is caused by immature brain development, it is likely that echolalia and self-talking, which occur in older autistic children, are the result of immature speech patterns.

Unlike normal children, who naturally connect language to the things in their lives at a remarkable rate, autistic children have to learn that objects have names. They have to learn that words communicate. All autistic children have problems with long strings of verbal information. Even very high-functioning people have difficulty following verbal instructions and find it easier to follow written instructions, since they are unable to remember the sequence of the information. My college math teacher once commented that I took excessive notes. He told me that I should pay attention and understand the concept. The problem was that it was impossible for me to remember the sequence of the problems without the notes. I learned to read with phonics and sounding out words, because I was able to understand speech by age three. Children with more severe auditory processing problems often learn to read before they can speak. They learn best if a written word is paired with an object, because many of them have very poor comprehension of spoken words.

As an adult my method for learning a foreign language may be similar to how a more severely impaired autistic child learns to understand language. I cannot pick words out of a conversation in a foreign language until I have seen them written first.

Two basic patterns of autistic symptoms can help identify which children will respond well to intensive, gently intrusive teaching methods, and which will not. The first kind of child may appear deaf at age two, but by age three he or she can understand speech. I was this kind. When adults spoke directly to me, I could understand them, but when they talked among themselves, it sounded like gibberish. The second kind of child appears to develop normally until one and a half or two and then loses speech. As the syndrome progresses, the ability to understand speech deteriorates and autistic symptoms worsen. A child that has been affectionate withdraws into autism as his sensory system becomes more and more scrambled. Eventually he may lose awareness of his surroundings, because his brain is not able to process and understand sights and sounds around him. There are also children who are mixtures of the two kinds of autism.

Children of the first kind will respond well to intensive, structured educational programs that pull them out of the autistic world, because their sensory systems provide a more or less accurate representation of things around them. There may be problems with sound or touch sensitivity, but they still have some realistic awareness of their surroundings. The second kind of child may not respond, because sensory jumbling makes the world incomprehensible. Gently intrusive teaching methods will work on some children who lose their speech before age two if teaching is started before their senses become totally scrambled. Catherine Maurice describes her successful use of the Lovaas program with her two children, who lost speech at fifteen and eighteen months of age, in her book, Let Me Hear Your Voice. Teaching was started within six months of the onset of symptoms. The regression into autism was not complete, and her children still had some awareness. If she had waited until they were four or five, it is very likely that the Lovaas method would have caused confusion and sensory overload.

My experience and that of others has shown that an effective teaching method coupled with reasonable amounts of effort should work. Desperate parents often get hooked into looking for magic cures that require ten hours a day of intensive treatment. To be effective, educational programs do have to be done every day, but they usually do not require heroic amounts of effort. My mother spent thirty minutes five days a week for several months teaching me to read. Mrs. Maurice had a teacher spend twenty hours a week on the Lovaas method with her children. In addition to participating in formal educational programs, young autistic children need a structured day, both in the school and at home. Several studies have shown that twenty to twenty-five hours a week of intensive treatment which required the child to constantly interact with his teacher was most effective. A neurologist gave my mother some very good advice: to follow her own instincts. If a child is improving in an educational program, then it should be continued, but if there is no progress, something else should be tried. Mother had a knack for recognizing which people could help me and which ones could not. She sought out the best teachers and schools for me, in an era when most autistic children were placed in institutions. She was determined to keep me out of an institution.

A controversial technique called facilitated communication is now being used with nonverbal people with autism. Using the technique, the teacher supports the person's hand while he or she taps out messages on a typewriter keyboard. Some severely handicapped people have problems with stopping and starting hand movements, and they also have involuntary movements that make typing difficult. Supporting the person's wrist helps to initiate motion of the hand toward the keyboard and pulls his fingers off the keyboard after he pushes a key to prevent perseveration and multiple pushing of a single key. Merely touching the person's shoulder can help him initiate hand movements.

Several years ago, facilitated communication was hailed as a major breakthrough, and wild claims were made that the most severely handicapped autistic people had completely normal intelligence and emotions. Fifty scientific studies have now shown that in the vast majority of cases, the teacher was moving the person's hand, as if it were a planchet on a Ouija board. The teacher was communicating, instead of the person with autism. A summary of forty-three studies in the Autism Research Review showed that 5 percent of nonverbal, severely handicapped people can communicate with simple one-word responses. In the few cases where facilitated communication has been successful, someone has spent many hours teaching the person to read first.

It is likely that the truth about facilitated communication is somewhere between wishful hand-pushing and real communication. Carol Berger, of New Breakthroughs in Eugene, Oregon, found that low-functioning autistics could achieve 33 percent to 75 percent accuracy in typing one-word answers. Some of the poor results in controlled studies may have been due to sensory overload caused by the presence of strange people. Reports from parents indicate that a few adults and children initially need wrist support and then gradually learn to type independently. But the person must know how to read, and facilitator influence cannot be completely ruled out until wrist or arm support is removed.

Parents who are desperate to reach their autistic children often look for miracles. It's hard not to get caught up in new promises of hope, because there have been so few real breakthroughs in the understanding of autism.

The Autistic Continuum

It appears that at one end of the spectrum, autism is primarily a cognitive disorder, and at the other end, it is primarily a sensory processing disorder. At the severely impaired sensory processing end, many children may be diagnosed as having disintegrative disorder. At a midpoint along the spectrum, autistic symptoms appear to be caused by equal amounts of cognitive and sensory problems. There can be mild and severe cases at all points along the continuum. Both the severity and the ratio of these two components are variable, and each case of autism is different. When a person with autism improves because of either educational or medical intervention, the severity of a cognitive or sensory problem may diminish, but the ratio between the two seems to stay the same. What remains inexplicable, however, are rigid thinking patterns and lack of emotional affect in many high-functioning people. One of the perplexing things about autism is that it is almost impossible to predict which toddler will become high-functioning. The severity of the symptoms at age two or three is often not correlated with the prognosis.

The world of the nonverbal person with autism is chaotic and confusing. A low-functioning adult who is still not toilet-trained may be living in a completely disordered sensory world. It is likely that he has no idea of his body boundaries and that sights, sounds, and touches are all mixed together. It must be like seeing the world through a kaleidoscope and trying to listen to a radio station that is jammed with static at the same time. Add to that a broken volume control, which causes the volume to jump erratically from a loud boom to inaudible. Such a person's problems are further compounded by a nervous system that is often in a greater state of fear and panic than the nervous system of a Kanner-type autistic. Imagine a state of hyperarousal where you were being pursued by a dangerous attacker in a world of total chaos. Not surprisingly, new environments make low-functioning autistics fearful.

Puberty often makes the problem worse. Birger Sellin describes in his book I Don't Want to Be Inside Me Anymore how his well-behaved son developed unpredictable screaming fits and tantrums at puberty. The hormones of adolescence further sensitized and inflamed an overaroused nervous system. Dr. John Ratey, at Harvard University, uses the concept of noise in the nervous system to describe such hyperarousal and confusion. Medications such as beta-blockers and clonidine are often helpful because they can calm an overaroused sympathetic nervous system.

Autistics with severe sensory problems sometimes engage in self-injurious behavior such as biting themselves or hitting their heads. Their sensory sensations are so disordered that they may not realize they are hurting themselves. Though a recent study by Reed Elliot published in the Journal of Autism and Developmental Disabilities showed that very vigorous aerobic exercise reduced aggression and self-injury in half of mentally retarded autistic adults, educational and behavioral training will help almost all people with autism to function better. Early intervention in a good program can enable about 50 percent of autistic children to be enrolled in a normal first grade. Though most autistics will not function at my level, their ability to live a productive life will be improved. Medication can help reduce the hyperarousal of many low-functioning older children and help them control their behavior. Many nonverbal autistics are capable of doing simple jobs such as washing windows or routine manual work. Few nonverbal autistic adults are able to read and are capable of doing normal schoolwork.

Many parents and teachers have asked me where I fit on the autistic continuum. I still have problems with rapid responses to unexpected social situations. In my business dealings I can handle new situations, but every once in a while I panic when things go wrong. I've learned to deal with the fear of traveling, so that I have a backup plan if, for example, my plane is late. I have no problems if I mentally rehearse every scenario, but I still panic if I'm not prepared for a new situation, especially when I travel to a foreign country where I am unable to communicate. Since I can't rely on my library of social cues, I feel very helpless when I can't speak the language. Often I withdraw.

If I were two years old today, I would be diagnosed with classic Kanner's syndrome, because I had delayed abnormal speech development. However, as an adult I would probably be diagnosed as having Asperger's syndrome, because I can pass a simple theory-of-mind test and I have greater cognitive flexibility than a classic Kanner autistic. All of my thinking is still in visual images, though it appears that thinking may become less visual as one moves along the continuum away from classic Kanner's syndrome. My sensory oversensitivities are worse than the mild difficulties some Kanner autistics have, but I do not have sensory mixing and jumbling problems. Like most autistics, I don't experience the feelings attached to personal relationships. My visual world is a literal one, though I have made progress by finding visual symbols to carry me beyond the fixed and rigid worlds of other people with classic Kanner autism.

In an article written by Oliver Sacks in The New Yorker, I was quoted as saying, «If I could snap my fingers and be nonautistic, I would not. Autism is part of what I am.» In contrast, Donna Williams says, «Autism is not me. Autism is just an information processing problem that controls who I am.» Who is right? I think we both are, because we are on different parts of the autism spectrum. I would not want to lose my ability to think visually. I have found my place along the great continuum.

Update: Diagnosis and Education

Both parents and teachers make the mistake of thinking a diagnosis of autism, PDD (Pervasive Developmental Disorder), ADHD (Attention Deficit Hyperactivity Disorder), or Asperger's is precise. It is not precise the way a diagnosis for measles or meningitis is precise. It is a behavioral profile and different doctors and psychologists often come up with a different diagnosis because they interpret the child's behavior differently. At the time of writing this update, there is no definitive brain imaging or laboratory test for the diagnosis of autism.

Since Thinking in Pictures was written, the mild Asperger diagnosis is being used more and more. At the many autism conferences that I attend, I am observing more and more very smart children with a diagnosis of Asperger's. Some of these children should be in a gifted and talented class instead of being sent to special education. There are other Asperger's individuals who may need special education in their area of weakness and be in an advanced class in their area of strength. I am worried that students who would be capable of a challenging career in science, engineering, or computers may be shunted into a special education rut. In fairness to special education teachers, it is difficult to work with a spectrum that can range from nonverbal to genius.

Diane Kennedy, author of ADHD Autism Connection, was one of the first people to write about the confusion of Asperger's with attention deficit problems. I talk to more and more parents of children with a diagnosis that switches back and forth between Asperger's and ADHD. Many parents have told me that stimulant ADHD medications such as Ritalin (metehylphenidate) and Adderall (a combination of four different types of amphetamines) have greatly helped their children. It is likely that some individuals on the high-functioning end of the autism spectrum share traits with ADHD. Children or adults who have more classical types of autism or are nonverbal often become agitated and worse on stimulants. A trial of only one or two pills is all that is needed to determine if stimulants will be helpful or terrible.

Brain Research and Early Diagnosis

During the last ten years, there has been an increased understanding of autistic brain abnormalities. A normal child's brain grows at a steady rate. Detailed brain scans of autistic children in Dr. Eric Courchesne's lab indicated that in the first year of life there is premature overgrowth of the brain followed by an arrest of growth. Children with greater amounts of abnormal overgrowth usually have more severe autism. Research has also shown that the serotonin systems in the autistic child's brain are highly abnormal. This may explain why doses for SSRI antidepressants often need o be kept very low to prevent agitation. The degree and pattern of abnormal overgrowth will be highly variable from child to child. David Amarel at the University of California found that the variability of overgrowth was greatest in low-functioning autism. He also discovered that the immune system is often abnormal and may affect the brain.

The excess of brain overgrowth causes the infant's head to become abnormally large between the ages of one and two. Later in childhood, the head size returns to normal due to later undergrowth of the brain. Measuring a young infant's head circumference (hat size) with a tape measure could be used as a simple screening tool for detecting babies who might be at risk for autism.

Other early screening tools that are being developed test for joint attention. Joint attention occurs when normal babies orient and follow an adult's gaze. When the adult is playing a little game, asking the baby to look at the pretty birdie, the baby will look where the adult is looking. The infant at risk for developmental problems will not follow an adult's gaze. Patricia Kohl at the University of Washington is working on another screening tool. This tool will detect children at risk for developmental problems who do not orient toward normal speech sounds. This is due to being unable to hear consonant sounds. Normal babies prefer to listen to «motherese»—expressive slowed down speech where the mother enunciates the words. Autistic babies prefer computerized warbling nonspeech sounds. The test would be conducted by observing the infant to determine which sounds he orients toward.

Early Education

Both scientific studies and practical experience have fully confirmed that young children with autism need at least twenty hours a week of intensive one to one teaching by an adult. All experts agree that the worst thing to do with an autistic two- to five-year-old is to let him watch TV all day. There is much debate about the best early education programs. I have observed that the best teachers tend to use the same methods regardless of the theoretical basis of the program. A review of teaching methods by Sally Rogers at the University of California at Davis indicated that discrete trial or ABA (Applied Behavioral Analysis) teaching methods were the most effective to get language started. This structured highly repetitive method helps jump-start language in young two- to five-year-olds. The discrete trial programs used today are usually more natural and less rigid than the older Lovaas method. To teach socialization and play skills methods such as Greenspan's floor-time and Dr. Lynn Kern Koegel's program are more effective. Dr. Koegel's book Overcoming Autism is full of practical teaching methods. In the floor-time method, the teacher engages the child in many interactive games and encourages social play.

Autism and PDD are highly variable and the methods that work for each child should be used. Dr. Koegel found that some little children respond well to a highly structured Lovaas-style program and other types of autistic children, who are more socially engaged, may make more progress with a less structured program. Do not get too single-minded on one method. Use things that work and eliminate things that do not work. Sometimes a combination of methods is best. For older high-functioning children, highly repetitious programs are boring and they need lessons that will stimulate their minds. In elementary school children a child's fixation can be used to motivate learning. If a child loves trains, then read a book about a train or do a math problem involving trains.

If shooting-type video games had been available when I was little, I would have become a total addict and I may not have developed more career-related interests such as building things or flying kites and airplanes. The video games with lots of rapid movement are the most addictive. For me, rapid movement video games would have just been another way to «stim» and «zone out.» I would rather encourage the older child to become really interested in doing science on a computer or learning programming. Free software is available that will turn a kid's computer into part of a super computer that crunches numbers on a real scientific project. The May 6, 2005, issue of Science is devoted to these fascinating projects. Looking at the NASA Web site and following a space probe during its journey is a wonderful way to use computers. The problem with video games is that both parents and teachers tell me that some students get so addicted that they have no other interests. I get hypnotized by screen savers with changing patterns that move rapidly. I cannot stop looking at them and for me to get any work done I have to shut them off. Video games or screen savers that move slowly do not have this effect.

Totally banning shooting-type games is probably a bad idea, but the time playing them should be severely limited. This is especially important for a child like me. They provide an activity that the autistic child can discuss with other kids at school and this may help the child socially. However, I want to direct the autistic child's interests into more constructive activities.

Genetics and Autism

Research during the last ten years confirms that autism, PDD, and Asperger's all have a strong genetic basis. Craig Newschaffer, Johns Hopkins School of Medicine, estimates that 60 to 90 percent of autism cases are genetic. Dr. Isabel Rapin and her colleagues at Albert Einstein College of Medicine reviewed papers published between 1961 and 2003. They concluded that interactions between multiple genes explain the highly variable nature of autism. Genome scans of families with many cases of autism indicate that at least ten genes are involved. They also found that the probability of having a second autistic child is 2 to 8 percent. Researchers have also confirmed previous studies that show that relatives of people with autism will often have many milder autistic-like symptoms. I have observed that the probability of having a child with low-functioning autism increases when both parents and their families have many autistic traits.

Many computer programmers exhibit autistic traits. Steve Silberman asked in an article entitled «The Geek Syndrome» in Wired magazine — are math and tech genes to blame? The computer and technical industries depend on people with attention to detail. The real social people are not interested in computers. Herbert Schreir of the Children's Hospital in Oakland, California, believes that intermarriage of «techies» explains why people have noticed high pockets of autism around Stanford and MIT Universities.

In 2004 and 2005, my webmaster for www.grandin.com (my livestock website) started giving me a list every month of the cities with the most hits on my webpage. Month after month, Redmond, Washington, where Microsoft is located and San Mateo, California, near Stanford University are in my top five cities. There is a total of one hundred cities on the list. The number one page downloaded is the first chapter of Thinking in Pictures. Even though my site is a livestock site, the autism book chapter gets the heaviest traffic. Is this because people in these areas are especially interested in the ways brains work, or does autism affect them more directly?

There are differences of opinion in the autism field about the relationship between autism and Asperger's. Are they really separate syndromes? Family and genetic studies done in the United Kingdom indicate that autism and Asperger's are part of the same spectrum. Research by Fred Volkmar at Yale showed that Asperger individuals with no speech delay are often poor at a visual thinking task such as the block design test on the WISC and high-functioning autistic individuals are more likely to be good at this test. In the block design test, the task is to assemble colored blocks to match patterns shown in a book. This difference could be explained by the differences in where the «computer cables» hook up. The underlying brain abnormality of underconnectivity problems would still be similar.

There is concern among people with Asperger's that genetic testing could eliminate them. This would be a terrible price to pay. Many gifted and talented people could be wiped out. A little bit of autism genetics may provide an advantage though too much creates a low-functioning, nonverbal individual. The development of genetic tests for autism will be extremely controversial.

Autism Epidemic

Many researchers agree that the increase in Asperger's syndrome is mostly increased detection. People who used to be labeled as science geeks or computer nerds are now diagnosed with Asperger's. Research in Sweden by Christopher Gillberg showed that some severe cases that used to be labeled mentally retarded are now labeled autistic. Another cause of the increase may be changes made to the DSMIV (Diagnostic and Statistical Manual) published by the American Psychiatric Association in 1994 to expand the diagnostic criteria to include Asperger's and Pervasive Developmental Disorders (PDD). The Centers for Disease Control (CDC) estimate that there are three to four autism cases per one thousand children. A CDC study in Atlanta, Georgia, indicated that 40 percent of all children on the spectrum are only diagnosed at school and 41 percent of special education students are on the autism spectrum. A fully verbal child with mild Asperger's will often not have any problems until he/she enrolls in school. Unfortunately there are severe cases of autism who do not receive services until they go to school. From my own observations there is one type of autism that I think has increased. The regressive type where the child loses language at age eighteen to twenty-four months. David Geier and Mark Geir, two autism consultants, state that exposure to mercury causes regression-type autism. Mercury has now been removed from many vaccines, but fish and power plant emissions are other sources of mercury. Other scientists question the effect of mercury in the incidence of autism.

There is increasing concern about environmental effects on the fetus during pregnancy. If these factors affect the incidence of autism, they probably could interact with susceptible genetics. An outside insult like toxic exposure could turn a brilliant Asperger baby into a nonverbal one. This is purely speculation. New research supports the idea that genetics susceptibility interacts with environmental insults. Scientists have developed a genetic line of mice that are highly susceptible to mercury toxicity. When the mice are given injections that mimic a vaccination schedule the normal mice have no ill effects and the susceptible mice develop autistic-like symptoms such as tail chewing and repetitive behaviors. Possibly there are some children who would have a similar susceptibility to mercury. Mady Horning at the Columbia University School of Public Health has a three-strikes model. The factors that all interact with each other to cause a developmental disability are:

1. Genetic susceptibility

2. Exposure to a toxic agent

3. The timing during development that exposure to a toxic agent occurs. A toxic agent may have no effect at one stage of development and bad effects at another stage.

Twin studies show further evidence of an interaction between environment and genetics. Mady Horning states that the concordance rate for autism in genetically identical twins is 90 percent. This means that 90 percent of the time both twins are autistic. In genetically different nonidentical twins the concordance rate is 35 percent and the autism rate in siblings is 4 percent. Further information on the mercury controversy can be found at the Autism Research Institute in San Diego, California, or in a new book by David Kirby entitled Evidence of Harm.