New technologies advance study of autism

Psychologists are key to designing neuroimaging studies that search for biological bases of autism.

By Hugh McIntosh

Dramatic advances in magnetic resonance imaging (MRI) are opening new windows onto the structures and processes of brain disorders?and enabling researchers to launch promising studies that may help unravel the mysteries of autism.

'It?s very much like how a microscope must have been to the biologists who first were able to look at cells,' says cognitive psychologist Patricia Carpenter, PhD, of Carnegie Mellon University. 'These are just incredibly powerful tools.'

MRI uses a magnetic field and strong pulses of radio waves to induce protons in the nuclei of various elements to emit characteristic radio signals, which are detected by a scanner and translated into an image by a computer.

Structural MRI has been available since the 1980s for studying brain anatomy, but recent advances in imaging analysis now permit measurements of the shape and volume of thin, complex structures, such as the cortex, in ways that were not previously possible.

Functional MRI (fMRI) allows researchers to see areas of the brain 'light up' when someone performs a brain-stimulating task, such as reading. It can be done repeatedly on children and adults because it does not involve the radioactive isotopes used in earlier functional imaging techniques. And it shows not only where brain activation is occurring, but also how hard that part of the brain is working.

Psychologists play an essential role in fMRI studies because of their knowledge of the cognitive processes that underlie many behavioral tasks. With that expertise, psychologists construct tasks that activate the brain in specific ways, allowing researchers to study cognitive processes with functional imaging.

Several studies on autism are under way using MRI and fMRI. Below are a few examples.

Imaging function

? At the University of California, Los Angeles, researchers speculate that autistic people have difficulty interpreting their emotions. To explore this idea neuropsychologist Susan Bookheimer, PhD, and her colleagues are using fMRI to study the connections between three levels of cognitive processing: perception, emotions and interpretation of emotion. The study will involve 60 high-functioning autistic people, ages 8 to 22.

The researchers will test how well participants process emotions by showing them short movie clips with neutral or strong emotional content. Because evidence from other studies suggests that the brain?s frontal lobe interprets emotions, the researchers will look for activation there when the brain?s emotion centers are stimulated by the movies, Bookheimer says.

The researchers predict that high-functioning autistic children?those who have language, who are not retarded, but who have social communication problems?are most likely to show less frontal lobe activation than children without developmental problems, indicating difficulties in making judgments about the emotions they are experiencing. ? In Pittsburgh, collaborators at two universities are using fMRI to explore the hypothesis that autism is a disorder of complex information processing, says child neurologist Nancy Minshew, MD, of the University of Pittsburgh. A related hypothesis is that these processing deficits are linked to neocortical systems and their interconnections throughout the brain.

The project will use fMRI to study language, problem-solving and visual-spatial systems in about 90 high-functioning autistic people ages 8 to 55. The researchers will examine the participants? brain areas involved in language by means of language tasks that, for example, start with simple declarative sentences and move on to more complex sentences, says Carnegie Mellon University?s Carpenter. Early results show activation in parts of the brain, such as Wernicke?s and Broca?s areas, traditionally associated with language production but also in other parts, such as the homologous regions in the right hemisphere.

'The picture that emerges is a set of areas?talking to each other,' Carpenter says. 'But the amount of team-playing depends on the particular game and how hard the game is to play.' The ability to study the parts of a complex cognitive system such as language will help scientists identify those parts of the system that are impaired in autism.

? Investigators at the Eunice Kennedy Shriver Center in Waltham, Mass., are using fMRI to study language processing in high-functioning adults with autism, ages 18 to 39. In particular, they are exploring whether the ability to name human-made objects and the ability to name natural objects lie in different brain areas, which is suggested by evidence that stroke patients may lose the ability to name artifacts but retain the ability to name animals.

Researchers want to explore this concept in autism by asking study participants to make semantic judgments about words. For example, they may ask, 'Does this word represent an artifact or a natural object?' The study should reveal whether the different terms activate different brain areas in normal subjects and, if so, whether these areas are activated in autism.

Shriver Center psychologist Helen Tager-Flusberg, PhD, speculates that different regions of the brain may be activated in autistic people compared with controls or that the same locations may be activated but to different degrees.

Because people with autism often have difficulty with social interactions, researchers are also interested in how these people process social emotional information, Tager-Flusberg says. Pilot work indicates that photos of faces expressing emotion activate the amygdala in people who have no developmental problems. The researchers will now use the same types of photos to test whether the same brain areas are active in people with autism.

Analyzing structure

Researchers use structural MRI to obtain a detailed look at the brain?s anatomy in attempts to explore the hypothesis that people with autism may have enlarged brains.

Some researchers suspect that autism could result from a breakdown in the system that normally trims unnecessary neurons in early childhood, Pittsburgh?s Minshew says. Using structural MRI, researchers are studying the size variations in many parts of the brain including cortex, white matter, limbic structures, corpus callosum, cerebellum and brain stem, as well as the whole brain.

At Yale University, researchers are using fMRI to investigate why people with autism often treat humans like objects. In their work, scientists study autistic participants who are performing two tasks?comparing photos of faces and comparing photos of objects?that activate two different brain areas in normal subjects. Researchers have found that the two tasks strongly activate just the object-processing area in autistic subjects.

'They?re using this object processing region of the brain when looking at faces,' says clinical psychologist Robert Schultz, PhD.

Thus, the patients treat the faces as common objects and not as anything special.

Shriver Center scientists will also use structural MRI to examine possible differences in the brain structure of two groups of autistic children with distinct language abilities and deficits, says Tager-Flusberg. 'Our autism subjects vary considerably in terms of their language abilities and deficits,' she says. 'We?re interested in whether that is reflected in their brain scans.'

In-depth analyses of the language abilities of these children suggest that the children fall into at least two broad subgroups, Tager-Flusberg says. Children in one group show age-appropriate syntax, semantics and articulation, but they have problems with reciprocal conversation, discourse and pragmatics (initiating an interaction, for example).

Children in the other group have similar deficits, but also have problems with articulation, especially the grammar and morphology of language.

Tager-Flusberg hypothesizes that these differences will be mirrored in structural MRI measurements of volume and morphology.

Hugh McIntosh is a writer in Chicago.