Researchers examine links between the brain and behavior in autism

Behaviors that are part of normal social development include noticing social information in the environment, paying attention and recognizing that the information is important, imitating other people and becoming sensitive to other people's emotional cues. Eventually, after developing these social skills, children form what is known as the theory of mind­the ability to recognize that others have feelings, beliefs and intentions that are different from their own. In the course of typical development, this occurs somewhere between the ages of 3 and 5. However, according to a widely held notion about the nature of autism, children with the disorder never fully develop a theory of mind.

Dawson and Meltzoff, both professors of psychology, are trying to better understand how development of social skills is stymied. They are investigating specific neuropsychological impairments in children with autism, and examining how differences in early social behaviors among individual children with autism are related to performance on behavioral tasks known to be linked to certain parts of the brain. This aspect of the study will dovetail with other parts of the longitudinal study aimed at determining whether there are different subtypes of autism.

The researchers are especially focusing on two abilities that typically develop in the first year of life­face recognition and emotion recognition. "We think that in terms of early social relationships, the abilities to recognize someone familiar and then to discern and make sense out of their emotional cues are going to be really critical," says Dawson.

Measuring these abilities in non-verbal children, like many children with autism, could be problematic. But, by plotting the pattern of electrical activity of the brain with a non-invasive technique known as event-related potential (ERP), the researchers can discern if a child's brain has registered differences or similarities between two stimuli presented one after the other. No overt behavioral response is required, so Dawson and Meltzoff can gather data for comparison, regardless of whether a child has language. They will use the ERP technology in conjunction with tasks designed to measure a child's ability to notice and recognize the importance of social information, imitate another person, register emotional cues, and other precursors to developing a theory of mind.

The ERP technology Dawson and Meltzoff are using will also enable them to create brain maps that show where the electrical activity is occurring. By presenting children in the study with tasks known to tap into certain areas of the brain and mapping where electrical activity occurs, the researchers will test hypotheses about the neurobiological basis of autism.

Researchers in the longitudinal study will map electrical activity in the brain using ERP techniques. Data is gathered by a net of sponge-tip electrodes that a child wears while he or she performs language and neuropsychological tasks.

"One of the areas of the brain we strongly suspect is impaired in autism is the medial temporal lobe, particularly the hippocampus and amygdala," says Dawson. "This area is specialized for recognition of the emotional significance of an event. For example, with a typically developing baby, if the mother is smiling at him or her, this part of the brain is perceptually recognizing a smile and recognizing it as being important­as emotionally significant. There may be certain stimuli that are particularly emotionally significant to the baby, like the mother's face. So it may be that if this part of the brain is impaired early on, the baby doesn't necessarily place any more importance on a face than an object."

The battery of tests used with children in the study will shed light on brain function in the medial temporal lobe as well as other areas of the brain suspected to be involved in autism, including the cerebellum and the prefrontal cortex. By measuring performance on these tests first at ages 3 to 4, then again at ages 6 to 7, Dawson and Meltzoff will be able to construct a neuropsychological profile of each child in the study that can be used to systematically investigate potential subtypes of autism, and they will chart the now murky course of early social-skill development.

"One of the interesting aspects of this project is that it allows Geri and me to collaborate," Meltzoff points out. "She brings her expertise in developmental psychopathology. I've been constructing tests within the framework of developmental theory. By combining both perspectives, and especially by having an umbrella program project that brings in other specialties, we have an opportunity to make significant breakthroughs in understanding autism, and also to use this information to help us revise theories of typical development."

Top of Page | Next Story | CHDD Home