Development and Psychopathology, 14 (2002), 613–634 Copyright 2002 Cambridge University Press Printed in the United States of America DOI: 10.1017.S0954579402003115 The neurobiological basis of autism from a developmental perspective NATACHA AKSHOOMOFF, KAREN PIERCE, AND ERIC COURCHESNE Children’s Hospital Research Center and University of California Abstract Autism is a neurobiological disorder that is diagnosed through careful behavioral assessment in early childhood. In this paper, we review recent studies that have attempted to reveal the underlying causes of autism using a variety of techniques. Particular emphasis is placed on techniques that have been used by a number of different laboratories, including structural magnetic resonance imaging and postmortem studies of neuroanatomy. Neurobiological and neuropsychological data from individuals across a wide age range are examined from a neurodevelopmental perspective. We discuss how these recent advances have led us to develop a growth dysregulation hypothesis of autism. Finally, we discuss how this hypothesis may lead to new innovations in autism research. The goal of research on the neurobiology of Even as risk factors (genetic and/or terato- genic) for autism are identified, we may still autism, particularly in our own laboratory, is to determine not only the underlying causes lack an explanation of how they are responsi- ble for specific neurodevelopmental abnor- of the disorder but also its developmental tra- jectory. In autism, as in any neurodevelopmen- malities and subsequently how these abnor- malities lead to the behavioral syndrome. To tal disorder, this is by no means a straightfor- ward problem. Autism is generally considered understand the neurobiological bases of au- tism, therefore, requires not only the discov- a polygenic or multifactorial disease; its ge- netic predisposition may be caused by the im- ery of “initial” triggering conditions (e.g., ge- netic mutations) but also the revelation of the pact of many genes, each contributing a small effect on phenotype, or by the interplay of multiple aberrant factors and processes that constitute the deviant epigenesis. To under- multiple environmental factors with multiple genes. Except for rare individual cases, which stand the causes of autism one must necessar- ily examine a variety of factors across multi- may or may not be similar in cause or syn- drome to idiopathic autism, the causes leading ple biological levels. In this paper, we discuss the neurobiological to this disorder are, as yet, unknown. bases of autism from a neurodevelopmental perspective. First, we briefly review neuroana- The authors were supported by funds from the National tomical findings from older children, adoles- Institute of Neurological Disorders and Stroke (R01 NS- cents, and adults with autism. Second, we dis- 19855), the National Institute of Mental Health (R01 cuss recent magnetic resonance imaging (MRI) MH-36840 and K01 MH01814), and Children’s Hospital data from toddlers and preschoolers with au- Research Center. The work in our laboratory is made pos- sible by support from these funds, as well as the families tism. These data suggest that abnormal neural who participate in our studies. growth patterns early in development lead to Address correspondence and reprint requests to: Dr. an array of consequences in anatomical and Natacha Akshoomoff, Laboratory for Research on the functional organization. Third, we then dis- Neuroscience of Autism, Children’s Hospital Research cuss how this “growth dysregulation hypothe- Center, 8110 La Jolla Shores Drive, La Jolla, CA 92037; E-mail: natacha@ucsd.edu. sis” can be used to help interpret recent func- 613