See related article on page 593 In recounting research results in the field of autism over the past 50 years, the most consistent and well-replicated finding remains that of elevated whole blood serotonin. Aberrant serotonin function has consistently been found to contribute to the genetics, neuropharmacology, and brain metabolism of persons with autism.1 In addition, some but not all randomized controlled trials have shown selective serotonin reuptake inhibitors (SSRIs) to improve specific associated symptoms (e.g. interfering repetitive behavior) in children and adults with autism.2 The data presented in this issue’s article by Makkonen and colleagues build upon this series of findings related to serotonin dysfunction in autism. Single-photon emission computed tomography (SPECT) with [123I] nor-β-CIT (a labeled tracer that has been used to determine serotonin [SERT] and dopamine [DAT] transporter binding capacity) was utilized to image 15 children with autism and 10 non-autistic comparison children. Due to concerns about unnecessary radiation exposure in healthy children, the comparison children had a number of different neurological diagnoses. Reduced SERT binding capacity was found in the medial frontal cortex (MFC), midbrain, and temporal lobe areas of the children with autism compared with the comparison children. After adjustment for age and correction for the assumed effects of sedation, only the reduction in the MFC remained significant. No difference was identified in DAT binding capacity between the groups. As the authors acknowledge, there are a number of limitations to their study. These include the lack of a healthy age- and sex-matched comparison group, for reasons addressed above; the need for sedation in the children with autism; the lack of use of a criterion standard autism diagnostic instrument, such as the Autism Diagnostic Interview-Revised; the relatively small sample size; and the unusually high level of cognitive functioning (mean Full-scale IQ=89) of the children with autism. On the other hand, there are a number of strengths to this investigation. Published brain imaging studies of neurochemical binding are extremely rare in pediatric populations, including children with autism, due to appropriate concerns regarding radiation exposure. For this reason, these preliminary data are unique, valuable, and a potentially important contribution to the literature. In fact, this study is believed to be the first investigation of SERT and DAT binding capacity in children and adolescents with autism. The authors speculate that the reduced SERT binding observed in the children with autism may represent less synapse sprouting and a sparse synaptic density in serotonergic pathways in autistic individuals. They conclude that the most evident reason for the reduced SERT binding capacity is a lesser number of serotonergic synapses and SERTs themselves. Autism is a ‘syndrome’ and its causes will ultimately be linked to a multitude of biological factors related to both nature and nurture. As the study by Makkonen et al. demonstrates, neuroimaging techniques, including SPECT, have the potential to contribute substantially to furthering our understanding of the pathophysiology of autism. It is unlikely that a single neurochemical, such as serotonin, will be determined to be central to the cause of autism. Undoubtedly, excitatory and inhibitory amino acids, neuropeptides such as oxytocin and vasopressin, and neurotrophic and other factors essential to brain development may be involved.3 It is interesting, though, that serotonin is believed to play a critical interactive and modulatory role with each of these systems. The involvement of serotonin in the pathophysiology and, potentially, the treatment of autism has withstood the test of time. An important next step to build upon the work of Makkonen and colleagues would be to consider a positron emission tomography investigation of SERT and/or serotonin receptor binding capacity in adults with autism. Studying an adult population would minimize the risks of radiation exposure and allow for the inclusion of a healthy control group. Adults with autism might also be better able to participate in the scans without the need for sedation. For such a study to proceed it would be imperative to ensure that the participants and/or their legal guardians were engaged fully in the informed written assent and consent process.