To the Editor: Dr. Klin et al. (1) presented evidence for abnormalities in the visual behavior of autistic subjects in filmed social interactions. They reported that autistic subjects looked more at objects, and when looking at faces, they fixated mostly on mouths instead of eyes. We studied the fixation patterns of autistic children shown faces and objects as well, but in static conditions, and did not find any abnormalities in fixation parameters (2, 3). Dr. Klin et al. suggested that their results were the reflection of a core social deficit. However, we feel that a different interpretation is warranted, given the fact that visual abnormalities were not seen with the static stimuli used in our studies.

Recent models of facial processing have favored a distinction between the perception of static aspects of faces and the perception of changes due to facial movements (4). Structural facial processing is thought to occur mainly in the lateral fusiform gyrus, while dynamic information is processed in the superior temporal sulcus. The lateral fusiform gyrus is located in the ventral object-processing system. There are indications that the superior temporal sulcus receives input from both the ventral and dorsal streams, which carry dynamic information. Therefore, the abnormal fixation patterns of autistic subjects in response to dynamic but not to static faces might reflect abnormal processing in the dorsal stream. There are several indications from studies of motion perception that dorsal-stream functioning is indeed abnormal in subjects with autism.

Because of a problem in dorsal-stream functioning, subjects with autism might be inclined to avoid dynamic stimuli, such as moving faces. The greater focus on the mouth if such subjects look at faces is surprising. However, Dr. Klin et al. reported a positive relationship between social competence and viewing time when subjects focused on mouths. Their suggestion "that by focusing on mouths these individuals with autism might attain some understanding of social situations (perhaps because of greater, focused attention on speech), whereas attention to eyes may not lead to any additional social insights" (p. 906) is interesting. The tendency to focus on mouths might reflect an effort of autistic subjects to overcome a basic perceptual problem. One of the hallmarks of the study by Dr. Klin et al. is a better definition of the phenotype of autism. We want to stress the importance of studying basic perceptual mechanisms in order to improve the chance of finding phenotypic markers.