Most children with pervasive developmental disorders (like autism) already present their symptoms and signs in the first year of life. Despite these early signs, few children are currently diagnosed before the age of three years. In the SOSO project, children suspected of having psychiatric disorders are observed in their behavior at the age of 16 months and followed until the age of 42 months.

According to the literature, playing is one of the behaviors that can differentiate between children with and without autism. Children with autism tend to play in a more stereotyped way, and the level of their play behavior tends to be lower compared to other children. During this project, we are able to observe these behaviors at a very young age, so we can investigate whether these differences are also visible at this young age, or whether they appear after receiving a specific level of play behavior. Therefore, we observe children who are suspected of having psychiatric disorders via two settings, in which we provoke play behavior. To analyse differences in play behavior we have developed ethograms, which we analyse using The Observer Video-Pro 4.0. Different observers use these ethograms to analyse child behaviors once their reliability exceeds 80%. The elements of these ethograms will be discussed during the presentation.

Another factor that might influence play behavior is the quality of attachment. The parents of a securely attached child can provide comfort and serve as a safe base for exploration and play; so, when observing child behavior, it might be important to consider their attachment. Trained observers judge the form of attachment of the children to their primary caretaker. Disorganisation of attachment is considered to be a major risk factor in the development of child psychopathology. It is not thought that disorganised attachment originates from physical problems of the child. Neurological abnormalities, however, may lead to pseudo-disorganised behavior, e.g. in children with autism. This may influence the behavior of the child. Using regression analyses, we try to get more insight into which factors influence play behavior in children suspected of being autistic.

To analyse whether differences in attachment and play behavior are a matter of different coping strategies, we also analyse cortisol levels (from saliva samples) and attachment classifications of these children. Many children have already participated in this project, and data will be presented for over 60 subjects.

Paper presented at Measuring Behavior 2002 , 4^th International Conference on Methods and Techniques in Behavioral Research, 27-30 August 2002, Amsterdam, The Netherlands

© 2002 Noldus Information Technology bv