Postadoption parenting and socioemotional development in postinstitutionalized children

Children adopted from institutions (e.g., orphanages) overseas are at increased risk of disturbances in social relationships and social understanding. Not all postinstitutionalized children exhibit these problems, although factors like the severity of deprivation and duration of deprivation increase their risk. To date, few studies have examined whether postadoption parenting might moderate the impact of early adverse care. Three groups were studied: postinstitutionalized and foster care children both adopted internationally and nonadopted children reared in their families of origin. The Emotional Availability (EA) Scales were assessed at 18 months in parent-child dyads. Parent emotional availability was found to predict two aspects of social functioning shown in previous studies to be impaired in postinstitutionalized children. Specifically, EA positively correlated with emotion understanding at 36 months; in interaction with initiation of joint attention at 18 months and group, it predicted indiscriminate friendliness as scored from a parent attachment interview at 30 months. Among the postinstitutionalized children but not among the children in other groups, higher EA scores reduced the negative association between initiation of joint attention and indiscriminate friendliness, thus suggesting that parenting quality may moderate the effects of early institutional deprivation.     

On the convergence of EM-like algorithms for image segmentation using Markov random fields

Inference of Markov random field images segmentation models is usually performed using iterative methods which adapt the well-known expectation–maximization (EM) algorithm for independent mixture models. However, some of these adaptations are ad hoc and may turn out numerically unstable. In this paper, we review three EM-like variants for Markov random field segmentation and compare their convergence properties both at the theoretical and practical levels. We specifically advocate a numerical scheme involving asynchronous voxel updating, for which general convergence results can be established. Our experiments on brain tissue classification in magnetic resonance images provide evidence that this algorithm may achieve significantly faster convergence than its competitors while yielding at least as good segmentation results.