Early indicators of schizophrenic outcomes were sought in a group of children of chronically ill schizophrenic women. A sample of pregnant women with varying degrees of mental illness were examined during the perinatal period and recruited into a 4-year longitudinal evaluation, which included cognitive, psychomotor, social, and emotional assessments at birth, 4, 12, 30, and 48 months of age. The mothers varied on mental health dimensions of diagnosis, severity of symptomatology, and chronicity of illness. Other factors included in the analyses were socioeconomic status (SES), race, sex of child, and family size. Hypotheses were tested to determine the relative impact of three sets of variables on the child's behavior: (1) specific maternal psychiatric diagnosis, (2) severity and chronicity of disturbance independent of diagnosis, and (3) general social status. We found that a specific maternal diagnosis of schizophrenia had the least impact. Neurotic-depressive mothers produced worse development in their children than schizophrenic or personality-disordered mothers. Both social status and severity/chronicity of illness showed a greater impact on development. Children of more severely or chronically ill mothers and lower-SES black children performed most poorly. These results do not support etiological models based on simple biological or environmental transmission of schizophrenia. The role of social and family environmental factors in predicting child cognitive and social-emotional competence was further evaluated using a multiple risk index. Children with high multiple environmental risk scores had much worse outcomes than children with low multiple risk scores.

Verbal IQ scores in a socially heterogeneous sample of 215 4-year-old children were highly related to a cumulative environmental risk index composed of maternal, family and cultural variables. Different combinations of equal numbers of risk factors produced similar effects on IQ, providing evidence (1) that no single factor identified here uniquely enhances or limits early intellectual achievement and (2) that cumulative effects from multiple risk factors increase the probability that development will be compromised. The multiple risk index predicted substantially more variance in the outcome measure than did any single risk factor alone, including socioeconomic status. High-risk children were more than 24 times as likely to have IQs below 85 than low-risk children.

It was hypothesized that maternal teaching style (MTS) and a child's attentional performance mediate the relation between contextual risk factors and intelligence quotient (IQ). One hundred fifty-nine 4-year-old children and their mothers in a longitudinal study of children at risk for mental disorder participated in a maternal teaching task, a Luria bulb-squeeze procedure, and a delayedmatch-to-sample (DMS) task in the laboratory. Maternal teaching, Luria, and DMS performance added significant variance to the risk-IQ prediction equation. The influence of MTS and attentional performance was examined through regression and structural equation-model analyses. Contextual risk was strongly related to a child's preschool IQ, but this connection may be mediated by maternal interaction and child attentional performance. The role that social context plays in defining the limits of the child's cognitive growth is discussed.