A study conducted in two types of laboratory settings and one school setting tested a model of the development of emotional expression. The model included: (1) societal, peer, and family influences; (2) self-factors, such as physiological, affective, motivation, cognitive, and personality variables; and (3) self-mediators, which act as filters through which other variables are accepted or rejected. All these factors are mutually influential and interactive. In the study, heart rate, gender, maternal expressiveness, and maternal anger were investigated as predictors of children's expressiveness. A total of 60 mothers and their children were chosen from a sample of 143 mothers who filled out the Family Expressiveness Questionnaire (FEQ). The 30 mothers scoring highest on both the positive and negative expressiveness FEQ subscales, and the 30 mothers scoring lowest on both subscales, were selected. Study methodology involved tests of these mothers' children's heart rates; the administration of a questionnaire to determine causes, speed, and duration of anger among mothers; and laboratory observations and teacher ratings of the children's emotional expressiveness. The study found that the children whose mothers reported both enduring anger and high expressiveness had the most depressed expressiveness levels. The combination of maternal negative feeling and negative expression appeared to have a greater effect than either factor by itself. Children whose mothers had low maternal anger ratings had the highest scores on positive and negative expression in story-telling situations; on positive expression in event situations; and on teacher ratings of children's happy and angry expressions. (AC)