Factors associated with the presence or absence of imaginary companions in 222 preschool children were investigated using a self-administered questionnaire completed by their parents. Section I of the Imaginary Companion Questionnaire was designed to elicit a variety of demographic data on the children and their families and was completed by all parents. Section II was devised to obtain data concerning the imaginary companion itself and was completed only by parents of those children (N -63) who currently or in the recent past had imaginary companions. Data on family structure, play activities, and personality characteristics of the children, as well as characteristics of their imaginary companions, were presented. Data from the present study indicate that reducing loneliness is one of the multiple functions served by imaginary companions.

The Splitting Index (SI), a self-report scale based on the writings of Kernberg (e.g., 1976) on self and object representations and the defense mechanism of splitting, was constructed. After development over the course of 6 pilot studies, the SI was validated through 2 further studies. Factor analyses revealed a 24-item scale with three 8-item subscales, measuring the splitting of self, family, and others' images. The SI and its subscales were demonstrated to be internally consistent and stable over a 4-week period. Convergent validity was supported by significant correlations with measures of borderline and narcissistic personality disorders, self-image stability, self-esteem, depression, and negative affectivity. Discriminant validity was demonstrated by near-zero correlations with two measures of cognitive complexity. Contrary to predictions, the SI was significantly correlated with the Dogmatism Scale (Rokeach, 1960), a third measure of cognitive complexity. Research and clinical applications of the SI are discussed.

The developmental progression of children's belief in three major figures oj early childhood was examined through structured interviews with children and questionnaires for parents. Belief in Santa Claus, the Easter Bunny, and the Tooth Fairy varied with the child's age and the level of parental encouragement of belief. However, belief in these figures was unrelated to other indices of the child's fantasy involvement.asual observation indicates that vari-C ous imaginary figures-Santa Claus, the Easter Bunny, the Tooth Fairy, and others-play significant roles in the lives of many young children. Yet pertinent professional writings are surprisingly sparse. A review of the literature revealed some scattered references to Santa Claus. The history of Santa as an American folk custom was examined by Barnett,' and a small body of psychiatric papers described negative reactions in adults occasioned by Chri~tmas.~. * Ge sell and Doll pioneered in outlining the age sequence of children's belief in Santa Claus some years ago. While authorities in child development continue to offer conflicting advice to parents on the desirability of the Santa myth,2s child development research on Santa Claus appears limited to the study of expectancy effects6 Some recent popular literature has focused on issues in the selection of Santa's helpers for department stores,20

The interrelations of a variety of indices of sociocognitive development (empathy, role-taking, logical cognition, and moral reasoning) were studied in delinquent and nondelinquent adolescents. Delinquent males (grouped into psychopathic, neurotic, and subcultural subgroups) and a matched nondelinquent comparison group were administered individually two empathy scales (the Davis Interpersonal Reactivity Index and the Mehrabian and Epstein Questionnaire Measure of Emotional Empathy), an adaptation of Byrne's and Flavell's role-taking tasks, two Piagetian cognitive tasks, and two of Kohlberg's structured moral dilemmas. Delinquents as a group displayed significantly more immature modes of role-taking, logical cognition, and moral reasoning than did nondelinquents. The delinquent subgroups, however, did not significantly differ from one another on these dimensions. Role-taking, logical cognition, and moral reasoning were significantly related to one another. Anticipated differences in level of empathy between the delinquent and nondelinquent adolescents failed to occur. A number of interpretations for this unexpected finding are offered.

Using Quay's typology, three equal groups (« = 12) of adolescent psychopathic, neurotic, and subcultural delinquent males and a matched nondeliquent control group were individually administered Kohlberg's structured moral dilemmas, two Piagetian tasks of cognitive development (pendulum and balance), and an adaptation of Flavell's role-taking task. Psychopathic delinquents were more immature in level of moral development than all other groups, which did not differ from one another. Psychopathic delinquents were significantly more concrete in their thinking on cognitive tasks than all other groups, which exhibited signs of early formal operational thinking. Psychopathic and, to a lesser extent, neurotic delinquents were deficient in role taking compared with controls and subculturals, who did not differ from one another. The findings provide further evidence against viewing delinquency as a unitary syndrome of deviance.