The predictive relation between neonatal behavior and temperament at 24 months was examined for 67 infants. Behaviors rated by trained examiners at both ages were aggregated into composite scores to provide a more reliable indicator of individual differences than single-item scores. Longitudinal analyses revealed several significant relations between the two ages, most notably that irritable neonates were rated as more upset, less attentive to stimuli, and less responsive to the staff at 24 months. The common variable describing behavior at both ages was related to emotionality. Examination of extreme groups on ratings of emotional tone indicated that the distressed toddler had high irritability ratings as a neonate, but the more pleasant toddler was not different from the average group in neonatal irritability scores. It was concluded that a modest predictive relation between temperament variables was demonstrated from the neonatal period to 24 months.In recent years there has been increasing interest in the study of infant temperament. One focus has been on problems of continuity of temperament from one age to the next and the concomitant problems of measurement (Buss & Plomin, 1984;Goldsmith & Campos, 1982;Hubert, Wachs, Peters-Martin, & Gandour, 1982;Rothbart & Derryberry, 1981). Several researchers propose a constitutional basis for temperament and predict some stability in the expression of temperament over time (Buss & Plomin, 1984;Rothbart & Derryberry, 1981). Rothbart & Derryberry (1981) maintain that temperament is a psychobiological concept involving a relation between the nervous system and behavior. Although the constitutional aspects may be influenced by maturation and experience, some degree of continuity is predicted. Assessment of the infant during the neonatal period allows for an evaluation of behavior before it is influenced by familial environmental and socializing factors, and it enables a search for the salient features of behavior that may be predictive of later temperament.Most attempts at establishing relations between newborn behavior and later temperament have met with only marginal success (see reviews by Goldsmith & Campos, 1982;Hubert et al., 1982; also Sameroff, 1978). In their research, Birns, Bar ten, and Bridger (1969) have demonstrated some behavioral stability in infancy: Irritability and sensitivity measured at 2 or 3 days of age in neonates still in the hospital were related to irritability This research was supported in part by a grant from the John D. and Catherine T. MacArthur Foundation, and Grant 14352 from the National Institute of Child Health and Human Development (R. S. Wilson, principal investigator).I am grateful to the parents and their infants who participated in the study, to the administrators and nursing staffs in the participating hospitals, and to the many co-workers who have contributed so much to the program, including R.

Temperament was assessed for newborn twins, using a comprehensive neonatal exam that focused on irritability, resistance to soothing, activity, and reinforcement value. The same infants were later assessed at 9 months in a structured laboratory setting, where measures of emotional tone were obtained under a wide variety of instigating conditions. Summary ratings of emotional activity were compiled for each infant, representing the preponderant reaction of the infant in both settings. Individual differences were markedly evident, ranging from wailing distress to smiling and contentment. The neonatal variables correlated significantly with emotional tone at 9 months (rs in mid-0.20 s, p < .05), and a further analysis revealed a generalized multiple correlation of R = 0.32 between the neonatal variables and the 9-month measures of temperament. When extreme groups on emotional tone were selected at 9 months (crying and distress vs. smiling and contentment), the groups were significantly discriminated by their neonatal scores, with 70% of the infants being correctly assigned to the appropriate extreme group. Overall, the irritable, difficult-to-soothe neonate was likely to be fussier and more distressed in the lab assessment than the more tractable neonate. The results affirmed a significant predictive linkage between neonatal behaviors and later measures of temperament and thus gave credence to the premise of some stability in infant temperament. Emotional activity appeared to be the core dimension stretching over ages, and it is discussed in relation to other formulations of infant temperament. The results demonstrated the utility of a comprehensive neonatal assessment specifically designed to measure temperament, which could be coordinated with laboratory measures of temperament at later ages.Research in infant temperament has become particularly active in the past few years, with a notable increase in observational studies to supplement the earlier questionnaire studies. Two recent reviews have drawn attention to some of the limitations in questionnaire studies and urged that controlled laboratory studies be performed (Goldsmith &

No abstract

No abstract

Cardiac and behavioral responses to a tactual stimulus were evaluated during the first sleep cycle for 3 groups of infants: 30 full terms, 30 nonintervened preterms, and 30 intervened preterms. Prior to testing, the latter group had received a regimen of multimodal sensory stimulation, which emphasized the tactual and vestibular modalities. The results showed that the intervention altered the preterms' sensory functioning mainly during active sleep. In this state, the full terms and the intervened preterms exhibited a significant cardiac acceleration to the stimulus, while the nonintervened preterms failed to do so. Similarly, the behavioral response of the intervened preterm more closely approximated that of the full terms. During quiet sleep the intervention did not affect either the cardiac or the behavioral response. In this sleep state, the full terms' cardiac response was that of monophasic acceleration, whereas both groups of preterms exhibited a biphasic response with a smaller initial acceleration which was followed by a deceleration below baseline. Behaviorally, the preterms responded less frequently and gave smaller responses than the full terms. Thus there were marked differences between preterms and full terms, and the intervention seemed to narrow this gap to some extent.