Anger Regulation In Disadvantaged Preschool Boys: Strategies, Antecedents, and the Development of Self-Control

 

The 20^th century witnessed a steady accumulation of data indicating that boys from low-income families are at elevated risk for problems of self-control (Bolger, Patterson, Thompson, & Kupersmidt, 1995; Campbell, Shaw, & Gilliom, 2000; Elder, 1979; C. Shaw & McKay, 1942). On average, socioeconomically disadvantaged boys, particularly those living in urban neighborhoods, are more disruptive and aggressive than their peers (Loeber & Farrington, 2000).  These differences are evident by school entry and become increasingly clear thereafter (Keenan & Shaw, 1997; Stevenson, Richman, & Graham, 1985).

While information on risk status is useful for prevention purposes, a focus on poor outcomes among disadvantaged boys has limited research in several ways. First, there has been inadequate recognition of within-group differences, even though most boys from low-income families develop a broad range of skills and do not become antisocial (Campbell et al., 2000). Second, because of the emphasis on outcome, relatively little is known about processes that account for the differences in adjustment within this at-risk population.

This study examined anger regulation, its developmental antecedents, and its relations with self-control in young boys from low-income families. A focus on how children regulate their emotions has been used successfully to examine normative development in children from middle class backgrounds but has been applied less often to high-risk samples (Cole, Michel, & Teti, 1994). Following the lead of other emotion regulation researchers, we adopt a functionalist view of emotions (Campos, Campos, & Barrett, 1989). Within this perspective, emotions are defined as “processes of establishing, maintaining, or disrupting relations between the individual and the internal or external environment on matters of significance to the individual” (Campos et al., 1989, p. 395).  From a functionalist viewpoint, anger can serve adaptive purposes by motivating efforts to eliminate barriers to a desired outcome (Saarni, Mumme, & Campos, 1998). However, consistent with the notion that emotions can disrupt, as well as enhance, skillful responding to environmental challenge, dysregulated anger may trigger aggressive, oppositional behavior and preempt prosocial problem-solving strategies (Berkowitz, 1962). To begin to characterize anger regulation and its development in low-income boys, we examined preschool regulatory strategies in relation to expressions of angry affect, child and maternal characteristics in toddlerhood, and indicators of self-control in first grade.

Anger Self-Regulation Strategies in Early Childhood

Although there is general agreement that the term emotion regulation refers to processes used to modulate the intensive and temporal features of emotional arousal (Gross, 1999; Thompson, 1994), the means by which modulation occurs remain, to a large degree, an open question. One approach to the study of emotion regulation has been to examine behavioral strategies hypothesized to manage emotion in relation to the dynamic features of affective expression. Much of the developmental work in this area has focused on infants and toddlers. Although strategies of regulation in the first two years of life may be relevant to older children, important developments during early childhood make this period worthy of attention in its own right. First, early childhood heralds an emergent sense of self-awareness (Kopp, 1989). The preschool child is cognizant of his or her status as object and agent and thus deliberately enlists emotion regulation strategies to maintain well-being and to uphold social standards of behavior. Second, the young child has a basic understanding of the relationship between distress and external causes (Kopp, 1989). An awareness of causes enables a range of new self-regulatory strategies, including focusing on less upsetting stimuli, learning more about the source of distress, and changing the offending cause. We consider in turn various strategies young children may use to regulate frustration.

 Redirection of attention can be more or less active, ranging from looking away (Fox, 1989) to engagement in alternative activities (Braungart & Stifter, 1991). Gaze aversion in infants predicts decreased distress following restraint and goal frustration (Buss & Goldsmith, 1998), while distraction with toys predicts decreased anger in response to frustration in toddlers (Calkins & Johnson, 1998; Grolnick, Bridges, & Connell, 1996). In a study that examined sex differences in strategy use among 2-year-olds, boys were more likely to distract themselves than girls (Raver, 1996). In this study, we assessed both active and passive reorienting strategies.

A second type of strategy involves maintaining or increasing attentional focus on the source of frustration. Research with toddlers suggests that this strategy increases child anger, particularly if the situational constraints are unalterable (Grolnick et al., 1996). However, the cognitive capabilities of the preschool child may support an additional, more adaptive type of ‘approach’ behavior, wherein the child attempts to learn more about the restrictions he or she faces. Although this type of strategy has not been examined in prior studies of emotion regulation in children, research on adult anger suggests that the acquisition of information regarding how and when sources of frustration will be removed reduces arousal (Dalgleish & Power, 1999). To examine the relation between information-gathering and anger in early childhood, this strategy was assessed apart from other ‘focus on source of frustration’ behavior.

Comfort-seeking behaviors comprise a third type of strategy that young children may employ to regulate frustration. Among 2-year-old children, this strategy is among those used most frequently in situations requiring the delay of gratification (Grolnick et al., 1996), although male toddlers appear to seek comfort somewhat less often than female toddlers (Raver, 1996). Because early childhood is a transitional period between the dependence on caregivers of toddlerhood and the relative autonomy of middle childhood, comfort-seeking behavior may figure less prominently in the regulatory repertoires of preschoolers than in those of 2-year-olds.

Developmental Antecedents of Anger Regulation in Early Childhood

The theoretical literature contains several accounts of how transactions among factors internal and external to the child shape the development of emotion regulation over the first years of life (e.g., Bridges & Grolnick, 1995; Calkins, 1994; Kopp, 1989). However, few researchers have examined longitudinal relations between such factors and subsequent strategies for coping with emotional challenge. Of interest here are factors in toddlerhood that may be related to the acquisition of strategies to regulate anger, including include child negative emotionality, attachment security, and parental control strategies.

Among internal factors, negative emotionality, or irritability, is thought to be particularly relevant to the way children learn to handle frustrating situations (Calkins, 1994). While moderate levels of distress are necessary for children to adopt and practice emotion regulation skills (Kopp, 1989), extreme irritability may impede the development of adaptive regulatory behaviors (Calkins, 1994). Irritable children may become too disorganized when frustrated to self-regulate effectively, or caregivers may ‘take over’ the regulatory process to prevent the escalation of negative affect. Consistent with these hypotheses, Braungart and Stifter (1996) found an inverse relationship between 5-month negative reactivity to frustration and 10-month regulation. An important but untested proposition is that the impact of child emotionality on early emotion regulation is moderated by aspects of the caregiving environment (Calkins, 1994). In this case, the deleterious effects of high irritability may be exacerbated by harsh, restrictive parenting or attenuated by supportive parenting.

Parent-child relationship factors figure prominently in theories of the development of emotion regulation (Calkins, 1994; Kopp, 1989). In the two first years, social interactions are primarily with the parents, with mothers handling the bulk of early childcare duties in most families. During this period, children’s increasingly insistent bids for autonomy combined with a limited understanding of the social and physical environment produce normative increases in anger expression (Goodenough, 1931).The quality of the primary attachment relationship and the manner in which parents, typically mothers, seek to control child behavior may have long-term significance for the ways children deal with this anger.

According to attachment theory, the quality, timing, and pacing of parent-child interactions over the first year shape enduring internal representations regarding the emotional demands of the environment and the availability of emotional support (Bowlby, 1969). Through a history of well-coordinated, positive encounters with primary caregivers, securely attached children are confident in their ability to cope with most situations and in the ability of caretakers to reinstate calm should the environment become over-arousing (Carlson & Sroufe, 1995). Later, these children are expected to remain organized in the face of stress and to seek the help of others when their own resources are overwhelmed. In contrast, insecurely attached children have experienced intermittent or insensitive caregiver responses when distressed and thus mistrust their own regulatory capacities and the regulatory support of others. These children are expected to respond inflexibly to future emotional challenges and to become dysregulated. In support of these claims, infant attachment security has been found to predict problem-solving ability (Frankel & Bates, 1990) and frustration proneness (Matas, Arend, & Sroufe, 1978) in early childhood.

In the second year, increases in mobility and in potential for harm to self and others require parents to exert increasing control over their children (Shaw & Bell, 1993). Concurrently, rapid developments in language skills render toddlers particularly vulnerable to repressive, hostile injunctions (Kopp, 1991). Kopp (1989) has speculated that parents who use warmth and verbal guidance when socialization goals run contrary to the personal goals of the child facilitate the development of flexible self-regulation. Accordingly, in a cross-sectional study of 2-year-olds, Calkins and Johnson (1998) found that maternal positive control was related to children’s use of distraction and constructive coping with frustration.

Although cultural factors have received little attention in the emotion regulation literature, there is evidence that parenting practices differ across ethnic groups, which in turn may lead to systematic differences in patterns of child self-regulation. For example, African American mothers are more likely than European American mothers to use an authoritarian style that emphasizes control over warmth (Baumrind, 1972). The differential impact in parenting styles on anger regulation in African American versus European American boys is hard to predict. Heightened expectations for control may lead to more autonomous self-regulation among African American boys. Alternatively, European American boys may benefit from increased warmth in the parent-child relationship. Thus, while the field lacks theoretical or empirical grounds on which to make specific predictions regarding cultural differences in the development of anger regulation, the pervasive influence of culture across the lifespan (Garcia Coll, Akerman, & Cicchetti, 2000) necessitates its consideration.

Anger Regulation and Self-Control at School Entry

School entrance introduces new standards and contexts for appropriate conduct. Children are now expected to suppress aggressive and destructive impulses, to cooperate with peers and adults, and to assert their own needs without violating the rights of others. These skills–the inhibition of disruptive behavior, cooperation with others, and appropriate self-assertion–have been identified as important elements of self-control in school age children (Bohart & Stipek, 2001; Caldarella & Merrell, 1997; Patterson, 1982).        Several theorists have suggested that patterns of emotion regulation, particularly the regulation of anger, play an important role in the development of self-control (Calkins, 1994; Cole & Zahn-Waxler, 1992; Eisenberg & Fabes, 1992). Data relevant to this hypothesis emerge from two research traditions. First, investigators have studied longitudinal relations between negative emotionality in infancy and adjustment in childhood (Rothbart & Bates, 1998). For example, Rothbart and colleagues (Rothbart, Ahadi, & Hershey, 1994) found that infant irritability predicted aggression at age 7. However, while studies of emotionality suggest that young children who have difficulty managing negative affect are at risk for maladjustment, they do not address the processes by which emotion becomes dysregulated or how regulatory skills might shape the acquisition of behavioral controls.

Work on individual differences in emotion regulation strategies is germane to these important questions. In an early paper on this topic, Eisenberg and Fabes (1992) proposed that children who are able to modulate intense negative emotions through the deployment of effective emotion regulation strategies will be less disruptive and more prosocial when stressed than children with impoverished regulatory capacities. In their research, Eisenberg and colleagues have focused primarily on attentional focusing/shifting as an index of emotion regulation. As hypothesized, they found that children low on attentional focusing/shifting were less constructive when angered than other children and were more likely to exhibit externalizing problems (Eisenberg, Fabes, Nyman, Bernzweig, & Pinuelas, 1994; Eisenberg et al., 1996). However, consideration of other regulatory strategies and other aspects of self-control may reveal a more complex story regarding the links between emotion regulation and social competence.

First, there may exist specific relations between various regulatory strategies and aspects of self-control. That is, an emotion regulation strategy that helps solve one type of social problem may be less useful in solving other problems, even if it is effective in reducing or inhibiting negative arousal (Thompson & Calkins, 1996). For example, shifting attention away from sources of frustration predicts lower levels of noncompliance and aggression (Eisenberg et al., 1996) but may be less beneficial in situations that require self-assertion (Crockenberg & Litman, 1990). Thus, children who ignore the provocative behavior of a peer may be less likely to resolve the situation in a way that is advantageous to them than those who seek information about the motives of the other child. Relatedly, children who can select from a number of effective emotion regulatory strategies according to situational demands may be best-prepared for the various social challenges of middle childhood. Access to a variety of prosocial solutions to peer conflict predicts lower rates of aggression in middle childhood (Lochman & Lampron, 1986). Similarly, versatility in the use of emotion regulation strategies may be an important indicator of regulatory skill.

Goals

This study examined patterns and correlates of anger regulation in young boys from low- income families. Our goals were threefold. First, we sought to explore relations between regulatory strategies and expressed anger in early childhood. We hypothesized that behaviors that shift attention away from sources of frustration or that clarify situational constraints would predict decreases in expressed anger, while those that increase focus on frustrating stimuli would predict increases in expressed anger. The second aim was to examine developmental antecedents of anger self-regulation. We expected that low negative emotionality, secure attachment, and positive maternal control in toddlerhood would predict higher levels of effective anger regulation strategy use, and that the relationship between child emotionality and strategy use would be moderated by maternal control. Based on prior work indicating parenting differences in African American versus European American mothers (Baumrind, 1972), we also conducted exploratory analyses to test for group differences in levels of anger regulation and to determine whether predictive relations between maternal control and strategy use are qualified by race.

The third goal of this research was to explore links between anger regulation in early childhood and self-control at school entry. We hypothesized that reliance on effective regulatory strategies would predict greater self-control, defined here in terms of lower levels of externalizing problems, higher levels of cooperation, and higher levels of appropriate assertiveness. Based on conceptualizations of emotion self-regulation as an important achievement of early childhood (Calkins, 1994; Kopp, 1989), we expected that these relations would hold after accounting for the effects of negative emotionality, attachment security, and maternal behavior. We also anticipated specific relations among particular regulatory strategies and individual aspects of self-control. Use of attention-shifting strategies was expected to predict lower levels of externalizing problems and higher levels of cooperation, while information gathering was expected to be the strongest predictor of high assertiveness. Additionally, we hypothesized that boys who employed a variety of effective regulatory strategies would receive higher scores on measures of self-control. To achieve these aims, we assessed child and maternal factors at 1½ years, behavioral strategies and angry affect during a delay of gratification task at age 3½, and school self-control at age 6.

 

Method

Participants

This study used data from a larger project on antisocial development in young, low-SES boys. Boys were recruited instead of girls based on prior research demonstrating higher levels of conduct problems in boys beginning in the preschool period (Achenbach, Edelbrock, & Howell, 1987; Richman, Stevenson, & Graham, 1982; Rose, Rose, & Feldman, 1989). Low-income participants were selected for the same reason (Loeber & Farrington, 2000). Recruiting took place at Women, Infants, and Children (WIC) Nutritional Supplement Program clinics throughout the metropolitan Pittsburgh area. Mothers with male infants between 6 and 17 months of age were asked to take part in a longitudinal study on child development. Of 421 mothers approached at the WIC sites, 310 participated in the first assessment at age 1½ while 282 participated in a second assessment at age 3½. Teacher ratings were available on 189 participants at age 6. The sample was comprised primarily of European American and African American children (54% and 40%, respectively), with a small number of biracial and Hispanic American participants. At the age-1½  visit, mothers were between 17 and 43 years of age. Mean per capita income was $2,892 per year ($11,568 for a family of four). Sixty-two percent of mothers identified themselves as married or living with a partner, 28% stated they were always single (never married), and 8% were separated, divorced, or widowed. Comparisons of boys who participated in the age 3½ assessment with those who did not revealed no significant differences on demographic variables or on child and family characteristics assessed at age 1½. Similarly, boys with teacher data at age 6 did not differ significantly from boys without these data on child and family characteristics at age 1½ or on any of the age-3½ anger regulation variables.

Procedures

Mothers and their sons completed laboratory or home assessments when boys were 1½, 3½, 5½, and 6 years of age. Assessments lasted approximately 2 hours and included structured parent-child interactive tasks and free-play periods. Assessment tasks were selected to vary in stress level so that mother and child behavior could be observed across a broad spectrum of conditions. Laboratory assessments were video-recorded through a one-way mirror and via a second, ceiling-mounted, remote-operated camera. The observational measures used in this study were derived from portions of the age-1½ visit and of the “Cookie Task” (Marvin, 1977) at age 3½. The relevant age-1½ tasks included the Strange Situation (Ainsworth, Blehar, Waters, & Wall, 1969), a 3 min clean-up task (Martin, 1981), and a 3 min teaching task (Matas et al., 1978).       Mothers and sons engaged in the 3 min Cookie Task midway through the age-3½ visit. This task was designed to simulate situations in which children must tolerate waiting for a desired outcome when there is little of interest in the immediate environment (e.g., waiting for parents to finish preparing dinner, waiting for a parent or sibling to end a telephone conversation so that he or she can play). Kopp (1991) notes that waiting is “extremely difficult for children, irrespective of age” (p. 44). For this task, the assessment room was cleared of all toys and activities while mother and child looked on. The mother was asked to sit at a table and complete questionnaires. In addition, the mother was given a clear bag with a cookie inside it and told to hold it in view but out of reach of her child during the task. The examiner then left the room. After 3 minutes, the examiner signaled the mother to give the cookie to the child by knocking on the one-way mirror.

Mothers completed questionnaires on child, parent, and family characteristics at each visit. Boys’ teachers completed questionnaires on school adjustment as part of an age-6 assessment.

Measures

Measures were grouped into four categories: (1) early child and maternal factors, (2)

behavioral regulatory strategies and angry affect, (3) school self-control, and (4) control variables.

Early child and maternal characteristics.

Negative emotionality. Negative emotionality at age 1½ was assessed with the Difficultness factor of the Infant Characteristics Questionnaire (ICQ; Bates, Freeland, & Lounsbury, 1979).

The ICQ, a measure of temperamental characteristics by maternal report, has demonstrated adequate reliability and validity and has shown longitudinal relations with preschool behavior problems (Bates, Maslin, & Frankel, 1985). The Difficultness factor assesses the intensity and frequency of fussy, irritable child behavior.

Attachment security. Attachment security at age 1½ was assessed using the Strange Situation (SS, Ainsworth et al., 1969). Videotapes of the SS were coded into one of four categories (A, B, C, D) according to the procedures described by Ainsworth and colleagues (1969) and by Main and Solomon (1986). Three coders, blind to scores for maternal behavior, were trained to reliability and tested for interrater agreement using attachment assessments from the lab of A. Sroufe. Interrater agreement on the A, B, C, and D categories ranged from 80% to 100% with a mean of 86%. Lacking hypotheses about specific insecure attachment categories and the development of anger regulation, we recoded categories A, C, and D into a single insecure group for comparison with the secure group.

Maternal control. Maternal control was assessed at age 1½ using the Early Parenting Coding System (EPCS; Winslow, Shaw, Bruns, & Kiebler, 1995). The EPCS was designed to capture a range of parenting behaviors typically exhibited in interactions with young children. Molecular and global codes were made from videotapes of the clean-up task and a teaching task in which the mother spent 3 min working with her son to construct a puzzle. For the purposes of this study, only codes relevant to positive versus negative control were used. These included three molecular codes–counts of verbal approval, physical approval, and critical statements–and three global codes–hostility, warmth, and punitiveness. Hostility was defined as the expression of anger by the mother toward the child as indicated by tone of voice and mannerisms. The warmth code assessed positive affect directed toward the child. Punitiveness was defined as the extent to which the mother was too strict or harsh, considering the child’s behavior. Originally, global codes were made on 4-point scales; however, it was necessary to convert the punitiveness and warmth codes to 3-point scales due to difficulties among coders in making more narrowly defined distinctions. Kappa reliability ranged from .83 to .94 for the individual codes. Principal components analysis of the six codes yielded a single factor with an eigenvalue greater than one. The individual variables were converted to z-scores then summed to form maternal control composites for the clean-up and puzzle tasks. Because the composite scores for the two tasks were positively correlated with one another (r = .47), they were summed to form a single measure of maternal control. High scores on this measure indicated warm, accepting control; low scores indicated harsh, hostile control.

Behavioral strategies and angry affect.

Child Affect Coding System. Child affective displays during the Cookie Task at age 3½ were coded from videotape using a scheme developed by Cole, Zahn-Waxler, and Smith (1994). Founded on the work of Ekman and Friesen (1978), Izard (1979), and others, this system uses facial action and vocal quality cues to determine the presence of basic emotions. Facial cues for anger include tightening or narrowing of the eyelids, tightening or pressing of lips, and clenching of teeth. Vocal cues include harsh, insistent vocal tone and increased volume and pitch. These indicators were used to rate the peak intensity of angry affect in each of 18 ten-sec coding intervals of the Cookie Task (all mother-child dyads completed the entire duration of the Cookie Task). Intensity was rated on a 0 to 3 scale with 0 indicating no anger, 1 indicating mild frustration or annoyance, 2 indicating moderate to intense vocal and facial expressions of anger without crying or shouting, and 3 indicating angry crying or shouting. Affect coders were unaware of the regulation hypotheses of this study. Reliability calculations were based on agreement with a master coder (first author) on 30 tapes, with agreement defined as both coders observing the same peak intensity in a given interval. Using this criteria, agreement for tapes containing expressions of anger (21 of the 30 tapes) was 88%; kappa was .76.

      Child Emotion Regulation Coding System. This system is based on Grolnick and colleagues’ (1996) work on emotion regulation in very young children. Five child behaviors of purported functional significance for affective expression were scored for presence/absence in each of the 10-sec coding intervals. These behaviors are as follows: active distraction (purposeful behaviors in which the focus of attention is not on the delay object or the task; may or may not be socially appropriate; includes fantasy play, exploration of the room, turning lights on and off, talking with mother, singing, dancing, etc.), passive waiting (standing or sitting quietly, not looking at cookie), information gathering (questions aimed at learning more about the waiting situation; does not include questions or statements indicating the child wants to change the situation), physical comfort seeking (e.g., touching mother, reclining on mother’s lap, requesting to be held), and focus on delay object or waiting task (speaking about, looking at, or trying to retrieve the cookie; speaking about or trying to end the waiting period, e.g., stating “I don’t want to wait,” requesting to leave the room). The codes are exhaustive; that is, boys engaged in at least one of these behaviors during each 10-sec interval. In addition, codes are mutually exclusive, with one exception: If boys engaged in distraction, passive waiting, information gathering, or focus on task/object while in ongoing physical contact with their mother (e.g., leaning against mother, sitting on her lap), they received a code for comfort seeking and the other strategy. Otherwise, only one strategy was coded at a time. Initial instances of comfort seeking (e.g., requesting to be held, climbing onto mother’s lap) were considered mutually exclusive with other strategies. Post-training reliability was adequate for all codes. Percent agreement with a master coder was 89% - 96%; kappas ranged from .64 to .79. Coders were unaware of the study hypotheses. There was no membership overlap for the behavioral strategies and affect coding teams.

School self-control.

Teacher’s Report Form (TRF; Achenbach, 1991). The teacher version of the Child Behavior Checklist assesses childhood problem behavior in the school setting. The 34-item broad-band Externalizing factor was used in this study. This factor contains items tapping disruptive behaviors, including aggression (e.g., “Gets in many fights”), noncompliance (e.g., “Disobedient at school”), and delinquency (e.g., “Swearing or obscene language”). First-grade teachers completed the TRF when boys were 6 years of age.

Social Skills Rating Scale (SSRS; Gresham & Elliot, 1990) is a three-part instrument used to assess social skills, problem behaviors, and academic competence in school-age children. Teachers completed the Social Skills Scale when boys were age 6. This scale assesses children’s ability to conform to standards of socially appropriate behavior within a context populated by both peers and adults. Two factors from this scale were used:  Cooperation, which measures children’s ability to comply with rules (e.g., “Follows directions,” “Easily makes transitions”), and Assertiveness, which measures the extent to which children assert their needs without infringing on the rights of others (e.g., “Questions unfair rules,” “Tells adults when treated unfairly”). One month test-retest reliability coefficients for these subscales are .75 to .88, respectively. SSRS factor scores converge with other measures of social skills (Gresham & Elliot, 1990).    

Control variables.

Maternal Behavior Coding System. Mothers were not given specific instructions on how to interact with their sons during the Cookie Task, raising the possibility that ensuing differences in maternal behavior were an important source of variation in boys’ use of regulatory strategies.

Thus, it was deemed necessary to account for these differences in longitudinal analyses predicting to and from strategy use. This measure assessed the extent to which the mother provided supportive structure during the waiting period. Behaviors considered supportive of self-regulation include suggesting activities that may help the child wait and intervening appropriately if the child becomes distressed. Conversely, behaviors viewed as discouraging self-regulation include teasing the child or returning the child’s attention to the cookie as he regains control of his emotions. Because mothers were occupied with completing the forms during the Cookie Task and were only intermittently involved with their child, a global measurement approach was chosen over a more molecular, time- or event-based strategy. Maternal behavior was coded on a single, 5-point scale, with 1 indicating low encouragement and high discouragement of effective self-regulation, 3 indicating equal levels of encouragement and discouragement of effective self-regulation, 5 indicating high encouragement and low discouragement of effective self-regulation, and 2 and 4 indicating intermediate levels. Agreement with a master coder was 91%, kappa was .68. Maternal regulatory behavior was rated by the emotion regulation coders. To check for bias due to coder overlap, separate coders rated child and maternal regulation on 30 tapes that were also rated by the master coder. Agreement with the master coder was comparable whether child and maternal ratings were made by same coder or by separate coders (i.e., kappa did not differ by more than +.05 for any code).

Wechsler Preschool and Primary Scale of Intelligence – Revised (WPPSI-R; Wechsler, 1989). We hypothesized that boys who sought information about situational constraints at age 3½ would be more assertive at age 6 because of their capacity to regulate anger by dealing directly and constructively with sources of frustration. An alternative explanation of any observed link between information gathering and assertiveness is that both are indicators of verbal ability: Children who tend to ask questions when stressed may be more likely to assert themselves in socially appropriate ways because they have good verbal skills, not because they are adept at controlling their emotions. To examine this possibility, we employed a measure of verbal ability derived from this widely used measure. Specifically, we used scores from a WPPSI-R short form comprised of the Information and Vocabulary subtests. The Information subtest assesses children’s store of general information, while the Vocabulary subtest measures word knowledge.

These subtests were selected because they contribute substantially to the Verbal factor (Mdn loadings = .72 and .65, respectively) and correlate moderately with the Verbal scale (rs = .73 and .66). Scaled scores for I and V were converted to VIQ scores according to procedures described by Sattler (1990). Boys completed the WPPSI-R short form during an age-5½ home visit. 

Results

Following the presentation of descriptive statistics and bivariate correlations among study variables, we describe (1) relations between behavioral regulatory strategies and expressed anger, (2) relations between early child and family factors and regulatory strategies, and (3) relations between regulatory strategies and school self-control.     

Descriptive Statistics and Bivariate Correlations

Table 1 contains means, standard deviations, and ranges of study variables. Attachment security is represented as a dichotomous variable with secure attachment coded 1 and insecure attachment coded 2. The mean attachment score indicates roughly equal numbers of securely and insecurely attached boys (53% versus 47%, respectively), which is similar to rates in other low-income samples (Erickson, Egeland, & Sroufe, 1985). Descriptive statistics for the individual scales that comprise the maternal control variable are presented in the Appendix. Scores for the five behavioral strategies reveal that boys were most likely to engage in distraction during the Cookie Task, followed by passive waiting and focus on the delay object/task. Comfort seeking and information gathering occurred less often. The mean peak anger, .90, indicates that, on average, boys expressed mild frustration during the task. This value is less than 1 because approximately one third (32%) of the sample did not express anger during the Cookie Task. Boys’ anger tended to decrease in a linear fashion across the 3 min period: Mean peak anger scores in the first, second, and third minute of the task were 1.5, .6, and .3, respectively. On average, mothers provided roughly equal levels of encouragement and discouragement of effective regulation during the Cookie Task. Finally, TRF and SSRS scores indicate slightly higher levels of externalizing problems and slightly lower levels of assertiveness and cooperation than were found in the standardization samples for these measures (Achenbach, 1991; Gresham & Elliot, 1990).

To test for race differences in anger regulation, African American and European American boys were compared on mean levels of strategy use and peak anger. To permit comparison, race was dummy coded as follows: African American = 1, European American = 2. No mean-group differences were found, t(280) ranged from .67 for distraction to 1.27 for passive waiting, all ps > .10, indicating that African American and European American boys did not differ significantly in the frequency with which they used individual regulatory strategies or in the intensity of expressed anger during the Cookie Task.  

Relations among study variables are presented in Table 2 using Pearson and point-biserial (for the dichotomous attachment security variable) correlation coefficients. Use of the regulatory strategies (with the exception of comfort-seeking) correlated significantly with the level of peak expressed anger. Additionally, a number of modest but significant correlations were observed among early child/maternal factors, regulatory strategies, and indicators of school self-control. These associations are examined more closely in the analyses that follow.                                  

Relations Between Behavioral Strategies and Angry Affect

We predicted that behaviors that shift attention away from the source of frustration or that clarify situational constraints would correspond with decreases in anger intensity while those that focus attention on the source of frustration would correspond with increases in anger intensity. While the correlations between behavioral strategies and peak anger intensity in Table 2 are consistent with this hypothesis, it is unclear from these data whether the strategies actually exert a regulatory effect. That is, do these behaviors produce changes in expressed emotion, or do they merely accompany the emotional arousal?

To address this issue, we used temporal contingency analyses to examine changes in anger following the occurrence of the five strategies. Following Buss and Goldsmith (1998), we located each coding interval in which a behavioral strategy occurred, then computed a change score for anger intensity (“no change,” “increase,” or “decrease”) from that interval to the subsequent interval. Observed frequencies of the three change scores were computed from the intervals in which a behavioral strategy occurred. For purposes of comparison, we also calculated expected frequencies of change scores for each behavior from intervals in which that behavior did not occur. We then performed χ² goodness-of-fit tests to determine whether observed frequencies differed from expected frequencies for each of the five behavioral strategies. Standardized residuals were used to isolate effects in significant χ² tests. Given our interest in modeling regulatory effects–that is, determining which behaviors are associated with changes in the expression of anger–the focus of these analyses was on the  “increase” and “decrease” contingencies. The meaning of “no change” scores is more ambiguous; in these cases, a behavior either maintained an affective state, or it simply accompanied that state (Buss & Goldsmith, 1998).

Table 3 contains the observed and expected frequencies for each of the three contingencies for each behavior. The χ² goodness-of-fit tests were significant for distraction, χ²(2) = 108.95, p < .001, information gathering, χ²(2) = 44.59, p < .001, passive waiting χ²(2) = 34.35, p < .001, and focus on object/task, χ² (2) = 152.78, p < .001. Examination of the standardized residuals for the “increase” and “decrease” contingencies revealed several different patterns across the four behaviors with an overall effect on expressed anger. Distraction and passive waiting were associated with low frequencies of increased anger; that is, anger displays increased in intensity less than expected after the occurrence of these behaviors. In contrast, in intervals following information gathering, anger decreased more than expected. Focus on object/task was associated with increases in anger, with greater-than-expected increases in intensity following its occurrence.

Before proceeding with the longitudinal analyses, we sought to determine wether the observed (nonsigificant) relation between comfort seeking and expressed anger was contaminated by the coding procedures. Recall that if boys engaged in distraction, passive waiting, information gathering, or focus on the delay object/task while in ongoing contact with their mother, they received a code for comfort seeking and the other relevant strategy. By adopting this procedure, we may have confounded the regulatory effects of comfort seeking with those of the other strategies. To examine this possibility, we ran a second contingency analysis in which only initial bids for physical contact (for which there was no overlap with other codes) were counted as comfort seeking. The results were unchanged from the original analysis, suggesting that the nonsignificant association between comfort seeking and expressed anger was not an artifact of the coding system.

Relations Between Early Child and Maternal Characteristics and Anger Regulation

We hypothesized that child negative emotionality, attachment security, and maternal control in toddlerhood would predict strategies used to regulate anger in early childhood. Hierarchical linear regression was used to control for maternal regulation during the Cookie Task, to sort out redundant predictors, and to test for an interaction between negative emotionality and maternal control. The sequence of variable entry for the prediction of regulatory strategies was as follows: maternal regulation during the Cookie Task, negative emotionality, attachment security, and maternal control. The order of entry of early child and family factors was based on the assumption that child emotionality exerts earlier and more proximal influence on anger regulation than attachment security, which in turn is more proximal in its influence than maternal control (Calkins, 1994). The negative emotionality X maternal control interaction term was entered last and rejected when not statistically significant.

Results of these analyses are presented in Table 4. The regression models were significant for distraction, passive waiting, and information gathering, Fs(6, 277) = 3.02, 2.47, and 2.65,  respectively, ps < .05, and nonsignificant for comfort seeking and focus on object/task. After accounting for the effects of maternal behavior during the Cookie Task and negative emotionality, securely attached boys were more likely than insecurely attached boys to engage in self-distraction (Ms = 11.28 vs. 9.80), to wait quietly (Ms = 5.2 vs. 3.1), and to ask questions about the task (Ms = 2.8 vs. 0.93). Maternal control also accounted for unique variance in distraction: Boys whose mothers relied on positive control at age 1½ used this strategy more frequently than those with harsh, negative mothers. Additionally, the interaction between negative emotionality and maternal control significantly predicted passive waiting. Follow-up analyses revealed that the slopes reflecting the association between negative emotionality and passive waiting differed as a function of maternal control. Consistent with our hypotheses, negative emotionality was associated with lower rates of passive waiting among boys whose mothers’ used negative control (as defined by 1 SD below the sample mean for maternal control), t(280) = 2.15, p < .05; however, negative emotionality and passive waiting had no significant association among boys whose mothers used positive control (as defined by 1 SD above the sample mean for maternal control), t(278) = .87, p = .39.

A final set of regression models predicting strategy use was computed to determine whether relations between maternal control and child anger regulation are moderated by race. For these analyses, race (dummy coded as African American = 1, European American = 2) was entered as a control variable in the hierarchical models described above, followed by a race X maternal control interaction term. Consistent with the mean-group difference tests described previously, race did not account for significant variance in strategy use, with standardized beta coefficients ranging from .003, SE = .39, p = .97 for distraction to .07, SE = .11, p = .35 for passive waiting. Of greater import for the question at hand, race did not interact with maternal control in the prediction of strategy use. Standardized beta coefficients for the interaction term ranged in absolute value from -.05, SE = .14, p = .79, for distraction to .11, SE = .48, p = .51, for information gathering.

Relations between Anger Regulation and School Self-Control

Do anger regulation strategies in early childhood predict self-control at the beginning of middle childhood? If so, do predictive relations hold after accounting for early child and maternal characteristics? To address these questions, teachers’ ratings of externalizing problems, cooperation, and assertiveness were regressed on maternal behavior during the Cookie Task; on age-1½ negative emotionality, attachment security, and maternal control; and on the five strategies. Maternal behavior was entered first, followed by the block of age-1½ predictors. The block of regulatory strategies was entered last.  The overall regression models were significant for externalizing problems and cooperation, Fs(9, 179) = 2.33 and 2.14, respectively, ps < .05. As seen in Table 5, with the exception of comfort seeking, each of the regulatory strategies was predictive of at least one indicator of self-control. Boys who directed attention away from the delay object for much of the delay task, either though active self-distraction or passive waiting, were lower on teacher-rated externalizing problems than boys who used this strategy less frequently. Boys who tended to focus on sources of frustration were rated by teachers as higher on externalizing problems and lower on  cooperation than their peers.

The model for assertiveness was nonsignificant, F(9, 179) = 1.30, p = .24. This finding is not surprising given that information gathering was the only strategy expected to predict assertiveness. When distraction, passive waiting, comfort seeking, and focus on the object/task were excluded, the model became significant F(5, 179) = 3.21, p < .05. To test the possibility that information gathering and assertiveness are linked due to a common association with verbal ability, we entered age-5½ verbal IQ in the regression equation prior to the information gathering score. The association between information gathering and assertiveness remained significant, β = .16, SE = .04, p < .05. Thus, the relationship between information gathering and assertiveness cannot be attributed solely to verbal skills.

The final analysis examined whether boys who employed a greater number of effective

regulatory strategies (i.e., those strategies that are associated with decreased anger: distraction, passive waiting, and information gathering) at age 3½ were better adjusted at age 6 than those who used fewer effective strategies. For purposes of comparison, we grouped boys by the number of effective strategies they used (1-3; all boys with age-6 data used at least one effective strategy). Seventeen boys used only one effective strategy, 65 used two effective strategies, and 97 used all three. Before comparing these groups on measures of self-control, we counted the total number of coding intervals in which boys engaged in one or more effective strategies. This score (M = 14.8, SD = 3.98) was used as a covariate to clarify whether versatility in anger regulation provides additional predictive information above and beyond the amount of time boys spent engaged in effective regulation.

A multivariate analysis of covariance revealed a significant group effect for number of effective strategies used, Wilk’s Λ = .90, F = 2.31 (6, 173), p < .05. Means and standard deviation of self-control scores by number of strategies and the results of univariate and post hoc tests are presented in Table 6. Boys who used all three effective strategies were nearly 1 standard deviation lower on externalizing scores than boys who used only one effective strategy. 

 

Discussion

Anger is a major element in the affective presentation of preschoolers (Goodenough, 1931; Radke-Yarrow & Kochanska, 1990), and of preschool boys in particular (Cole, Zahn-Waxler et al., 1994; Eisenberg et al., 1994), yet by school entry many children can manage their emotions effectively in moderately frustrating conditions. While this transition is remarkable, there is a dearth of information about the processes by which young children regulate their anger, the factors that shape regulatory skills, and the role these skills play in the development of self-control. The present study examined anger regulation in low-income boys, a population for whom regulatory competence may have particularly wide-ranging implications.

Relations Between Behavioral Strategies and Angry Affect in Early Childhood

Results of the temporal contingency analyses suggest that behavioral strategies affected boys’ anger in some instances. Specifically, anger increased less than expected following distraction and passive waiting and decreased more than expected following information gathering. While both patterns are consistent with an inhibitory effect, these findings suggest that information gathering exerted more influence on anger insofar as it seemed to reduce, rather than merely maintain, the level of affective intensity. In contrast, focusing on the delay object or task was associated with subsequent increases in anger intensity. Comfort seeking was unrelated to changes in boys’ anger.

These findings conform to theoretical models of emotion regulation in children (Thompson, 1994) and, in the case of attentional control, which has been considered in prior research, converges with results from infant (Braungart & Stifter, 1991; Buss & Goldsmith, 1998) and toddler (Calkins & Johnson, 1998) samples. As this study is among the first to examine the effects of regulatory behavior on anger in preschool children, it is of interest to consider these findings in light of the developmental level of the participants. In the first year of life, visual shifting affords infants a degree of internal control over sensory stimulation. Johnson, Posner, and Rothbart (1991) found that infants’ ability to disengage attention is associated with their susceptibility to negative emotions. The present results suggest that while motor and cognitive advances in the second and third year support a variety of new ways by which young children may actively modulate exposure to evocative stimuli, the basic regulatory function of attention shifting is preserved. 

If attention shifting behaviors represent attempts to control intake of evocative stimuli, boys’ efforts to learn more about situational constraints imposed during the frustration task could reflect attempts to alter or refine their appraisals of such stimuli (Thompson, 1994). Asking questions about when and how obstacles to goals will be removed may be a particularly sophisticated regulatory strategy for preschool children, both in terms of what is targeted for change (the construal of an event) and the means by which change is sought (language). Clearly, this strategy is functional only in the presence of others; nonetheless, verbal information gathering may presage efforts to alter interpretations of arousing events independently (Stein, Trabasso, & Liwag, 1993).            

Several factors may have accounted for the lack of effect for comfort seeking. One possibility is that children of this age reserve comfort seeking as a regulatory strategy for stressors that are more aversive or more novel than the task used in the present investigation or for situations in which their personal coping resources are depleted. Second, boys may have relied on other strategies to modulate anger because mothers, as possessors of the cookie, were perceived as a source of frustration. These possibilities as well as the broader literature underscore the importance of context in the conceptualization and assessment of emotion regulation (Grolnick et al., 1996; Buss & Goldsmith, 1998). Further research is needed to clarify links between characteristics of the emotional challenge and regulatory strategy use in young children.           

In this investigation, emotion regulation was conceptualized in terms of temporal relations

between behavioral strategies and expressed affect. Specifically, emotion regulation was inferred when affect changed (or did not change) in a regular fashion following the commission of targeted behaviors. Of course, there may also exist temporal regularities among regulatory strategies which in turn may be related to emotional states. Cohn and Tronick (1983) documented a periodic cycling among behavioral states in infancy which, they concluded, inhibits the onset of negative affect. Analogous patterns may exist in childhood, raising interesting questions for future research. For example, are certain sequences of regulatory strategies more effective than others for regulating children’s anger? Are regulatory sequences stable across time and contexts? Are parents’ regulatory efforts more helpful at certain points in the sequence than at other points?¹

Developmental Antecedents of Anger Regulation in Early Childhood         

Consistent with developmental models of emotion regulation, intrinsic and extrinsic factors in toddlerhood were related to young boy’s use of  emotion regulation strategies under frustrating conditions. When multivariate analyses were employed to control for the effects of maternal behavior during the frustration task and to identify nonredundant predictors, attachment security at age 1½ accounted for unique variance in self-distraction, information gathering, and passive waiting. Maternal control at age 1½ contributed additional variance to the prediction of distraction and interacted with negative emotionality in the prediction of passive waiting.

Attachment theory suggests that children who have experienced a history of responsive caregiving develop enduring expectations that their needs will be met (Bowlby, 1969), thereby enabling adaptive responses to emotional challenge (Carlson & Sroufe, 1995). This study provides support for this claim and sheds light on the diversity of strategies employed by securely attached children in the face of frustration. Specifically, boys classified as secure at age 1½ were more likely to disengage from frustrating stimuli and to seek information regarding when and how obstacles would be removed. This versatility in emotion self-regulation may help explain relations between attachment security and various indices of competence found in prior research (Belsky & Cassidy, 1994).

Our results suggest that the links between the parent-child relationship and anger regulation are multiple: Maternal control was predictive of self-distraction after accounting for attachment security. In considering the mechanisms that may account for this relationship, we note that our measure of positive versus negative maternal control assessed specific control techniques and affect toward the child. We expect that both components of maternal control may be relevant to the development of anger regulation. Children of positive mothers may shift attention to less frustrating aspects of the environment because this strategy has been modeled in dyadic interactions and reinforced via positive maternal regard. Indeed, mothers rated high on positive control often made a game of the age-1½ clean-up and puzzle tasks to increase enjoyment for their sons and to encourage compliance; these control efforts were accompanied by warmth and positive feedback. In contrast, rejecting mothers do not provide these positive socialization experiences; in addition, they model anger as a primary affective response to challenging situations and a means of influencing others (Denham, Workman, Cole, Weissbrod, Kendziora, & Zahn-Waxler, 2000).

Prior work implicates dispositional emotionality as a primary precursor of individual differences in emotion regulation (Calkins, 1994; Eisenberg & Fabes, 1992). Our results suggest that negative emotionality does not influence preschool anger regulation by itself; it does so only in conjunction with maternal control. In particular, high negative emotionality predicted lower rates of passive waiting only among boys whose mothers were high on negative control. This finding is consistent with Patterson’s (1982) coercion model, wherein parents reinforce difficult child behavior through hostile, inconsistent caregiving. In this case, mothers who are punitive or rejecting in their attempts to reduce (or to reduce exposure to) child irritability may unwittingly hamper the acquisition of effective regulatory strategies. In turn, deficits in child anger regulation may increase the likelihood of coercive interactions in middle childhood.

Anger Regulation and Self-Control at School Entry

Consistent with the results of Eisenberg and colleagues (1994, 1996), boys who reoriented attention away from sources of frustration at age 3½, either through active self-distraction or passive waiting, were lower on teacher-rated externalizing problems than their peers. Boys who focused attention on sources of frustration were perceived as less cooperative and higher on externalizing problems than those who used this strategy infrequently.

Attention-shifting was not predictive of all measures of self-control, however. As hypothesized, the tendency to shift attention toward or away from sources of frustration was unrelated to assertiveness. Rather, assertiveness associated uniquely with information gathering. To our knowledge, these results are the first to show that particular regulatory strategies relate in specific ways to different aspects of self-control. This pattern adds further weight to a point argued persuasively by Thompson and Calkins (1996): The ‘effectiveness’ of a given strategy depends not only on its impact on emotional experience, but also on the demands of the social context in which it is used. Thus, while attention shifting and information gathering both minimize anger under frustrating circumstances, the former strategy might be most useful in situations that require immediate cooperation or compliance, while the latter may be most useful in situations that are more ambiguous or that require children to solve social problems on their own initiative.

In light of the perspective advocated by Thompson and Calkins, it is not surprising that boys who used three different effective regulatory strategies were viewed by teachers as less aggressive and disruptive than boys who used just one. This result suggests that boys who can draw on alternative strategies when one approach proves ineffective are less likely to act out in the face of frustration. Taken together, the findings on school outcomes indicate that skill in the use of particular strategies and flexibility in the application of multiple strategies are both important achievements on the path to self-control.

Several limitations warrant caution in the interpretation of these results. First, the study was intended to illuminate socioemotional processes associated with the development of self-control in low-income boys. While similar relations between regulatory strategies and angry affect and between attentional control and social competence have been observed in middle-class samples of young boys and girls (Calkins & Johnson, 1998; Grolnick et al., 1996; Eisenberg et al., 1994; Raver, 1996), the generality of other findings is unclear. For example, the impact of temperamental factors on anger regulation may vary across socioeconomic groups. Low-income families face unique stressors that could elevate the relative strength of external influences (Garcia, Shaw, & Gilliom, 1999). For instance, financial strain and maternal job loss predict increased punishment and harsher discipline practices (McLoyd, Jayartane, Ceballo, & Borquez, 1994). These processes may have partially overridden effects of negative emotionality in this sample. Additionally, prior work with toddlers indicates sex differences in the use of anger regulation strategies (Raver, 1996). Differences also may be evident in early childhood, the stage at which sex differences in self-control first appear. Clearly, further work with diverse samples of boys and girls is needed.

Second, anger regulation was assessed in only one context. The management of frustration may be evaluated in any situation in which personal goals are blocked. Responses to parental limit-setting in the home, confrontations with sibling and peers, and encounters with objects that exceed children’s developmental capabilities will provide useful, complementary information about the regulation of anger in childhood. For example, assessment of anger in natural environments can yield a broader range of anger than may be obtained in the laboratory. While the paradigm used in this study permitted a fairly fine-grained analysis of anger regulation, few boys became extremely angry and a significant number did not exhibit any anger at all. Work is needed to determine whether the choice and/or effectiveness of specific regulatory strategies vary with the intensity of anger.

A third limitation is the reliance on an observational measure of emotion regulation. Advances in the assessment of central and peripheral nervous system activation allow a more complete understanding of emotion and emotional control (Fox, 1989). Psychophysiological measures may be particularly helpful to determine the extent to which strategy use is influenced by rising emotional arousal. Finally, it is important to note that the significant effects in this study were, without exception, modest in magnitude. Stronger relations may emerge across shorter time periods or with more comprehensive measures of anger regulation.

In summary, this study provides new information about anger regulation in young boys from high-risk environments. Young boys drew from a variety of behavioral strategies in response to frustration, and these strategies predicted specific changes in angry affect. Individual differences in strategy use were related to child and maternal factors in toddlerhood and to indices of self-control at school entry. If replicated, the results point the way toward foci for interventions with at-risk children. Efforts to support parents as they deal with the challenges of infancy and toddlerhood may enhance their children’s ability to regulate anger, which in turn could increase self-control and prevent conduct problems. In cases where deficits in self-control are already present, these problems may be addressed by teaching effective anger regulation strategies.

References

Achenbach, T. M. (1991). Manual for teacher’s report form and 1991 profile. Burlington: University of Vermont, Department of Psychiatry.

Achenbach, T. M., Edelbrock, C., & Howell, C. (1987). Empirically-based assessment of the behavioral/emotional problems of 2- and 3-year-old children. Journal of Abnormal Child Psychology, 15, 629-650.

Ainsworth, M. D., Blehar, M., Waters, E., & Wall, S. (1978). Patterns of attachment: A psychological study of the Strange Situation. Hillsdale, NJ: Erlbaum.

Bates, J. E., Freeland, C. A., & Lounsbury, M. L., (1979). Measurement of infant difficultness.             Child Development, 50, 794-803.

Bates, J. E., Maslin, C., & Frankel, K. (1985). Attachment security, mother-child interaction, and temperament as predictors of behavior-problem ratings at age 3 years. In I. Bretherington & E. Waters (Eds.), Growing points in attachment theory and research. Monographs of the Society for Research in Child Development, 50(1-2, Serial No. 209, pp. 167-193).

Baumrind, D. (1972). An exploratory study of socialization effects on black children: Some black-white comparisons. Child Development, 43, 261-267.

Belsky, J., & Cassidy, J. (1994). Attachment: Theory and evidence. In M. Rutter & D.

Hay (Eds.), Development through life: A handbook for clinicians (pp. 373-402). Oxford: Blackwell.

Berkowitz, L. (1962). Aggression: A social psychological analysis. New York: McGraw Hill.

Bohart, A. C., & Stipek, D. J. (Eds.). (2001). Constructive and destructive behavior: Implications for family, school, and society. American Psychological Association: Washington, DC. 

Bolger, K. E., Patterson, C. J., Thompson, W. W., & Kupersmidt, J. B. (1995). Psychosocial adjustment among children experiencing persistent and intermittent family economic hardship. Child Development, 66, 1107-1129.

Bowlby, J. (1969). Attachment. New York: Basic Books.

Braungart, J. M., & Stifter, C. A. (1991). Regulation of negative reactivity during the strange situation: Temperament and attachment. Infant Behavior and Development, 14, 349-364.

Braungart, J. M., & Stifter, C. A. (1996). Infant responses to frustrating events: Continuity and change in reactivity and regulation. Child Development, 67, 1767-1779.

Bridges, L. J., & Grolnick, W. S. (1995). The development of emotional self-regulation in infancy and early childhood. In N. Eisenberg (Ed.), Review of personality and psychology (pp. 185-211). Newbury Park, CA: Sage.                     

Buss, K. A., & Goldsmith, H. H. (1998). Fear and anger regulation in infancy: Effects on the temporal dynamics of affective expression. Child Development, 69, 359-374.

Caldarella, P., & Merrell, K. W. (1997). Common dimensions of social skills of children and adolescents: A taxonomy of positive behaviors. School Psychology Review, 26, 264-278.

Calkins, S. (1994). Origins and outcomes of individual differences in emotion regulation. In N. A. Fox (Ed.), The development of emotion regulation: Biological and behavioral considerations. Monographs of the Society for Research in Child Development, 59,(2-3, Serial No. 240, pp. 53-72). Chicago: University of Chicago Press.

Calkins, S., & Johnson, M. C. (1998). Toddler regulation of distress to frustrating events: Temperamental and maternal correlates. Infant Behavior and Development, 21, 379-395.

Campbell, S. B., Shaw, D. S., & Gilliom, M. (2000). Early externalizing behavior problems: Toddlers and preschoolers at risk for later maladjustment. Development & Psychopathology, 12, 467-488.

Campos, J., Campos, R. M., & Barrett, K. C. (1989). Emergent themes in the study of emotional development and emotion regulation. Developmental Psychology, 25, 394-402.

Carlson, E., & Sroufe, A. (1995). Contribution of attachment theory to developmental psychopathology. In D. Cicchetti & D. Cohen (Eds.), Developmental psychopathology: Vol. I. Theory and methods (pp. 581-617). New York: Wiley.

Cohn, J. F., & Tronick, E. Z. (1983). Three-month-old infants reactions to simulated maternal depression. Child Development, 54, 185-193.

Cole, P. M., & Zahn-Waxler, C. (1992). Emotional dysregulation in disruptive behavior disorders. In D. Cicchetti & S. Toth (Eds.), Rochester symposium  on developmental psychopathology: Developmental perspectives on depression (pp. 173-209).  Rochester: University of Rochester Press.

Cole, P. M., Michel, M. K., & Teti, L. (1994). The development of emotional regulation and

dysregulation: A clinical perspective. In N. A. Fox (Ed.), The development of emotion regulation: Biological and behavioral considerations. Monographs of the Society for             Research in Child Development, 59,(2-3, Serial No. 240, pp. 73-100). Chicago: University of Chicago Press.

Cole, P. M., Zahn-Waxler, C., & Smith, K. D. (1994). Expressive control during a

disappointment: Variations related to preschoolers’ behavior problems. Developmental Psychology, 30, 835-846.                                               

Dalgleish, T., & Power, M. J. (Eds.). (1999). Handbook of cognition and emotion. John

Wiley and Sons: Chichester, England.

Denham, S. A., Workman, E., Cole, P. M., Weissbrod, C., Kendziora, K. T., & Zahn-Waxler, C. (2000). Prediction of externalizing behavior problems from early to middle childhood: The role of parental socialization and emotion expression. Development & Psychopathology, 12, 23-45.

Eisenberg, N., & Fabes, R. A. (1992). Emotion, regulation, and the development of social

competence. In M. S. Clark (Ed.), Review of personality and social psychology, Vol. 14.        Emotion and social behavior (pp. 119-150). Newbury Park, CA: Sage.

Eisenberg, N., Fabes, R. A., Guthrie, I. K., Murphy, B. C., Maszk, P., Holmgren, R., & Suh,

K. (1996). The relations of regulation and emotionality to problem behavior in elementary school children. Development & Psychopathology, 8, 141-162.

Eisenberg, N., Fabes, R. A., Nyman, M., Bernzweig, J., & Pinuelas, A. (1994). The relations of

emotionality and regulation to children’s anger-related reactions. Child Development, 65, 109-128.

Ekman, P., & Friesen, W. F. (1978). Facial action coding system (FACS): A technique for the     measurement of facial action. Palo Alto, CA: Consulting Psychologists Press.

Elder, G. H. (1979). Historical change in life patterns and personality. In P. B. Blates & O. G.

Blim (Eds.), Life-span development and behavior: Vol. 2 (pp. 117-159). New York: Academic Press.

Erickson, M., Egeland, B., & Sroufe, L. A. (1985). The relationship between quality of

attachment and behavior problems in preschool in a high-risk sample. In I. Bretherton & E. Waters (Eds.), Growing points in attachment theory and research. Monographs of the Society for Research in Child Development, 50 (pp. 147-186).

Fox, N. A. (1989). Psychophysiological correlates of emotional reactivity in the first year of life.       Developmental Psychology, 25, 364-372.

Frankel, K. A., & Bates, J. E. (1990). Mother-toddler problem solving: Antecedents in

attachment, home behavior, and temperament. Child Development, 60, 810-819.

Garcia, M. M., Shaw, D. S., & Gilliom, M. (1999, April). Early family contributors to emotion regulation at school entry: a longitudinal study of boys from low-income families. Paper

presented at the meeting of the Society for Research on Child Development, Albuquerque, NM.

Garcia Coll, C., Akerman, A., & Cicchetti, D. (2000). Cultural influences on developmental

processes and outcomes: Implications for the study of development and psychopathology.         Development and Psychopathology, 12, 333-356.

Goodenough, F. L. (1931). Anger in young children. Minneapolis: University of Minnesota

Press.

Gresham, F. M., & Elliot, S. N. (1990).  Social Skills Rating System Manual. Circle Pines, MN:

American Guidance Service.

Grolnick, W. S., Bridges, L. J., & Connell, J. P. (1996). Emotion regulation in 2-year-olds:

Strategies and emotional expression in four contexts. Child Development, 67, 928-941.

Gross, J. J. (1999). Emotion regulation: Past, present, future. Cognition and Emotion, 13,

551‑573.

Izard, C. E. (1979). The maximally discriminative facial movement coding system (MAX).

Newark, DE: University of Delaware, Instructional Resource Center.

Johnson, M. H., Posner, M. I., & Rothbart, M. K. (1991). Components of visual orienting in early

infancy: Contingency learning, anticipatory looking, and disengaging. Journal of Cognitive          Neuroscience, 4, 335-344.

Keenan, K., & Shaw, D. S., (1997). Developmental and social influences on young girls' early

problem behavior. Psychological Bulletin, 21, 95‑113.

Kopp, C. B. (1989). Regulation of distress and negative emotions: A developmental view.

Developmental Psychology, 25, 343-354.

Kopp, C. B. (1991). Young children's progression to self‑regulation. In M. Bullock (Ed.), The

development of intentional action: Cognitive, motivational, and interactive processes (pp. 38‑54). Basel, Switzerland: Karger.

Lochman, J. E., & Lampron, L. B. (1986). Situational social problem-solving skills and self-

esteem of aggressive and nonaggressive boys. Journal of Abnormal Child Psychology, 14, 605-617.

Loeber, R., & Farrington, D. P. (2000). Young children who commit crime: Epidemiology,

developmental origins, risk factors, early interventions, and policy implications.                         Development & Psychopathology, 12, 737-762.

Main, M., & Solomon, J. (1986). Discovery of a new, insecure-disorganized/disoriented

attachment pattern: Procedure, findings, and implications for the classification of behavior. In M. Yogman & T. B. Brazelton (Eds.), Affective development in infancy (pp. 95-124). Norwood, NJ: Ablex.

Martin, J. A. (1981). A longitudinal study of the consequences of early mother-infant interaction:

A microanalytic approach. Monographs of the Society for Research in Child Development,      46(3).

Marvin, R. S. (1977). An ethological-cognitive model for the attenuation of mother-infant

attachment behavior. In T. M. Alloway, L. Krames, & P. Pliner (Eds.), Advances in the study of communication and affect: Vol. 3. The development of social attachments (pp. 25-60). New York: Plenum.

Matas, L., Arend, R. A., & Sroufe, L. A., (1978). Continuity of adaptation in the second year:

The relationship between quality of attachment and later competence. Child Development,        49, 547-556.

McLoyd, V. C., Jayaratne, T. E., Ceballo, R. & Borquez, J. (1994). Unemployment and work

interruption among African American single mothers: Effects on parenting and adolescent socioemotional functioning. Child Development, 65, 562-589.

Patterson, G. R. (1982). A social learning approach: Vol. 3. Coercive family process. Eugene,

OR: Castalia.

Radke-Yarrow, M. & Kochanska, G. (1990). Anger in young children. In N. L. Stein, B.

Leventhal, & T. Trabasso (Eds.), Psychological and biological approaches to emotion (pp. 297-312). Hillsdale, NJ: LEA.

Raver, C. (1996). Relations between social contingency in mother-child interaction and 2-year-

olds’ social competence. Developmental Psychology, 32, 850-859. 

Richman, N., Stevenson, J. & Graham, P. (1982). Preschool to school: A behavioral study. New

York: Academic Press.

Rose, S. L., Rose, S. A., & Feldman, J. F. (1989). Stability of behavior problems in very young

children. Development & Psychopathology, 1, 5-19.

Rothbart, M. K., Ahadi, S. A., & Hershey, K. L. (1994). Temperament and social behavior in

childhood. Merrill-Palmer Quarterly, 40, 21-39.

Rothbart, M. K., & Bates, J. E. (1998). Temperament. In W. Damon (Series Ed.) & N. Eisenberg

(Vol. Ed.), Handbook of child psychology: Vol. 3. Social, emotional, and personality development (5th ed., pp.105-176) New York: Wiley.

Saarni, C., Mumme, D. L., & Campos, J. J. (1998). Emotional development: Action,

communication, and understanding. In W. Damon (Series Ed.) & N. Eisenberg (Vol. Ed.), Handbook of child psychology: Vol. 3. Social, emotional, and personality development (5th ed., pp.237-310). New York: Wiley.

Sattler, J. M. (1990). Assessment of children (3rd ed.). San Diego: Jerome M. Sattler.

Shaw, C., & McKay, H. (1942). Juvenile delinquency and urban areas. Chicago: University of

Chicago Press.

Shaw, D. S., & Bell, R.Q. (1993). Developmental theories of parental contributors to antisocial

behavior. Journal of Abnormal Child Psychology, 21, 493-518.

Stein, N. L., Trabasso, T., & Liwag, M. (1993). The representation and organization of emotional

experience: Unfolding the emotion episode. In J. M. Haviland & M. Lewis (Eds.), Handbook     of emotions (pp. 279‑300). New York: The Guilford Press.                  

Stevenson, J., Richman, N., & Graham, P. (1985). Behaviour problems and language abilities at

three years and behavioural deviance at eight years. Journal of Child Psychology and                      Psychiatry, 26, 215-230.

Thompson, R. A. (1994). Emotion regulation: A theme in search of a definition. In N. A. Fox

(Ed.), The development of emotion regulation: Biological and behavioral considerations. Monographs of the Society for Research in Child Development, 59,(2-3, Serial No. 240, pp. 25-52). Chicago: University of Chicago Press.

Thompson, R. A., & Calkins, S. D. (1996). The double-edge sword: Emotion regulation for

children at risk. Development and Psychopathology, 8, 163-182.

Wechsler, D. (1989). Manual for the Wechsler Preschool and Primary Scale of Intelligence –        Revised. San Antonio: The Psychological Corporation.

Winslow, E. B., Shaw, D. S., Bruns, H., & Kiebler, K. (1995). Parenting as a mediator of

child behavior problems and maternal stress, support, and adjustment. Paper presented at the biennial meeting of the Society for Research in Child Development, Indianapolis, IN.

Appendix

Mean Scores and Standard Deviations of Maternal Control Scales (N = 310)

Scale                                                       M                      (SD)                Min.                   Max.

Clean-Up Task

Global Hostility                                 1.10                     (.40)                  1                         4

Global Warmth                                 2.82                     (.44)                  1                         3

Global Punitiveness                           1.12                     (.38)                  1                         3    

Critical Statements^a                           5.63                   (5.50)                  0                       37

Verbal Approval^a                             10.43                   (9.49)                  0                       44

Puzzle Task

Global Hostility                                 1.06                     (.27)                  1                         4

Global Warmth                                 2.85                     (.43)                  1                         3

Global Punitiveness                           1.15                     (.44)                  1                         3    

Critical Statements^a                           4.94                   (4.28)                  0                       23

Verbal Approval^a                               5.97                   (5.47)                  0                       25

^a Scores reflect counts.

Author Note

Miles Gilliom, Daniel S. Shaw, Joy E. Beck, Michael A. Schonberg, JoElla L. Lukon, and Emily B. Winslow, Department of Psychology, University of Pittsburgh.

This study is based in part on the first author’s master’s thesis submitted to the faculty of the University of Pittsburgh. The work was supported by Grant MH50907 and National Research Service Award 1F31MH12226 from the National Institute of Mental Health. The third and fourth authors contributed equally to this project.   

We thank Celia Brownell, Susan Campbell, and Jeffrey Cohn for valuable feedback on a previous version of this paper. We are indebted to the families of the Pitt Mother and Child Project for many years of generous participation. 

Correspondence concerning this article should be addressed to Miles Gilliom, 1269 Sixth Avenue, Apartment 1, San Francisco, California, 94122. Electronic mail may be sent to potatoseason@yahoo.com.

Footnotes

¹ We thank an anonymous reviewer for suggesting this analysis.

² We thank an anonymous reviewer for raising some of these questions.

Table 1

Mean Scores and Standard Deviations of Study Variables

Variable                                                  M                     (SD)                  Min.                   Max.

Age-1½ child/maternal variables (N = 310)

Difficultness^a                                   23.41                   (6.52)               -1.74                   2.30

Attachment security^b,c                         1.47                 (.50)              1 (53%)                2 (47%)

Maternal control^b                                .00                   (1.00)             -3.39                     1.74 

Age 3½ anger regulation variables^b (N = 282)

Waiting strategies^d                              

Distraction                                  8.99                   (5.42)                   1                       18  

Passive waiting                             3.62                   (4.80)                   0                       18

Information gathering                1.71                   (1.39)                   0                         8  

Comfort seeking                         2.03                   (4.69)                   0                       18  

Focus on delay object/task          3.32                   (4.65)                   0                       18  

Anger intensity                                    .90                     (.93)                   0                         3  

Maternal regulation                          3.34                   (1.12)                   1                         5  

Age 6 school self-control variables^e (N = 189)

Externalizing                                   52.46                 (10.68)                 39                       91  

Assertiveness                                    10.95                   (4.07)                   0                       19

Cooperativeness                              13.50                   (4.77)                   1                       20

^a Maternal report. ^b Observational measure. ^c 1 = Securely attached; 2 = Insecurely attached. ^d Scores reflect the number of 10-sec coding intervals in which a behavior was observed. ^e Teacher report.

Table 2                                                  

Correlations Among Age-1½ Child and Maternal Variables, Age-3½ Anger Regulation Variables, and Age-6 Self-Control Variables