Emotional Self-Regulation 1 Running Head: PRESCHOOL DEVELOPMENT OF EMOTIONAL SELF-REGULATION STRATEGIES Preschool children’s development of emotional self-regulation strategies in a sample at- risk for behavior problems

Relatively little longitudinal research has been conducted on the development of children’s emotional self-regulation skills (SRS) use after infancy, particularly for children at high risk for problems in regulating emotions. The current study attempted to validate Kopp’s theory (1989) on the development of children’s regulation of negative emotions during the first few years of life. This study used a repeated observational assessment of a delay of gratification task at ages 2, 3, and 4 with both variableand person-oriented analyses with a low-income sample of children initially selected on the basis of early problem behavior. Results were consistent with Kopp’s theory on the developmental progression of children’s emotional self-regulation strategies. Children initially used more other-oriented SRS (i.e. physical comfort seeking) and transitioned to greater use of independent SRS (i.e. information gathering) by age 4. Results are discussed as a platform for future research that might examine the development of SRS in toddlers and preschoolers. Emotional Self-Regulation 3 Preschool children’s development of emotional self-regulation strategies in a sample atrisk for behavior problems Emotional self-regulation strategies (SRS) encompass a variety of deliberate or automatic cognitive and physiological processes, including anticipating and dealing with existing or anticipated stressors by modulating behavior (Aspinwall & Taylor, 1997). Kopp’s (1989) model and related theoretical accounts (e.g., Blair, 2002; Calkins, 2004; Karoly, 1993) have suggested pathways for the development of emotional SRS; however, most of this research has relied on concurrent or cross-sectional data. A related body of literature suggests that less adaptive use of emotional SRS predicts increased risk for externalizing and internalizing problems (Buckner, Mezzacappa, & Beardslee, 2003; Gilliom, Shaw, Beck, Schonberg, & Lukon, 2002; Silk, Shaw, Forbes, Lane & Kovacs, 2006). Thus, there is a need for longitudinal research on normative changes in emotional SRS and individual differences in emotional SRS trajectories. To address this gap in the literature and inform future theoretical, empirical, and applied innovations, the present study examined the development of emotional SRS across the toddler and preschool periods using variableand person-oriented approaches with a sample of boys at risk for the development of early-starting conduct problems. Development of emotional self-regulation strategies in young children It has been postulated that the achievement of emotional self-regulation begins very early in life (Kopp, 1989) and continues to develop throughout the early school years (Kochanska & Knaack, 2003; Posner & Rothbart, 2000), changing most rapidly during the toddler and preschool periods (Kochanska & Knaack, 2003; Vaughn, Kopp, & Krakow, 1984). Through the development of neurological skills (e.g., attention in infancy) and experiences interacting with caregivers (Blair, 2002; Calkins & Howse, 2004; Posner & Rothbart, 2000), the child’s brain gradually becomes better prepared to handle emotion-inducing situations (Posner & Rothbart, 2000). It is this process that is Emotional Self-Regulation 4 hypothesized to contribute to individual differences in children’s ability to utilize different SRS effectively, leading to individual differences in behavioral control (Calkins & Howse, 2004). Kopp’s (1989) theory suggests that in the first two months of infancy, children primarily use reflexive SRS (e.g., head turning) to alleviate distress. By about the third month, children move toward more voluntary forms of self-soothing and social communication (e.g., crying brings a caregiver to soothe the infant; Kopp, 1989). This notion has received empirical support, with most strategies after the initial months of infancy shown to involve social communication directed at caregivers (Grolnick, Bridges & Connell, 1996; Mangelsdorf, Shapiro, & Manzolf 1995; Rothbart, Ziaie, & O’Boyle, 1992). For example, Brungart-Reiker and Stifter’s (1996) longitudinal study of 5and 10month old infants indicated that older infants attempted to use communication with their caregiver more and avoidance of the stressful situation less than younger infants. Kopp also has postulated that after the first few months, infants decrease some of their dependence on the caregiver and begin to engage in basic forms of voluntary selfregulation when they encounter distressing situations. For example, Mangelsdorf and colleagues (1995) found young infants (6 months of age) used more reflexive strategies such as gaze aversion to avoid a negative stimulus, while older infants (12 and 18 months of age) used more self-soothing and self-distraction, a finding that has been replicated (Parritz, 1996). As the child moves from late infancy to the toddler period, the model suggests that children become more purposeful and organized rather than reactive in their selfregulation and begin distracting themselves from the source of distress. Research on this issue also has received some empirical validation (Diener & Mangelsdorf, 1999; Grolnick, Bridges & Connell, 1996; Rothbart, Ziaie, & O’Boyle, 1992). For example, Grolnick and colleagues (1996) found that during a separation task, 2-year-olds’ use of Emotional Self-Regulation 5 active distraction was related to lower negative emotions, whereas physical comfort seeking was not necessarily related to declines in negative emotions. It is believed that from toddlerhood through the preschool years, children continue to develop more sophisticated regulation strategies in conjunction with their burgeoning advances in cognitive and language skills, abilities that increase their selfcontrol (Calkins & Howse, 2004; Dodge, 1989; Kochanska, Murray & Coy, 1997; Kopp, 1989; Thompson, 1990). During the second and third years children are hypothesized to begin to understand how certain situations tend to arouse specific emotions and can therefore organize and monitor their behavior more effectively to obtain control over their behavior in these situations (Calkins, 2004; Calkins & Howse, 2004). Data generally support these suppositions. In one cross-sectional study of preschool children, age differences were found indicating that 3-year-old children used strategies that focused on the desired object more than 4-year-old children (Stansbury & Sigman, 2000). In Kochanska and colleagues’ (2001) longitudinal study of effortful control (EC) development, a concept similar to emotional self-regulation, moderate intra-individual stability was found in children’s performance on observational EC tasks from 2 to 5 years of age, with children demonstrating an increasing ability to delay gratification over time (Kochanska, Murray, & Harlan, 2000; Kochanska et al., 2001). By the end of the preschool period, Kopp (1989) hypothesizes that children have developed new and increasingly complex ways of regulating their emotions, a supposition corroborated by concurrent data (Diener & Mangelsdorf, 1999; Mendez, Fantuzzo, & Cicchetti, 2002). For example, Silk and colleagues (2006) used a crosssectional study of 4-, 5-, and 7-year old children in a delay of gratification task and found that older children used significantly more distraction and less focus on the delay object than younger children. Preschool children’s increasing proficiency with the use of SRS such as active distraction allows them to select behavioral means to improve a Emotional Self-Regulation 6 distressing situation, including the use of such strategies as playing with enjoyable toys or initiating positive interactions with peers (Denham, 1998). However, Cole, Michel and Teti (1994) note that when preschool children become over-aroused or lack the coping ability for a specific situation, they may regress to formerly used, less developmentally sophisticated strategies. It should be noted that in addition to developmental changes due in large part to biological maturation, parenting and other socialization factors most likely directly and indirectly influence children’s development of SRS (e.g., Calkins & Howse, 2004; Posner & Rothbart, 2000). However, the focus of the current paper is to describe the longitudinal change in low-income children’s SRS use rather than focus on specific moderating factors. Variable-centered Approaches to Understand Developmental Change As the preceding discussion demonstrates, there appears to be a high degree of consensus about the probable developmental pathway for children’s emotional SRS, based primarily on Kopp’s (1989) model and extant empirical support. However, with the exception of research on the stability of EC and developmental change in global measures of self-regulation (e.g., Berlin & Cassidy, 2003; Kochanska et al., 2000, 2003; Pauli-Pott, Mertesacker, & Beckmann, 2004; Raikes et al., 2007; Rothbart, Ziaie, & O’Boyle, 1992; Spinrad, Stifter, Donelan-McCall, & Turner, 2004), little longitudinal research has examined the development of emotional SRS in early childhood; therefore lacking data on the developmental progression of emotional SRS among the same individuals across time. When a longitudinal design has been used, typically there were few assessments (i.e., two) and/or a long interval between assessments (e.g., assessments at ages 3 and 6), making it difficult to draw conclusions about patterns and rate of change. The current study seeks to advance our understand of the development of children’s SRS in the toddler and preschool period by providing three waves of Emotional Self-Regulation 7 longitudinal data on a cohort of boys utilizing the same observational coding system to examine the stability and development of SRS during early childhood. In addition, few studies on the development of SRS have included low-income children or samples at heightened risk for emotional or behavioral problem even though it is especially important to examine emotional developmental issues within low-income, high-risk samples (Raver, 2004). Though some researchers have suggested that emotional SRS may not differ between low-income and middle-income samples (Garner & Spears, 2000), most researchers agree that the environmental stressors that children growing up in low-income homes experience leaves them at a greater risk for poor emotional and social outcomes (Evans, 2004; Raver, 2004; Thompson & Calkins, 1996). Examination of longitudinal change in low-income samples has the potential to advance our understanding of the developmental progression of emotional SRS for children at increased risk for later problem behavior. Thus, the first goal for the present study was to examine changes in emotional SRS strategy use over time using a variable-oriented approach in a high-risk sample of boys. Individual Differences in Emotional SRS and Outcomes In addition to cross-sectional and limited longitudinal reports on normative change in emotional SRS, the value of Kopp’s (1989) model and related theoretical perspectives are validated by research linking proficiency with emotional SRS during early childhood and important social and behavioral outcomes (e.g., Gilliom et al., 2002; Raver, Blackburn, Bancroft, & Torp, 1999; Silk et al., 2006). For example, preschooland school-age children’s emotional SRS use has been related to concurrent internalizing and externalizing outcomes (Silk et al., 2006; Stansbury & Zimmerman, 1999). Similarly, parent and teacher ratings and observational measures of children’s use of specific emotional SRS (e.g., seeking comfort from a caregiver or actively distracting oneself in an emotion-invoking situation) have been associated with children’s Emotional Self-Regulation 8 behavioral outcomes at school (Blair, Denham, Kochanoff, & Whipple, 2004; Eisenberg, Fabes, Nyman, Bernzweig, & Pinuelas, 1994; Gilliom et al., 2002) and at home (Feldman & Klein, 2003; Stansbury & Zimmerman, 1999). For example, children found to be adept at shifting their focus in an emotional situation were not only less likely to show increases in negative affect during the situation, but even years later were less likely to exhibit externalizing behavior and more likely to be cooperative in school (Eisenberg et al. 1996; Gilliom et al., 2002). In addition, the development and implementation of ineffective SRS appears to be related to variables associated with externalizing behavior such as negative emotionality and self-control (Gilliom et al., 2002). The consistent associations demonstrated between early emotional SRS and social, behavioral, and cognitive outcomes suggests a need for research on individual differences in emotional SRS development across toddlerhood and the preschool period. Although examination of individual differences at a single time point informs the understanding of worthy SRS targets for intervention, examining individual differences in emotional SRS over time can help to uncover non-normative trajectories that may lead to later problems. For example, some children may begin to use planful SRS such as active distraction in early childhood, but their tendency to use these strategies may not increase as rapidly as the majority of their peers’ use of these strategies, leaving them at risk for increased behavior problems as the social and behavioral demands increase in the preschool environment. Person-centered approaches provide a valuable complement to variable-oriented approaches by examining trajectories of groups of children that use specific emotional SRS across the toddler and preschool periods (Nagin, 1999; Shaw, Gilliom, Ingoldsby, & Nagin, 2003). Two important features of the present sample make the current analysis of emotional SRS development particularly noteworthy. First, members of the sample were selected based on their increased risk for conduct problems. As reviewed above, links Emotional Self-Regulation 9 exist between externalizing problems at school entry and a tendency to less frequently use planful strategies more frequently focus on distressing events (Gilliom et al., 2002). Thus, young children who attain the ability to use more adaptive emotional SRS show more self-control and less conduct problems. It may be the specific strategies that children learn to use over time that promote self-control and decrease risk for conduct problems (Gilliom et al., 2002). Person-centered analysis of the at-risk sample in the current study can contribute to our understanding of longitudinal deviations from the normative development of strategy use. Secondly, our sample consisted of boys. Although gender differences in emotional SRS in early childhood have received limited attention, it is reasonable to speculate that boys may show unique developmental trajectories of emotional SRS across the toddler and preschool periods. The majority of research supports a gender difference in rates of conduct problems with boys beginning to show higher rates than girls around beginning at age 4 (Keenan & Shaw, 1997). Given the importance of socialization for both the gender difference in conduct problems and for the development of emotional SRS usage (Keenan & Shaw, 1997; Kopp, 1989), it would be reasonable to expect a number of boys to show wide variability in trajectories of SRS from the toddler to preschool period. In sum, the broad goal of the current study was to examine the longitudinal development of boys’ emotional SRS using Kopp’s (1989) theory to guide hypotheses. First, using a variable-oriented approach, we expected the mean ratio of boys’ use of individual types of SRS to support Kopp’s model. Accordingly, we expected to see increased use of more sophisticated and planful strategies and the decrease of less sophisticated, more emotion-focused strategies, such as self-soothing, as a function of age. Second, it was anticipated that when children’s change in SRS over time was modeled using a person-oriented approach, the majority of children would follow a Emotional Self-Regulation 10 pattern similar to Kopp’s model (i.e., use of more sophisticated strategies over time. However, due to the inclusion of boys screened on the basis of high risk for earlystarting conduct problems, it was anticipated that a smaller group of boys would be identified that deviated from Kopp’s model, specifically continuing to use less sophisticated SRS through the preschool period (e.g., continued use of self-soothing and less use of active distraction). Method