Emotional overinvolvement (EOI) in parents’ Five Minute Speech Samples (FMSS; Magaña-Amato, 1993) is thought to measure overconcern and enmeshment with one’s child. Although related to maladaptive outcomes in studies of adult children, FMSS EOI evidences varied relations with behavior problems in studies with young children. These mixed findings may indicate that certain FMSS EOI criteria reflect inappropriate and excessive involvement with adult children, but do not indicate maladaptive processes when parenting younger children. The first study evaluated the relevance of the FMSS EOI construct for young children’s adjustment given mixed relations in the extant literature that have prompted concern regarding the validity of EOI as a measure of pathological parental overinvolvement with young children. Building upon findings from the first study, the second study examined whether the relation of self-sacrifice/overprotection (SSOP), the most prominent and theoretically pernicious EOI criterion, with elevated levels of child internalizing and attention/hyperactivity problems four years later (at age 8) could be accounted for by observed parenting (i.e., poor support, high intrusion, high hostility) during an intervening assessment at age 6.

These studies utilized a community sample of 223 child-mother dyads (47.9% female; Mage_W1 = 49.08 months, Mage_W2 = 73.34 months, Mage_W3 = 97.66 months; 56.5% Hispanic/Latina). In the first study, I evaluated the relations of each FMSS EOI criterion with changes in child behavior problems from preschool to first grade (using data from Waves 1 and 2). Findings from the first study indicated that that both the self-sacrifice/overprotection (SSOP) and statements of attitude (SOAs) FMSS EOI criteria predicted increased externalizing problems. In contrast, excessive detail and exaggerated praise were not related to child externalizing behavior problems, and emotional display was not evident in this sample. None of the FMSS EOI criteria evidenced significant relations with internalizing behavior problems. We found evidence of moderation of the effect of SOAs by gender, such that SOAs contributed to increased externalizing problems among boys but not girls. Relations did not differ by maternal race/ethnicity (Hispanic/Latina vs. non-Hispanic/Latina mothers).

Results from study 2 provided support for the primary theoretical assertion of the expressed emotion literature, which posits that what parents say about their child and the parent-child relationship reflects (or guides) how they interact with their child on a day-to-day basis (Chambless, Bryan, Aiken, Steketee, & Hooley, 1999). Higher levels of SSOP at age 4 predicted higher levels of maternal insensitivity at age 6, and maternal insensitivity predicted higher levels of internalizing (but not attention/hyperactivity) problems at age 8. Test of the indirect effects indicated that SSOP exerted a significant indirect effect, via maternal insensitivity, on internalizing problems, but only a direct effect on attention/hyperactivity problems. Moreover, none of the conditional indirect effects were significantly different from zero, suggesting an absence of moderated mediation by child gender, maternal race/ethnicity, and single mother status. Taken together, these studies suggest that the FMSS SSOP construct may offer a cost-effective, culturally valid, and clinically valuable screening tool for the detection of pathological parental attitudes that may confer elevated risks for insensitive parenting practices and/or child adjustment difficulties.

Children of color can be misperceived as being much older than they are. However, anecdotal evidence of this phenomenon has outpaced its empirical investigation. Utilizing a longitudinal sample of 245 children (50% assigned female at birth, 88.8% non-white), the current study examined adults’ perceptions of children’s age (APCA) over time (i.e., ages 4, 6, 8, 10, and 12 years) to address questions regarding the shape of growth in APCA across childhood, predictors of growth, and how growth parameters (i.e., slope and intercept) may predict children’s educational (i.e., school grades and disciplinary actions) and socioemotional functioning (i.e., externalizing behaviors and social competence) at age 12. Four hundred adult raters provided over 15,000 ratings across more than 1,000 headshots of children from preschool to early adolescence. First, results indicated that growth in APCA proceeded linearly as perceived age ratings steadily increased with chronological age. There was also significant variation in patterns of linear change over time and estimated age values at age 12. Second, child sex assigned at birth and skin tone emerged as significant predictors of APCA over time. Adults’ overestimations of boys’ ages were greater than girls in early development (i.e., 4 to ~5.5 years), but girls were seen as older than boys in later development (i.e., ~ 5.5-12 years old). Children with darker skin tone ratings were perceived as older than those with lighter skin tones at all ages. Third, the APCA slope and intercept growth parameters did not significantly predict early adolescents’ educational and socioemotional functioning at age 12. Lastly, given the influence of pubertal status on the appearance of age, a supplementary, within-time analysis including the child’s pubertal status as a predictor of APCA at age 12 showed that the once-significant sex effect became non-significant, while the interaction between sex and skin tone became significant. Specifically, at age 12, adults rated girls with lighter skin tone as older than girls with darker skin tone, but skin tone was not related to APCA for boys. Together, these findings extend the evidentiary base for research on APCA while illuminating promising avenues for investigations of APCA in future research.

Dissociative processes in children and adolescents entail alterations in memory, identity, and perception that reflect and/or precipitate disconnects across biological, emotional, cognitive, and behavioral systems. Children of maltreatment survivors are at heightened risk for various forms of psychopathology, including problematic dissociation. However, given that most parents who were maltreated as children do not go on to maltreat their own children, these intergenerational effects may be mediated by more subtle parenting processes that thwart children’s emergent capacities for self-regulation and integration, such as caregivers’ reduced ability to reflect on child needs and behaviors and insensitive caregiving practices characterized by hostility, intrusiveness, and low support.

Drawing on five data waves with a diverse sample of 250 maternal caregiver-child dyads (50% female children, 88.8% nonwhite, 36.7% poverty), this dissertation examined intergenerational effects of caregiver history of maltreatment on children’s dissociation trajectories from early childhood through the transition to adolescence. Further, I evaluated explanatory simple and sequential mediation paths wherein I hypothesized that the severity of a caregiver’s history of maltreatment would predict higher levels of child dissociation at age 6 and slower declines in dissociation from ages 6 to 12 via lower caregiver reflectiveness and more insensitive caregiving. Further exploratory analyses tested these pathways separately for girls and boys.

A factor of curves (FOCUS) model provided a multivariate representation of children’s dissociation starting values and change over time as reported by teachers, caregivers, and examiners at ages 6, 8, 10 and 12. In line with study hypotheses, insensitive caregiving in the wake of maltreatment emerged as an initiating factor for children's dissociation. However, contrary to hypotheses, a history of maltreatment was positively associated with caregiver reflectiveness. In turn, reflectiveness predicted less insensitive caregiving and a resilient pattern wherein children’s dissociation scores began relatively low in early childhood and remained more stable across middle childhood as compared to children with less reflective and more insensitive caregivers. This dissertation has important implications for future research and applied efforts to mitigate intergenerational maltreatment effects and promote children’s resilience to the development of pathological dissociation.

The purpose of this investigation was to examine the development of self-regulation across preschool and early childhood, with an emphasis on identifying basic developmental processes, implications for adjustment, and contextual influences. Study 1 utilized a path analytic framework to assess longitudinal transactional relations between emotion and behavior regulation. Study 2 utilized a latent change analytic framework to assess the development of self-regulation and the influence of context. The use of these two complementary paradigms sought to clarify individual patterns of growth and stability, and to elucidate the mechanism behind potential cross domain effects.

Data were drawn from a diverse sample of 250 children who completed annual observational assessments of regulation during laboratory tasks at ages four, five, and six. Additional measures included parent report of adversity exposure, teacher report of social competence, internalizing problems, and externalizing problems, and laboratory-measured academic achievement.

Study 1 found modest stability of regulation across early childhood, and bidirectional prospective relations between emotion and behavior regulation. Better behavior regulation at age four predicted better emotion regulation at age six, while better emotion regulation at age five predicted better behavior regulation at age six. Each cross-domain longitudinal relation controlled for the stability of prior within-domain regulation and concurrent cross-domain relations. Behavior regulation evidenced direct relations with adjustment in multiple domains, with indirect effects of emotion regulation emerging through its effect on behavior regulation.

Study 2 found that emotion and behavior regulation evidenced different patterns of growth. Emotion regulation improved from age four to five, and remained relatively constant from age five to six. Furthermore, higher levels of cumulative adversity predicted a decrease in the slope of emotion regulation trajectories, such that children who experienced more adversity improved less in emotion regulation over time. There was no significant effect of adversity on initial levels of regulation. Although behavior regulation appeared to increase in a linear fashion, a latent change model could not be fit successfully, which precluded assessing the effect of adversity on behavior regulation, or interrelations between regulatory trajectories. Implications are for the development of context-sensitive interventions and increased specification of key domains of self-regulation in investigations.

Prior theory and research examining relations between trauma exposure and prosocial behavior (e.g., Rao et al., 2011; Vollhardt, 2009) has led some to suggest that altruism may be born of suffering (ABS; Staub & Vollhardt, 2008). However, extant research on these relations has focused on communally-experienced traumatic events, with little consideration accorded to individually-experienced events and the potential significance of contextual and developmental influences on these relations. This dissertation adopted a developmental lens of analysis to systematically evaluate extant research on childhood interpersonal trauma (i.e., maltreatment, bullying victimization, violence exposure) and prosocial behavior (i.e., behavior intended to benefit others; Batson & Powell, 2003) with particular emphasis on contextual factors that may modify these associations (e.g., sample age and geographic region, trauma type and assessment, prosocial type and assessment). In this comprehensive meta-analysis, I collected 35,383 articles from various sources and retained 24 studies as unique assessments of the impact of childhood interpersonal trauma on prosocial behavior. Meta-analytic procedures revealed a significant negative effect of childhood interpersonal trauma on prosocial behavior (r = -.16, 95%, CI [-.23, -.09], k = 26, I2 = 87.57,  = .03). However, this association was qualified by several moderators. Specifically, the negative association between childhood interpersonal trauma and prosocial behavior was stronger for samples in the U.S., child survivors of maltreatment as opposed to other trauma types, studies using administrative records to assess trauma exposure, and studies using teacher, peer, and examiner informants to assess the child’s prosocial behavior outside the family context as opposed to caregiver or child self reports. Younger samples evidenced marginally stronger and more negative associations between trauma exposure and prosocial behavior. Although I explored the potential moderating impact of trauma assessment scale (i.e., dichotomous versus continuous), prosocial behavior type (i.e., global versus specific assessments), and prosocial behavior measurement modality (i.e., observed versus reported), no other significant moderating effects emerged. Findings speak to the importance of considering the role of development as it may refine the ABS model. Further, I explicate several promising avenues for future research aimed at clarifying whether, when, and for whom altruism may follow from suffering.

Parental psychological control (PPC) is characterized by parental manipulative, inhibiting, and/or invalidating tactics to control youth’s socioemotional development, which are linked to negative youth adjustment outcomes. However, PPC effects on youth’s adjustment may vary across domains, and, consistent with transactional models of development, youth’s adjustment may reciprocally influence PPC expression. This investigation evaluated unidirectional and transactional relations between PPC and youth’s internalizing problems, externalizing problems, and parent-youth attachment security from late childhood (i.e., age 10) to early adolescence (i.e., age 12) as expressed in diverse groups based on youth’s sex assigned at birth, ethnicity-race, and neighborhood risk. Parent-youth dyads (N = 214; 49.5% female; 45.8% Latine; 36.5% in poverty) were drawn from a longitudinal study of child development. PPC was observationally rated by trained coders, youth’s internalizing and externalizing symptoms were reported by examiners, parent-youth attachment security was indicated by youth’s reports, and neighborhood risk was composited across multiple Census and FBI indicators. After controlling for family socioeconomic status, bivariate latent difference score models showed within-person variability in PPC and youth’s adjustment from ages 10 to 12, with PPC demonstrating growth, whereas youth’s externalizing problems and parent-youth attachment security, but not youth’s internalizing problems, declined. Transactional analyses supported a parent effect model wherein PPC predicted less change (i.e., more stability) in youth’s internalizing problems, and a youth effect model wherein youth-reported attachment security predicted increased PPC. Although changes in PPC were not related to changes in youth’s adjustment, there was a significant association between changes in youth’s externalizing problems and changes in PPC for families residing in higher-risk neighborhoods. In higher-risk neighborhoods, youth’s externalizing problems predicted less growth in observed PPC, but this relation was not significant for families in lower-risk neighborhoods. This study revealed directionally and contextually nuanced relations between PPC and youth’s multi-domain adjustment. The obtained pathways speak to the reciprocity of parent and youth effects such that PPC is not unilaterally deleterious for youth’s development, and youth’s adjustment may differentially influence PPC as a function of broader contextual risks. Therefore, applied efforts to enhance the parent-youth relationship in early adolescence must entail careful consideration of both transactional and contextual dynamics to promote positive parenting and youth’s adjustment.

A robust body of literature has documented positive associations between parents’ psychopathology and children’s adjustment difficulties. However, prior studies have focused on mothers’ psychopathology, using cross-sectional designs, with predominantly clinical and European American samples of young children. The current investigation addressed these gaps in the literature by evaluating concurrent and prospective associations of mothers’ and fathers’ depressive symptoms with youths’ internalizing and externalizing symptomatology from early adolescence to young adulthood in a community sample of European American and Mexican American families.

Drawing on five waves of longitudinal data from 392 families (52% female; Mage_W1 = 12.89, SD = .48; Mage_W5 = 21.95, SD = .77; 199 European American and 193 Mexican American families; 217 intact and 175 stepfather families), the first aim of this study was to document the unique contributions of both mothers’ and fathers’ depressive symptoms to youths’ symptomatology, as well as the transactional effects of youths' symptomatology on mothers’ and fathers’ depressive symptoms, from ages 12 to 22. The second aim of this study was to evaluate the explanatory role of theoretically-specified mediating mechanisms underlying these effects (i.e., co-parent’s depressive symptoms, parental acceptance, parental rejection, and perceived mattering to each parent). In both set of analyses, the obtained relations were tested across groups defined by family ethnicity (i.e., European American and Mexican American), family structure (i.e., intact and stepfather families), and youth gender.

After employing a novel technique to harmonize the data across waves, a series of trait and time-varying cross-lagged models evaluated between- and within-person differences in mothers’ and fathers’ depressive symptoms, youths’ symptomatology, and transactional relations among these three reporters across time. Overall, the findings suggested that both mothers’ and fathers’ between- and within-person differences in depressive symptoms were associated with youths’ internalizing and externalizing symptoms. However, whereas each parent’s depressive symptoms uniquely contributed to youths’ internalizing symptoms, only mothers’ depressive symptoms influenced youths’ externalizing symptoms. Although reciprocal effects of youths’ internalizing symptoms on parents’ depressive symptoms were not significant, youths’ externalizing symptoms predicted changes in mothers’ depressive symptoms over time. Moderation analyses revealed distinct transactional patterns by family ethnicity and child gender, but not by family structure.

A series of mediation analyses revealed that, although parents’ rejecting behavior and youths’ perceptions of mattering to each parent were related to youths' symptomatology, there were only two significant indirect effects, both of which accounted for the influence of fathers’ depressive symptoms on youths’ symptomatology. Specifically, fathers’ depressive symptoms contributed to higher rates of youths’ internalizing and externalizing symptoms via higher levels of mothers’ depressive symptoms and lower levels of fathers’ acceptance of the child, respectively. Moderation analyses revealed that the indirect effect from fathers’ depressive symptoms to youths’ internalizing symptoms via mothers’ depressive symptoms was significant among European American families, but not among Mexican American families. Together, these analyses revealed dynamic transactions among family members’ symptomatology that point to opportune times and targets for intervention and prevention efforts aimed at mitigating the negative impact of parents’ depressive symptoms on youths’ adjustment.

ABSTRACT OF THE DISSERTATION

Associations Between Narrative Features and Resilience in

Newly Emancipated Foster Youth

by

Izabela Kate Grey

Doctor of Philosophy, Graduate Program in Psychology

University of California, Riverside, June 2014

Dr. Tuppett M. Yates, Chairperson

This dissertation evaluated prospective relations between emancipated youth's narrative content (i.e., depicted themes of positive and negative affect) and coherence (i.e., organization) in a modified Five Minute Speech Sample (FMSS), which prompted youth to reflect on their experiences in foster care. Specific aims evaluated 1) unique and interactive associations of youth's narrative content and narrative coherence with psychosocial adjustment, 2) childhood correlates of youth's narrative content and coherence, and 3) mediating models wherein narrative features were expected to explain associations between childhood experiences and youth adjustment. Participants were 172 recently emancipated foster youth (66% female; Mage= 19.62 years, SD = 1.11; 34.1% Hispanic, 31.1% African American, 15.9% White European American, and 18.9% multiracial/other) who completed extensive face-to-face interviews as part of an ongoing study of youth's adaptation to aging out.

Both narrative content, particularly negative affect, and narrative coherence were related to age-salient adjustment. However, narrative content was more strongly related to emotional well-being (i.e., positive narrative content was associated with self-esteem and life satisfaction, and negative narrative content was related to internalizing and externalizing problems), whereas narrative coherence was more strongly related to relational outcomes (i.e., romantic and peer relationship quality, and peer attachment). Moderation analyses revealed unique associations of narrative content and coherence with varied outcomes, but the degree of narrative coherence did not significantly alter relations between narrative content and adjustment. Although narrative features were related to youth's childhood experiences (e.g., child maltreatment, child welfare experiences, and mentoring relationships), mediation analyses suggested only modest explanatory relations between these childhood experiences and later adjustment via narrative content and coherence with the strongest indirect paths evident through negative narrative content.

These findings support the utility of the FMSS as a brief narrative assessment tool. The results are discussed with particular attention to processes of narrative representation and meaning making as salient developmental influences in the wake of foster care, and in development broadly. Implications for future practice and policy aimed at supporting positive development among transition-aged foster youth are also discussed.

New policy efforts to support transition-aged foster youth (TAY) were implemented through the 2008 Federal Fostering Connections to Success Act in response to studies showing a) that foster youth who exit the child welfare system at 18 struggle in multiple domains during the transition to adulthood, and b) extended foster care supports through age 21 are associated with improved adaptation. However, there remains a need to clarify factors that influence policy uptake by TAY. Youths’ and workers’ satisfaction with the foster care service provision process may influence youths’ decision making regarding whether or not to opt out of extended care and the quality of social workers’ service provision, respectively. This study examined communication characteristics between TAY and their social workers as related to youth- and worker-reported satisfaction and relationship quality. Youth-worker dyads (N = 51; M=18.69 years, SD = .44; 56.9% female) were audio-recorded during a routine monthly meeting. Instrumental and affective communication features were rated in 5-minute samples of full speech, verbal content (i.e., transcripts), and non-verbal tone (i.e., content filtered speech). Principle Component Analyses (PCA) identified primary communication factors of TAY and social workers. Bivariate analyses revealed significant correlations among communication factors within and across channels, as well as with youths’ and workers’ reported satisfaction and perceived relationship quality. Youth expressed their true feelings about their workers and foster care in their full speech, and, relatedly, workers were most attuned to features of youth communication present in full speech. In contrast, workers expressed their true feelings about the youth in the content of their speech, and relatedly, youth were most attuned to features of worker communication present in the transcribed content of workers’ speech. At the dyadic level, planned contrasts evaluated the importance of communicative congruence between workers and youth. Workers were responsive to congruence in positivity, whereas youth were more responsive to congruence in negativity. These findings highlight the relevance of communication-based research and training for applied efforts to support TAY in and beyond the child welfare system.

Pretend play is a powerful mechanism of and context for children’s cognitive, affective, and social development. Research has documented relations between features of pretend play, such as the quality of expressed fantasy and the quantity of expressed affect, and varied developmental outcomes. Although there is theoretical and empirical support for the importance of both cognitive and affective facets of play in development, little is known about changes in the form and function of play across childhood. This investigation evaluated the structure and development of children’s pretend play in a diverse community sample of 250 children across three longitudinal play assessments at ages 4, 6, and 8. The data supported a two-factor play structure across childhood and in diverse groups with regard to child gender, race/ethnicity, and poverty status. Multilevel models documented the level and linear change patterns in cognitive and affective play factors over time. Although Hispanic race/ethnicity was associated with lower levels of the cognitive play factor at age 4, and poverty status was associated with lower levels of the cognitive play factor and the affective play factor at age 4, no predictors accounted for the significant variance that remained in the linear change parameter for the cognitive, but not the affective, factor. Together, these findings extend the literature on pretend play by evaluating competing models of the structure of play and documenting patterns of cognitive and affective development in children’s pretend play over time and in diverse groups. Implications are discussed to inform research on play and adjustment and guide practice and policy efforts that foster both.