Trauma and Maternal Caregivers as Risks for Executive Function Deficits Among Youth Who Have Sexually Harmed

Abstract

Despite robust research linking trauma events to the commission of sexual harm by youth, the majority of victims do not become victimizers, imploring us to study potential interceding influences. Some research indicates that intermediary factors like attachment characteristics, trauma symptomatology, and executive functioning may be critical in understanding sexual harm committed by youth. This study explored relationships between trauma events, trauma symptoms, and attachment characteristics, and their relationship to executive functioning in a sample of 196 youth who committed sexual harm. Results revealed bivariate associations between trauma events, symptomatology, and maternal attachment characteristics. Structural equation modeling revealed numerous direct and indirect effects on the path to deficits in executive functioning, and that sexual abuse, above and beyond other forms of traumatic events, contributed to executive functioning deficits via trauma symptoms. The results are contextualized and implications are discussed.

Keywords

adolescent sexual abusers attachment trauma executive functioning youth who sexually harm

Introduction

Researchers investigating social determinants of sexual harm committed by youth have long paid attention to trauma events, particularly sexual victimization, as the victim-to-victimizer hypothesis of sexual offending has been well established in the literature (e.g., Burton & Meezan, 2004). However, given that the vast majority of victims do not become victimizers (Widom, 2017), investigators have explored potential mediating factors explaining the relation between trauma and risk for the commission of sexual harm among samples of youth who have sexually harmed¹ (YSH). For example, trauma symptomatology (Boonmann et al., 2016) and executive functioning deficits (Adjorlolo & Egbenya, 2016; Burton et al., 2016; Veneziano et al., 2004; J. Yoder & Precht, 2019) are indicated risks for the commission of sexual harm and may require further study to better understand etiological models (Creeden, 2009; J. Yoder et al., 2019). The experience of early life trauma can be compounded by caregiver–child interactions that increase the likelihood of trauma symptomatology and executive functioning difficulties. Trauma experiences may interrupt attachment formation such that early caregiver–child relational experiences can create risk for trauma events and those same traumatic events may contribute to insecure attachment styles (Grady et al., 2017, 2018; J. R. Yoder et al., 2018). Therefore, it is critical to pinpoint how caregiver–child relational interactions, or attachment characteristics, co-occur with traumatic events to contribute to outcomes associated with the commission of sexual harm by youth. This article endeavors to contribute to this body of literature by exploring trauma events, caregiver attachment characteristics, trauma symptoms, and their relationship to executive functioning in a sample of YSH.

Trauma Experiences

There is now robust support for the victim-to-victimizer hypothesis of sexual harm committed by youth, as a disproportionate number of those who have committed sexual harm have histories of sexual victimization (Aebi et al., 2015; Pereda et al., 2009). There is also increasing evidence that YSH have additional abuse experiences concurrent with sexual victimization, compounding the effects of early adversity (Levenson et al., 2017; J. Yoder et al., 2017). Across studies, YSH show elevated rates of sexual abuse relative to youth who commit delinquent² crimes (Burton et al., 2011; Levenson et al., 2017; Seto & Lalumière, 2010; van Wijk et al., 2006), but there are also high rates of other forms of abuse. For example, among a large sample of male YSH, 13% had a history of sexual abuse (compared with 5% of delinquent youth), 20% had a history of physical abuse (15% of delinquent youth), and 11% had a history of physical neglect (6% of delinquent youth; Levenson et al., 2017): All differences were statistically significant. Furthermore, cumulative or multiple forms of trauma may be higher among YSH when compared with delinquent youth (Levenson et al., 2017; J. Yoder et al., 2020).

As researchers have conclusively found certain trauma experiences (i.e., sexual and physical abuse) to be discriminating risks, there is a need to focus on how these experiences co-occur with other childhood adversity to contribute to concomitant risks for the commission of sexual harm. For example, there are additional adversity risk factors that increase the likelihood for the commission of sexual harm, including family or household problems like mental health, substance use, witnessing violence, or criminality (Levenson & Grady, 2016; Wanklyn et al., 2012; J. Yoder & Caserta, 2016). Although such risks are often experienced at equivalently high rates between YSH and delinquent youth (Seto & Lalumière, 2010), we can greatly benefit from isolating a YSH sample and testing adversity covariation and subsequent effects on health and behavioral outcomes like trauma symptoms and executive functioning.

Caregiver Attachment Characteristics and Relational Experiences

Attachment theory has long been posited as an etiological explanation underlying the commission of sexual harm (Grady & Shields, 2018; Marshall & Marshall, 2016; McKillop et al., 2012). Etiological models drawing on attachment theory are increasingly becoming more integrated with the addition of multiple evidence-driven risks for youth sexual violence (Miner et al., 2010, 2016; J. R. Yoder et al., 2016). One specific model (see Grady et al., 2017) accounts for the relation between trauma and attachment, suggesting that trauma experiences may interrupt attachment formation, thereby creating risk for criminogenic needs that contribute to sexual harm. Researchers have begun to explore this theory by testing associations between trauma events, child–caregiver relational experiences or attachment characteristics, and resulting attachment styles. For example, Grady and colleagues (2018) found connections between physical abuse and avoidant and ambivalent attachment styles. Others have evaluated how trauma events and relational interactions may be experienced concurrently. In one study, adverse (harsh and inconsistent) maternal relational experiences were associated with traumatic events and insecure attachment; however, trauma experiences did not mediate the relationship between maternal relations and attachment styles, and most of the variance in insecure attachment was explained by maternal relational experiences (J. R. Yoder et al., 2018). This might suggest that trauma events may instead covary with relational experiences to contribute to risk outcomes. The same study noted that YSH reported greater adverse maternal caregiving and associated trauma events than delinquent youth.

This body of evidence suggests that early relational experiences, measured in the current study as youth-reported caregiver attachment characteristics (e.g., my mother cares about my feelings), can co-occur with traumatic events to contribute to risk for the commission of sexual harm among youth. This is consistent with neurological research suggesting that early relational interactions can covary with traumatic experiences to influence youth development (Ludy-Dobson & Perry, 2010; Perry, 2009; Perry et al., 1995). It is clear that further research is needed to determine how caregiver attachment characteristics co-occur with early traumatic events to contribute to risk pathways for the commission of sexual harm. Maternal attachment characteristics are specifically salient, as research on children and adolescents has overwhelmingly found maternal caregivers to be the primary sources of support for youth from childhood through early adulthood, thereby providing the foundation upon which self-regulation skills are learned. For example, in a study of high school and college-age people, Rosenthal and Kobak (2010) found that biological mothers were the most frequently identified primary attachment figures (53%), followed by friends (13%), romantic partners (12%), fathers (11%), and siblings (7%). In their discussion, they cited multiple studies with similar findings. Similarly, studies of youth with severe behavioral health problems have overwhelmingly found maternal caregivers to be identified as their primary caretakers (e.g., Gao et al., 2017).

Trauma Symptoms

Despite the high rates of trauma exposure among YSH, abuse and relational experiences are not necessarily deterministic; the vast majority of abused youth do not go on to repeat cycles of abuse (Widom, 2017). Given this, it is important to explore other possible indicators that influence the path between being a victim and becoming a victimizer among samples of YSH. One potential intermediating influence is the way in which trauma and attachment characteristics are internalized and mentally coded. There are known associations between traumatic events, attachment characteristics, and trauma symptomatology (King et al., 2011; Turner et al., 2012). For example, in one nationally representative sample of children, aged 2 to 9 years, emotional abuse experiences and parenting practices were the most prominent in associations with traumatic symptoms (Turner et al., 2012). Nevertheless, the experience of traumatic events does not necessarily denote the expression of trauma symptomatology; among one sample of juvenile justice youth, 93% reported at least one traumatic event, yet only 12% met diagnostic criteria for posttraumatic stress disorder (PTSD; Wilson et al., 2013). Therefore, assessing indicators of trauma symptomatology, relative to using diagnostic cutoffs, may be a more comprehensive way to assess internalization of traumatic events.

Interestingly, relative to delinquent youth, YSH have more internalizing difficulties, including isolation, anxiety, and low self-esteem (Seto & Lalumière, 2010; Van Wijk et al., 2006, 2007). As such, traumatic and adverse attachment characteristics may be internalized differently for some youth and expressed as trauma symptomatology or other internalization problems, such as helplessness and hopelessness (Brown & Grady, 2019), that create propensities for the commission of sexual harm. Indeed, researchers have found that, compared with delinquent youth, YSH have stronger relationships between childhood abuse and internalizing mental health problems (Boonmann et al., 2016).

Researchers are only beginning to identify the lasting neurological impacts of trauma events and associated symptomatology. Importantly, a meta-analysis revealed strong and consistent relationships between PTSD and executive functioning broadly (Polak et al., 2012), and researchers have found a relationship between general trauma symptoms and executive functioning in youth samples (Nooner & Leaberry, 2013). More notably, researchers who compared maltreated youth with PTSD to maltreated youth without PTSD found that those with PTSD showed greater deficits in neurological functioning (Kavanaugh & Holler, 2014). As such, trauma symptomatology may be a mediator in the relationships between adverse relational and trauma experiences and neurological deficits. Parsing out these associations among YSH may tell us more about the relative experience of these adversities.

Executive Functioning Deficits

Childhood adversity can affect developing neurological zones and may have specific impacts on the prefrontal cortex controlling executive functioning (Steinberg, 2008). This can result in myriad psychological and behavioral problems (Jaffee & Maikovich-Fong, 2013), including the commission of sexual and nonsexual harm (Nikulina & Widom, 2013). Researchers have found that traumatic events can significantly delay the formation of neural networks responsible for executive functioning required for monitoring, planning, organizing, regulating, and decision making (Steinberg, 2008; Zelazo et al., 2013). Early adverse relational experiences, when coupled with victimization experiences, may disrupt normative prefrontal neurological development via executive functioning and create vulnerabilities for behavioral problems (Perry et al., 1995). Nevertheless, the role of trauma symptomatology in this relationship has yet to be fully explored, particularly among YSH.

There is reason to believe that executive functioning deficits can be significant markers for the commission of sexual harm by youth, as research has found direct and indirect effects of poor executive functioning on delinquent and sexual offending behaviors (Burton et al., 2016; J. Yoder et al., 2019). Nevertheless, emerging research is somewhat inconclusive regarding whether or not this represents an elevated risk factor for this group. In one study, researchers compared YSH with a control group with similar backgrounds and found that YSH were slower to complete an executive functioning examination and had significantly more difficulty shifting their attention (Kelly et al., 2002). In another study, researchers found equivalent deficits in executive functioning when comparing youth who had only sexually harmed and youth who committed sexual and nonsexual crimes (Veneziano et al., 2004). Finally, in a recent study that compared youth who had committed nonsexual crimes only with youth who committed both sexual and nonsexual crimes, the latter demonstrated higher deficits in specific metacognition capacities, including working memory, planning and organizing, and organization of materials, while other features of executive function were not distinguishable between groups (J. Yoder & Precht, 2019). Consequently, the evidence is mixed regarding whether executive functioning deficits create a higher risk for sexual violence perpetration beyond other forms of violence. Therefore, because it is still an indicated risk for youth sexual violence, it is critical to discern risk antecedents in the formation of executive functioning deficits.

The relationship between trauma and executive functioning is just beginning to be explored among researchers studying YSH. One recent study found associations between physical, sexual, and emotional abuse and neglect on all features of executive functioning represented by both behavioral regulation index (BRI; that is, inhibition, working memory, behavioral shift, and emotional control) and metacognition index (MI; that is, cognitive shift, organization of material, task completion; J. Yoder & Precht, 2019). In another study, researchers found that executive functioning mediated the relationship between sexual abuse victimization and the commission of sexual harm (J. Yoder et al., 2020). In the same study, when controlling for physical abuse and domestic violence, sexual victimization contributed to deficits in metacognition. However, extant research on executive functioning capacities among YSH has not consistently included measures of relational experiences and therefore has not accounted for the potential covariation of trauma events with attachment characteristics in linking to such deficits. Given that the amount of time mothers spend with their children is inversely related to neuropsychological deficits (Beaver et al., 2010), understanding how trauma events and maternal attachment characteristics co-occur to contribute, directly and indirectly via trauma symptomatology, to executive functioning is an important next step in understanding risk for youth sexual violence.

Current Study

Despite a strong call for research on the neurodevelopmental impacts of childhood trauma and attachment in the commission of sexual harm by adults (Levenson et al., 2016; Levenson & Socia, 2016) and youth (Creeden, 2009; R. Dillard et al., 2019), there are still few studies that have collectively explored the co-occurring impact of trauma events and relational experiences on trauma symptoms and executive functioning among YSH. It is clear that there is more to learn about developmental pathways that create risk for YSH. This study tests covariation among trauma events and maternal attachment characteristics, and investigates how they contribute to trauma symptomatology and executive functioning in a sample of YSH. This study has three organizing research hypotheses:

Hypothesis 1 (H1): There is covariation between trauma events and maternal attachment characteristics.

Hypothesis 2 (H2a): There are direct relationships between trauma events, maternal attachment characteristics (exogenous variables), and trauma symptoms (mediator).

Hypothesis 2 (H2b): There are direct relationships between trauma symptoms (mediator) and executive functioning deficits (endogenous variable).

Hypothesis 3 (H3): Trauma symptoms mediate the relations between trauma events, maternal attachment characteristics, and executive functioning.

Method

Sample and Procedures

For this study, University Institutional Review Board approvals were granted, and the study was approved by the State Department of Youth Services. Youth adjudicated of a sexual crime who were residing in six residential and correctional facilities in a Midwestern state were recruited to participate in a battery of standardized and researcher-developed measures on risk and protective factors for committing sexual harm. The surveys included questions pertaining to attachment with caregivers, victimization experiences, executive function problems, and additional questions relative to youth offending characteristics. The content of the surveys and purpose of the study were explained to the youth by the research team. Youth were told that there were no rewards associated with participation and no punishment associated with nonparticipation. All youth who completed the surveys assented and voluntarily agreed to participate. Pencil and paper surveys were administered to youth in a small group format (8–10 youth per group) in classroom settings in each facility. For youth with reading complications (n = 4; 2%), the surveys were read aloud by trained graduate student research assistants. Direct care staff were present for supervision, safety, or if a youth were to become distressed; however, there were no reports of distress after taking this survey.

The sample consisted of 196 males who have been adjudicated of a sexual crime. The average age of the overall sample was 17 years (SD = 2.23 years), and youth completed ninth grade (SD = 1.65 grades). Overall, 95 youth (53.1%) identified as White, 52 youth (29.1%) identified as Black or African American, 10 youth (5.6%) as Latino or Hispanic, two youth (1.1%) identified as Asian, eight youth (4.5%) identified as Native American, and 12 youth (6.7%) identified as Other. Youth indicated they were raised in two-parent families (103; 58.8%), single-mom families (45; 25.7%), single-dad families (4; 2.3%), families with grandparents or other relatives (13; 7.5%), or foster families (10; 5.7%). This sample was used in two previously published studies (Burton et al., 2016; Morrell & Burton, 2014).

Measures

Trauma events: Childhood Trauma Questionnaire (CTQ)

Childhood trauma was measured using the CTQ (Bernstein et al., 1994). The CTQ is a 25-item standardized measure that asks youth to report how often (0 = never, 1 = rarely, 2 = sometimes, 3 = often, 4 = very often) they experienced abuse while growing up. Although there are five subscales, for the purpose of this study, only the physical and sexual abuse subscales were included in the analyses given that researchers have found these abuse indicators to be the most germane in association with sexual harm perpetrated by youth (Knight & Sims-Knight, 2004). These items are similar to dichotomous (0 = yes, 1 = no) items included in the original Adverse Childhood Experiences (ACEs) study (see Felitti, 2009; Felitti et al., 1998). Physical abuse (e.g., I was punished with a belt, board, cord, or some other hard object; people in my family hit me so hard it left me with bruises or marks; M = 1.23, SD = 1.17) had strong internal consistency (α = .891). Sexual abuse (e.g., someone molested me; someone tried to make me do sexual things; M = 0.89, SD = 1.25) also had strong internal consistency (α = .910). The items from the Physical and Sexual Abuse subscales were extracted to create latent variables and were modeled as exogenous factors.

Trauma events: Domestic Trauma Experiences (DTE)

Another form of developmental adversity was captured through a 12-item questionnaire, DTE, used in other studies of juvenile justice samples (Burton et al., 2011). This tool assessed youth’s experiences of indirect childhood adverse events in their homes of origin with example items such as “having a parent with an alcohol or drug problem” or “lots of moves and/or homelessness.” The response items were 0 = no, 1 = yes; the DTE scale (M = 0.42, SD = 0.33) demonstrated good internal consistency (α = .834). The items from the scale were extracted to create a latent variable modeled as an exogenous factor.

Trauma events: Witnessing Violence Survey (WVS)

Developmental adversity was also captured through a 16-item WVS developed by a residential program in a Midwestern state and adapted by researchers for use in other studies of juvenile justice samples (Burton et al., 2011). The tool assessed various forms of witnessing violence with example questions such as “I have seen a stranger being stabbed” or “I have seen a relative killed.” The response items were 0 = never, 1 = once or twice, 2 = a few times, 3 = many times; the WVS scale (M = 1.13, SD = 0.73) demonstrated strong internal consistency (α = .921). The items from the scale were extracted to create a latent variable modeled as an exogenous factor.

Maternal attachment characteristics

The Inventory of Parental and Peer Attachment (IPPA; Armsden & Greenberg, 1989) was used to assess characteristics of attachment between youth and their caregivers. The tool assesses the quality of the attachment with the caregiver relative to trust, communication, and alienation. The IPPA is comprised of 25 items in each of the three (Mother, Father, and Peer) subscales. For the purpose of this study, the composite score from only the maternal subscales was used given that the majority of youth in the sample indicated a maternal caregiver as their primary caregiver (n = 148, 75.5%). The response format is a 5-point scale ranging from 0 = almost never or never true to 4 = almost always or always true. Example items are as follows: “My mother expects too much from me”; “My mother accepts me as I am”; or “When we discuss things, my mother cares about my point of view.” The measure has demonstrated strong internal reliability and cross-validity (Armsden, 1986; Armsden & Greenberg, 1989; Buist et al., 2004; Lewis et al., 1987). The maternal subscales of trust and communication were combined due to high collinearity (<.9; M = 3.05, SD = 1.95) and demonstrated strong internal consistency (α = .969). The Maternal Alienation subscale (M = 1.13, SD = 0.73) also revealed good internal consistency (α = .755). The items from the maternal subscales were extracted to create separate exogenous latent variables.

Trauma symptoms: Trauma Symptom Checklist for Children (TSCC)

Trauma symptoms were measured with the TSCC (Briere, 1996), the most frequently used trauma symptom measure in North America for youth 8 to 16 years old (Wolpaw et al., 2005). It is a 54-item standardized measure for youth who have been exposed to traumatic events in clinical and nonclinical populations and juvenile justice settings. In the current study, the 10 items in the Clinical subscale for posttraumatic stress were used. The response format is a 4-point scale ranging from 0 = never to 3 = almost all of the time. The Posttraumatic Stress subscale (M = 0.87, SD = 0.72) demonstrated strong internal reliability in this sample (α = .913). The items were extracted to create a latent endogenous and mediating variable.

Executive functioning

Executive functioning was measured using the Behavior Rating Inventory of Executive Function–Self Report (BRIEF-SR; Guy et al., 2004). The BRIEF assesses youth’s executive functioning deficits and includes 80 items measured on a 3-point scale (0 = never, 1 = sometimes, 2 = often). The instrument was validated with two overarching dimensions of executive functioning: BRI (α = .938) and MI (α = .954). Some example items include the following: “I blurt things out,” “I have trouble changing from one activity to another,” or “I have trouble thinking of a different way to solve a problem.” Taken together, the items create the Global Executive Composite (GEC; M = 2.5, SD = 0.27) score with strong internal reliability (α = .972). The GEC, by averaging items, was used as an endogenous observed index in the analyses.

Data Analysis

This study conducted a structural equation model (SEM) that included observed indices and latent factors represented by items from validated and researcher-developed instruments. These items were extracted and tested to determine model fit. Specifically, this study used subscales of the CTQ, IPPA, TSCC, DTE, and WVS to create latent factors by running a confirmatory factor analysis (CFA) to validate the factors’ structures prior to running path analyses. The BRIEF was the only instrument modeled as an observed index, reflected by a total composite score, due to the number of observed items embedded (= 80) and the ensuing challenges with model convergence. Therefore, we modeled all other instruments as latent variables that allowed for good fit, model convergence, and sufficient power to detect model specificity and target coefficient effects (Wang & Rhemtulla, 2020). To achieve good model fit, some minor item deletions and modeling decisions were made that will be discussed further.

CFA

SEM using Mplus (Version 8; Muthén & Muthén, 2018) was used to test the research questions. SEM is a useful analytical approach to determine measurement fit and modeling relationships between theoretically driven constructs of interest simultaneously (Bowen & Guo, 2012). CFA was used to test the measurement model and confirm that the items representing physical and sexual abuse were good fitting items. CFA fit statistics (comparative fit index [CFI], Tucker–Lewis index [TLI], root mean square error of approximation [RMSEA], and chi-square) were used to assess model fit (Bowen & Guo, 2012). The cutoffs for each fit index are presented in Table 1 in relation to the model fit for each model in this study.

Table 1.

Measurement Model Fit Indices and Factor Loadings.

Model 1. Maternal alienation fit indices: χ²(1458) = 1,824, p < .001, CFI = .954, TLI = .952, RMSEA = .037 (CI = [.031, .042])			Model 2. Maternal trust and communication fit indices: χ²(1991) = 2,401, p < .001, CFI = .968, TLI = .966, RMSEA = .033 (CI = [.028, .038])
Model fit cutoff	RMSEA (Hu & Bentler, 1999; Iacobucci, 2010)	RMSEA 95% CI (Hu & Bentler, 1999; Iacobucci, 2010)	Chi-square (function of df and number of parameters, not robust to sample size; T. A. Brown, 2006; Hu & Bentler, 1999; Iacobucci, 2010)		CFI/TLI (T. A. Brown, 2006; Hu & Bentler, 1999; Iacobucci, 2010)
Model fit cutoff	<.06	(0–.06)	>.05		≥.95
Latent construct	Observed items	Factor loadings	Latent construct	Observed items	Factor loadings
Sexual abuse			Sexual abuse
	CTQ 24	.915***		CTQ 24	.912***
	CTQ 25	.900***		CTQ 25	.909***
	CTQ 30	.964***		CTQ 30	.964***
	CTQ 31	.920***		CTQ 31	.918***
Physical abuse			Physical abuse
	CTQ 4	.950***		CTQ 4	.943***
	CTQ 14	.958***		CTQ 14	.966***
	CTQ 17	.698***		CTQ 17	.710***
	CTQ 20	.859***		CTQ 20	.850***
Maternal alienation			Maternal trust/communication
	IP8	.583***		IP1	.877***
	IP10	.563***		IP2	.836***
	IP11	.565***		IP4	.881***
	IP17	.891***		IP12	.862***
	IP18	.855***		IP13	.760***
	IP23	.689***		IP20	.903***
Trauma symptoms				IP21	.905***
	DTS01	.759***		IP22	.875***
	DTS02	.793***		IP5	.798***
	DTS03	.868***		IP7	.852***
	DTS10	.764***		IP15	.849***
	DTS12	.895***		IP16	.808***
	DTS32	.722***		IP19	.896***
	DTS33	.830***		IP24	.833***
	DTS35	.829***		IP25	.871***
	DTS43	.835***	Trauma symptoms
	DTS50	.691 ***		DTS01	.766***
Domestic trauma				DTS02	.791***
	DM4A	.698***		DTS03	.868***
	DM4B	.637***		DTS10	.757***
	DM4C	.504***		DTS12	.896***
	DM4D	.545***		DTS32	.720***
	DM4E	.540***		DTS33	.831***
	DM4F	.871***		DTS35	.831***
	DM4G	.951***		DTS43	.837***
	DM4H	.798***		DTS50	.680 ***
	DM4I	.705***	Domestic trauma
	DM4J	.904***		DM4A	.701***
	DM4K	.674***		DM4B	.638***
	DM4L	.563***		DM4C	.465***
	DM4M	.575***		DM4D	.502***
Witnessing violence				DM4E	.560***
	DM11A	.736***		DM4F	.889***
	DM11B	.896***		DM4G	.962***
	DM11C	.870***		DM4H	.781***
	DM11E	.954***		DM4I	.724***
	DM11F	.863***		DM4J	.902***
	DM11G	.707***		DM4K	.658***
	DM11H	.906***		DM4L	.538***
	DM11I	.900***		DM4M	.590***
	DM11J	.825***	Witnessing violence
	DM11K	.616***		DM11A	.745***
	DM11L	.784***		DM11B	.891***
	DM11M	.704***		DM11C	.867***
	DM11N	.496***		DM11E	.956***
	DM11O	.809***		DM11F	.863***
	DM11P	.574***		DM11G	.704***
				DM11H	.905***
				DM11I	.900***
				DM11J	.836***
				DM11K	.615***
				DM11L	.771***
				DM11M	.693***
				DM11N	.530***
				DM11O	.808***
				DM11P	.574***

Note. CFI = comparative fit index, TLI = Tucker–Lewis index, RMSEA = root mean square error of approximation; CI = confidence interval.

***

p < .001. All are reported as two-tailed values.

The CFA tests initially showed a poor fitting model. Using modification indices and residual correlation matrices (>.2), three poor fitting items were systematically reviewed. These were considered for deletion if it was theoretically appropriate and the modification indices reflected a problematic pattern (Bowen & Guo, 2012; Byrne et al., 1989). Then, a systematic method was taken to delete one item at a time, determine model fit, and retaining one item while deleting another one (Bowen & Guo, 2012). This process resulted in two total item deletions for the CTQ items, including “I believe I was sexually abused” and “I got hit or beaten,” and one item deletion for the WVS items, including “I have been beat up”; this item fails theoretically to align with the overarching focus of the instrument on witnessing violence.

Notably, when testing the measurement portion of the model via the CFA, a recurring model fit problem emerged with modification indices indicating challenges with the maternal alienation and trust and communication variables and indicators that were simultaneously modeled. The items were cross-loading on the contrasting latent variables (i.e., maternal alienation indicators were cross-loading on the trust and communication latent variable). Theory was used to guide next steps in modeling with first attempting to combine all items as one latent variable reflecting maternal attachment. This resulted in problems with maternal alienation items that required deletion leading to one alienation item and fit indices that never quite met the cutoff thresholds. Given that maternal alienation and trust/communication are distinct constructs, the decision was made to model the maternal alienation latent variable separately from the maternal trust and communication latent variables, leading to two separate SEM models.

Structural model testing

A general SEM was used to analyze the structural portions of the models. First, the covariation between the exogenous variables was tested. The exogenous variables were modeled this way given the theoretically driven hypotheses and the challenge of parsing out the time order sequence of maternal attachment characteristics and trauma events in the prefaces for these question sets in the survey. Youth were asked to think of maternal attachment characteristics and trauma events as experiences “while they were growing up”; it is difficult to ascertain whether maternal attachment precedes or follows traumatic events. Trauma symptoms and executive functioning were prefaced as events that occurred “over the past 6 months” and “over the past 3 months,” respectively. Executive functioning was regressed on trauma symptoms; trauma events, including sexual abuse, physical abuse, domestic trauma events, and witnessing violence; and maternal alienation and trust and communication, separately. We also included racial identity (Black and White, separately dummy coded) and age as control variables, modeling them on both endogenous variables. To determine whether trauma symptoms mediated the relation between attachment constructs and trauma events and executive functioning, the indirect effects were tested after initial results were run.

The weighted least square (WLS) estimator was used; the WLS estimator considers data to be ordinal (Flora & Curran, 2004) and addresses nonnormally distributed data that characterize most ordinal-scaled data (Flora & Curran, 2004) by not making distributional assumptions like the default maximum likelihood (ML) estimator (Bollen, 1989). Mplus uses a pairwise present technique to handle missing ordinal nonnormally distributed data by retaining all cases with at least one data point, similar to full information maximum likelihood (FIML).

Power

SEM models often require power analyses to detect both model specificity and power to detect effects (Wang & Rhemtulla, 2020); the former relies heavily on degrees of freedom and the latter relies on a complex amalgamation of number of observed items and item reliability in the measurement model and the number and value of estimated parameters (Wang & Rhemtulla, 2020). With small sample sizes, like the sample in this study, power can be enhanced by a larger number of information in the measurement model (a greater number of reliable indicators). To assess power for model specificity, the lowest number of degrees of freedom from the models (maternal alienation) was assessed. This model yielded a total df = 1824. As 80% power is needed to detect statistical significance based on α = .05, a minimum sample size of n = 200 is required with df = 100. With a sample size of 196 in this study, and minimum df = 1824, there was 100% power to detect statistical significance based on α = .05. For this conservative test of not close fit, ε₍₎ = 0.05 and ε_a = 0.01, where ε₍₎ is the null value of RMSEA and ε_a is the alternative value of RMSEA (MacCallum et al., 1996).

To assess power to detect effects, we ran power simulations using pwrSEM (Wang & Rhemtulla, 2020). We specified the model by precisely including the number of indicators, latent constructs, observed indices, paths, mediation, and factor loadings. We set parameter values to estimate power for the model results that were achieved. Then, we estimated the power using 100 simulations and α = .05. Power calculations showed that the alienation model (the model with the smallest amount of information) had sufficient power to detect effects: Y ~ X = .85, Y ~ M = 1.00, M ~ X = .98, and X ~ X = 1.00.

Results

CFA

The final models testing the direct relationships and mediation demonstrated good fit. The factor loadings linking the CTQ items to physical and sexual abuse latent variables were all strong (>.45; Costello & Osborne, 2005; Tabachnick & Fidell, 2001). The components of the measurement model (latent factors with corresponding observed items and their loadings for each model) are presented Table 1.

H1: Covariation

The covariation results revealed that the only nonstatistically significant association was between sexual abuse and witnessing violence in both the maternal alienation and trust and communication models. The significant results, presented as standardized values, are illustrated in Figures 1 and 2, and all standardized results are presented in Table 2.

Figure 1.

Maternal alienation model: Trauma symptoms as a mediator between maternal alienation, sexual abuse, and executive function.

Figure 2.

Maternal trust and communication model: Trauma symptoms as a mediator between sexual abuse and executive function.

Table 2.

Covariation and Path Results.

Model 1: Covariation	ϕ (SE)Sexual abuse	ϕ (SE)Maternal alienation	ϕ (SE)Domestic trauma experiences	ϕ (SE)Witnessing violence
Physical abuse	.722 (.05)***	.495 (.07)***	.745 (.05)***	.432 (.07)***
Sexual abuse	—	.322 (.09)**	.588 (.08)***	.137 (.09)
Maternal alienation	—	—	.464 (.09)***	.329 (.09)***
Domestic trauma experiences	—	—	—	.273 (.08)**
Model 2: Covariation	ϕ (SE)Sexual abuse	ϕ (SE)Maternal trust and communication	ϕ (SE)Domestic trauma experiences	ϕ (SE)Witnessing violence
Physical abuse	.721 (.05)***	−.320 (.07)***	.744 (.05)***	.433 (.07)***
Sexual abuse	—	−.211 (.09)*	.587 (.08)***	.137 (.09)
Maternal trust and communication	—	—	–.323 (.08)***	−.276 (.08)***
Domestic trauma experiences	—	—	—	.278 (.08)**
Model 1: Maternal alienation path results	Trauma symptoms		Executive functioning
Model 1: Maternal alienation path results	γ/β (SE)	p	γ/β (SE)	p
Direct paths
Physical abuse	−.130 (.21)	.541	−.272 (.24)	.252
Sexual abuse	.362 (.02)	.029*	−.043 (.02)	.830
Maternal alienation	.254 (.11)	.022*	.051 (.12)	.654
Domestic trauma experiences	.003 (.14)	.985	.189 (.18)	.304
Witnessing violence	.063 (.14)	.518	.233 (.10)	.022*
Trauma symptoms	—	—	.372 (.09)	<.001***
Age	.040 (.09)	.665	−.100 (.10)	.324
Racial identity: Black	−.070 (.11)	.552	.065 (.13)	.603
Racial identity: White	−.016 (.11)	.886	.152 (.12)	.194
Indirect paths
Maternal alienation	—	—	.095 (.05)	.05*
Sexual abuse	—	—	.135 (.06)	.032*
R ²	.191 (.06)***		.232 (.08)***
Model 2: Maternal trust and communication path results	Trauma symptoms		Executive functioning
Model 2: Maternal trust and communication path results	γ/β (SE)	p	γ/β (SE)	p
Direct paths
Physical abuse	−.055 (.02)	.832	−.266 (.23)	.248
Sexual abuse	.352 (.18)	.05*	.047 (.19)	.811
Maternal trust and communication	−.020 (.09)	.832	−.079 (.09)	.388
Domestic trauma experiences	.052 (.14)	.715	.184 (.18)	.313
Witnessing violence	.096 (.10)	.334	.226 (.10)	.029*
Trauma symptoms	—	—	.384 (.09)	<.001***
Age	.037 (.08)	.665	.092 (.036)	.012*
Racial identity: Black	−.063 (.11)	.570	.086 (.020)	.000*
Racial identity: White	−.007 (.11)	946	.242 (.022)	.000*
Indirect paths
Sexual abuse			.135 (.07)	.048*
R ²	.145 (.06)***		.223 (.08)***

Note. γ represents coefficient estimates between exogenous and endogenous variables, and β represents coefficient estimates between endogenous variables.

p < .05. **p < .01. ***p < .001. All are reported as two-tailed values.

H2: Direct Effects

Results from the maternal alienation model revealed that maternal alienation and sexual abuse had statistically significant direct effects on trauma symptoms; 19.1% of the variance in trauma symptoms was explained by the exogenous variables. Results also revealed that trauma symptoms and witnessing violence had statistically significant direct effects on executive functioning deficits; 23.2% of the variance in executive functioning was explained by the model. There were no effects of race or age on the trauma symptoms or executive functioning. Results from the maternal trust and communication model revealed that sexual abuse had statistically significant direct effects on trauma symptoms; 14.5% of the variance in trauma symptoms was explained by the exogenous variables. Results also revealed that trauma symptoms and witnessing violence had statistically significant direct effects on executive functioning deficits; 22.3% of the variance in executive functioning was explained by the model. There were no effects of race or age on the trauma symptoms or executive functioning. For all path coefficients, standard errors, and p values, refer to Table 2. For a visual presentation of the results, refer to Figures 1 and 2.

H3: Indirect Effects

For the maternal alienation model, indirect effect results revealed that there was a total indirect effect of trauma symptoms on the relation from sexual abuse to executive functioning. There was also an indirect effect of trauma symptoms on the relation from maternal alienation to executive functioning. For the maternal trust and communication model, indirect effect results revealed that there was a total indirect effect of trauma symptoms on the relation from sexual abuse to executive functioning.

Discussion

Given the strong connection between trauma and relational experiences among YSH (Creeden, 2009; Grady et al., 2017), it was unsurprising that we found significant relationships between nearly all trauma events in the model and maternal attachment characteristics in this study. However, direct effects were not observed between all of the trauma events and trauma symptoms, underscoring how trauma events may be uniquely processed by youth. YSH have histories marked with significant sexual trauma (Seto & Lalumière, 2010), so this finding in our study was not surprising. However, it was unexpected to learn that physical abuse experiences did not stand out in relation to trauma symptoms given that physical abuse has been linked to attachment and interpersonal difficulties in this population (Grady et al., 2018). Physical abuse, while traumatizing for many who experience it, might be less complex for youth to process, whereas the experience of sexual abuse for children might feel more multifaceted given the accompanying secrecy, confusion, self-blame, and sometimes positive attributions associated with it (Blanco et al., 2015; R. L. Dillard, 2020). Indeed, problems in maintaining interpersonal relationships have been associated with distorted cognitive schemas among YSH who were themselves sexual abuse victims (Burton & Meezan, 2004).

Results revealed that sexual abuse, above and beyond physical abuse and other forms of traumatic events, contributed to executive functioning deficits via trauma symptoms. This might shed light on the work of researchers who found that sexual abuse has more powerful impacts than physical abuse on internalization and internal processes that affect regulation capacity; sexual abuse experiences, beyond other forms of abuse, were found to have significant associations with anger difficulties, suicidal ideation, and thought disturbance, with greater effects for YSH (Boonmann et al., 2016). Others have found sexual abuse to be a key indicator in all forms of metacognition deficits representing executive functioning (J. Yoder et al., 2019). In addition to the distinction between physical and sexual abuse noted above, it is possible that the relational nature of the experience of sexual abuse, violating normative social and/or familial boundaries, is more disorienting than physical abuse and therefore distinctively contributes to youth experiencing trauma symptoms and subsequent executive functioning impairments.

One interesting finding from this study was that witnessing violence was directly linked to executive functioning. This suggests that YSH who hear or see violence take place in their households or communities have more difficulties with planning, organizing, understanding thought patterns, or transitioning between concepts. Although this is one of the first studies of YSH to find a direct effect of witnessing violence on executive functioning, it is consistent with extant research that suggests exposure to indirect forms of violence creates risk for impairments in executive functioning more broadly among youth (Berkowitz, 2003; Menon et al., 2018). It is possible that having a consistent and predictably safe home environment in childhood and adolescence contributes to neurodevelopment reflective of high-level executive functioning regardless of exposure to interpersonal trauma.

In addition, there was an indirect effect of maternal alienation on executive functioning, as this was explained by trauma symptoms with covariation between maternal alienation and sexual abuse. It is possible that, for some of these youth, it was not only the experience of sexual abuse but the experience of sexual abuse concurrent with feeling alienated from a maternal caregiver that increased the likelihood of psychological trauma symptoms. This is consistent with the finding that youth with compounding forms of adversity can have greater trauma symptoms relative to youth experiencing chronic victimization of the same type (Turner et al., 2010). It is also possible that maternal alienation is internalized and processed similarly to experiences of sexual abuse, as feelings of alienation by maternal caregivers has been found to have a significant impact on cognitive processing (Natarajan et al., 2011). In a study of youth exposed to community violence, researchers found that isolation and estrangement (i.e. feelings of alienation) serve as partial mediators in the path to PTSD symptomatology (O’Donnell et al., 2006).

The covariation between maternal alienation and sexual abuse and their respective direct and indirect association with trauma symptoms and executive functioning emphasizes that we can begin to parse out the temporal ordering of trauma and attachment in studies of traumatized youth. This study is aligned with extant research indicating that trauma interferes with neural pathways that develop higher order executive functioning (De Bellis, 2001), and begins to identify the role that problems in maternal attachment characteristics may have had in the development of executive functioning. As trauma and attachment problems co-occur, it makes sense that a youth who have been sexually abused may also report problems in attachment with caregivers. Although we do not have data about the specific perpetrators of sexual abuse in this study, the findings in this study still speak to the multiple and co-occurring developmental risks that are often present in the lives of these youth (Burton et al., 2011).

Maternal caregiver characteristics, specifically alienation, had a direct effect on trauma symptoms. This finding supports the work of researchers who found that children with trauma histories have insecure attachments that lead to struggles in interpersonal relationships more generally (Baer & Martinez, 2006). More specifically, these results suggest how integrated attachment theories of sexual violence can be further refined to consider the temporal sequencing or the covariation of risks in the development of sexually harmful behaviors. For instance, while traumatic events may interrupt attachment formation (see Grady et al., 2017), theoretical models can also account for the co-occurrence and cumulative impact of early adversities, including direct and indirect forms of abuse and relational experiences. Alternatively, longitudinal research is needed to test integrated theories to better account for the time-ordered sequence of risks that this study could not capture.

Although it is possible that maternal relationships are unique in contributing to the outcomes examined in this study, we suspect that this is a “primary caregiver” effect, which is most often reported as maternal (Gao et al., 2017) but not limited to maternal figures. Youth-reported maternal alienation may characterize more adversity and contribute to greater challenges with traumatic internalization and subsequent abilities in maintaining attention and focus necessary for emotional reciprocity and other higher order thinking. As such, consistent with attachment theory more broadly, it is possible that alienation from one’s primary caregiver is experienced as trauma and carries enduring consequences in neurodevelopment that result in deficits in executive functioning (Antonucci et al., 2018).

The connection between trauma symptoms and executive functioning was not surprising and corresponds with other research suggesting there are clear neurological implications resulting from trauma events (Perry, 2009; Perry et al., 1995), particularly in samples of YSH (Burton et al., 2016; J. Yoder et al., 2019). It is possible that trauma symptom expression represents traumatic stress that includes the residual impact of traumatic events that were responsible for activating the body’s fight or flight response, releasing the stress hormone cortisol (De Bellis, 2001; De Bellis et al., 2011). This biological response to traumatic stress can contribute to impairment in executive functioning (Augusti & Melinder, 2013). In other words, traumatic stress and trauma symptoms may be similar constructs (Kerig & Becker, 2010) responsible for the same impact on executive functioning. This suggests a sequential linear connection between adversity, internalization of that experience, and subsequent neurological impairments.

It is also possible that the relationship between trauma symptoms and executive functioning exposed the shared variation of trauma symptomatology and deficits in executive functioning as measured and understood in this project. In short, PTSD symptomatology is operationalized as the manifestation of functional difficulties following exposure to a traumatic event, including two or more of the following: angry or irritable outbursts, reckless behavior, hypervigilance, exaggerated startle response, problems with concentration, and sleep disturbance (Diagnostic and Statistical Manual of Mental Disorders [5th ed.; DSM-5]; American Psychiatric Association, 2013). Executive functioning is understood as a complex system of behaviors that involve the conscious control of thought and action. Broadly, these include response inhibition, reasoning, processing speed, and the ability to shift working memory for task completion (Salthouse & Davis, 2006). In other words, there is great overlap in symptomatology associated with PTSD and deficits in executive functioning. Therefore, when executive functioning deficits are measured among those who have experienced trauma, we cannot know whether we are identifying deficits in executive functioning that exist as a consequence of trauma or a consequence of something unrelated. As with traumatic events, deficits in these domains are necessary but not sufficient to receive a diagnosis of PTSD.

Limitations

The results of this study should be examined in the context of its limitations. First, at 196, our sample size was borderline for meeting the minimum suggested sample size of 200 for SEM (Wang & Rhemtulla, 2020). Therefore, a larger sample size would have increased the power. As explained in the “Method” section, we addressed this limitation by increasing reliable indicators, and subsequent power analyses gave us confidence in the findings despite this limitation. Second, although inferences of temporality among the variables studied can be made based on typical developmental sequence and the clinical experiences of the authors, the cross-sectional design of this study does not allow for making causal predictions. Therefore, research on YSH should be prospective and longitudinal to make causal determinations.

We used the victim-to-victimizer hypothesis as a framework rather than a rationale for generating and testing our hypotheses. A stronger test of the victim-to-victimizer hypothesis of the commission of sexual harm by youth would require a comparison group. Furthermore, the results of this study are not generalizable to all YSH in a few respects: This was a male sample and therefore is not generalizable to female YSH. Also, residentially based youth are, by definition, those with more severe problem behaviors than community-based samples of youth. Therefore, results are not necessarily generalizable to community-based samples. We also caution against generalizing to other residential samples given the purposive nonprobability sampling employed in this study.

Measures of executive functioning should be triangulated with additional measures for precision. Other studies have measured executive function with specific neuropsychological tests (see Veneziano et al., 2004) rather than a self-report tool used in this study. The use of multiple measures, including neurological scans, may be necessary for accuracy in assessing executive function deficits. The measurement of executive functioning can be refined to also consider its dynamic nature; given that this article examined the role that trauma plays in executive functioning deficits, further research needs to consider whether variation in executive functioning over time is attributed to trauma experiences over time. Longitudinal research in this area is needed. Furthermore, it is possible that some deficits in executive functioning observed were unrelated to trauma measured in this study, but instead due to another unmeasured variable, such as a traumatic brain injury (Sesma et al., 2008). In addition, the instrument we used to measure attachment, the IPPA, does an adequate job of capturing a youth’s feelings toward his parents but does not explain the reciprocal experience of attachment itself. The implications of this limitation on future research are described in the following section. Finally, these data were self-reported, which may lead to some social desirability bias in this sample where youth are reporting what they believe researchers want to hear.

Implications

One implication of this study is to serve as a reminder to practitioners to evaluate symptoms of trauma thoroughly before choosing a treatment recommendation. As mental health clinicians who focus on the assessment and treatment of YSH, we agree with the recent findings of Lateef and Jenney (2020), who concluded that there is a great need for trauma-informed care in promoting the wellness of these youth more broadly. This study underscores the impact of trauma among YSH housed in institutional settings. Treatments such as trauma-focused cognitive-behavioral therapy (TF-CBT) should be modified and evaluated for YSH to comprehensively address trauma as the most germane risk factor for sexual violence. Efficacious treatments like TF-CBT can support reductions in trauma-related symptoms, augment relational experiences between youth and caregivers, and lower risk for ongoing sexual violence. Similarly, this study shows how trauma events and symptomatology are associated with executive functioning impairments, suggesting the need for advanced screening to detect these deficits among YSH. Comprehensive screening of regulation and metacognition might support targeted treatment that could contribute to mitigating the risk of committing sexual harm by some youth.

An additional implication of this study is the need for researchers to more deeply explore the causal mechanisms in the victim-to-victimizer relationship by testing relations between risk pathways among YSH. For example, witnessing domestic and community violence, which is highly correlated to poverty (Graif & Matthews, 2017; Hernandez et al., 2014), was found in this study to directly affect executive functioning. Understanding the potential role of poverty in all risks for the commission of sexual harm by youth, including trauma and executive functioning, might be an important step in uncovering potential sexual abuse prevention opportunities.

The instrument used to measure caregiver attachment characteristics assesses youth-reported relational experiences instead of styles of attachment. These are correlated but distinct constructs. The IPPA does an adequate job of capturing youth-reported interactions with caregivers by way of experiences of alienation, trust, and/or communication, but it does not necessarily explain the reciprocal experience of interaction patterns, including the mutual experiences of support, love, belonging, and acceptance. Furthermore, it does not assess attachment styles or the formal classifications resulting from these relational experiences. Therefore, more research needs to be done to understand the relationship between parent–youth interaction characteristics and attachment styles. Perhaps a closer look at the interactions between youth and caregivers and the resultant attachment formation (J. Yoder et al., 2020) would fill this research gap.

Footnotes

Acknowledgements

The authors express gratitude to David L. Burton for his mentorship and assistance in this project. They also thank the thoughtful feedback of reviewers at various stages of this study.

Authors’ Note

The authors take responsibility for the integrity of the data and the accuracy of the data analyses, and have made every effort to avoid inflating statistically significant results.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Adam Brown

Notes

References

Adjorlolo

Egbenya

D. L.

(2016). Executive functioning profiles of adult and juvenile male sexual offenders: A systematic review. The Journal of Forensic Psychiatry & Psychology, 27(3), 349–375.

Aebi

Landolt

M. A.

Mueller-Pfeiffer

Schnyder

Maier

Mohler-Kuo

(2015). Testing the “sexually abused-abuser hypothesis” in adolescents: A population-based study. Archives of Sexual Behavior, 44, 2189–2199.

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: American Psychiatric Publishing.

Antonucci

L. A.

Taurisano

Coppola

Cassibba

(2018). Attachment style: The neurobiological substrate, interaction with genetics and role in neurodevelopmental disorders risk pathways. Neuroscience & Biobehavioral Reviews, 95, 515–527.

Armsden

(1986). Coping strategies and quality of parent and peer attachment in late adolescence [Meeting session]. First Biennial Meeting of the Society for Research on Adolescence, Madison, WI, United States.

Armsden

G. C.

Greenberg

M. T.

(1989). Inventory of parent and peer attachment (IPPA). University of Washington.

Augusti

E. M.

Melinder

(2013). Maltreatment is associated with specific impairments in executive functions: A pilot study. Journal of Traumatic Stress, 26(6), 780–783. https://doi.org/10.1002/jts.21860

Baer

J. C.

Martinez

C. D.

(2006). Child maltreatment and insecure attachment: A meta-analysis. Journal of Reproductive and Infant Psychology, 24(3), 187–197.

Beaver

K. M.

Vaughn

M. G.

Delisi

Higgins

G. E.

(2010). The biosocial correlates of neuropsychological deficits: Results from the national longitudinal study of adolescent health. International Journal of Offender Therapy and Comparative Criminology, 54(6), 878–894. https://doi.org/10.1177/0306624x09345993

10.

Berkowitz

S. J.

(2003). Children exposed to community violence: The rationale for early intervention. Clinical Child and Family Psychology Review, 6(4), 293–302.

11.

Bernstein

D. P.

Fink

Handelsman

Foote

Lovejoy

Wenzel

… Ruggiero

(1994). Initial reliability and validity of a new retrospective measure of child abuse and neglect. The American Journal of Psychiatry, 151, 1132–1136.

12.

Blanco

Nydegger

L. A.

Camariollo

Trinidad

D. R.

Schramm

Ames

S. L.

(2015). Neurological changes in brain structure and functions among individuals with a history of childhood sexual abuse: A review. Neuroscience and Biobehavioral Reviews, 57, 63–69.

13.

Bollen

K. A.

(1989). A new incremental fit index for general structural equation models. Sociological Methods & Research, 17(3), 303–316.

14.

Boonmann

Grisso

Guy

L. S.

Colins

O. F.

Mulder

E. A.

Vahl

…Vermeiren

R. R. J. M

. (2016). Childhood traumatic experiences and mental health problems in sexually offending and non-sexually offending juveniles. Child and Adolescent Psychiatry and Mental Health, 10(1), 45. https://doi.org/10.1186/s13034-016-0127-2

15.

Bowen

N. K.

Guo

(2012). Structural equation modeling. Oxford University Press.

16.

Briere

(1996). Trauma Symptom Checklist for Children (TSCC): Professional manual. Psychological Assessment Resources.

17.

Brown

Grady

M. D.

(2019). Helplessness and hopelessness in adolescents who commit sexual and nonsexual crimes. Victims & Offenders, 14(1), 15–31.

18.

Brown

T. A.

(2006). Confirmatory factor analysis for applied research. Guilford Press.

19.

Buist

K. L.

Deković

Meeus

W. H.

Aken

M. A. V.

(2004). Attachment in Adolescence:. Journal of Adolescent Research, 19(6), 826–850. https://doi.org/10.1177/0743558403260109

20.

Burton

D. L.

Demuynck

Yoder

J. R.

(2016). Executive dysfunction predicts delinquency but not characteristics of sexual aggression among adolescent sexual offenders. Sexual Abuse: A Journal of Research and Treatment, 28(8), 707–721. https://doi.org/10.1177/1079063214556357

21.

Burton

D. L.

Duty

K. J.

Leibowitz

G. S.

(2011). Differences between sexually victimized and nonsexually victimized male adolescent sexual abusers: Developmental Antecedents and Behavioral Comparisons. Journal of Child Sexual Abuse, 20(1), 77–93. https://doi.org/10.1080/10538712.2011.541010

22.

Burton

D. L.

Meezan

(2004). Revisiting recent research on social learning theory as an etiological proposition for sexually abusive male adolescents. Journal of Evidence-based Social Work, 1(1), 41–80. https://doi.org/10.1300/j394v01n01_04

23.

Byrne

B. M.

Shavelson

R. J.

Muthén

(1989). Testing for the equivalence of factor covariance and mean structures: The issue of partial measurement invariance. Psychological Bulletin, 105(3), 456.

24.

Costello

A. B.

Osborne

J. W.

(2005). Best practices in exploratory factor analysis: Four recommendations for getting the most from your analysis. Practical Assessment, Research & Evaluation, 10(7), 1–9.

25.

Creeden

(2009). How trauma and attachment can impact neurodevelopment: Informing our understanding and treatment of sexual behaviour problems. Journal of Sexual Aggression, 15(3), 261–273. https://doi.org/10.1080/13552600903335844

26.

De Bellis

M. D

. (2001). Developmental traumatology: The psychobiological development of maltreated children and its implications for research, treatment, and policy. Development and Psychopathology, 13, 539–564.

27.

De Bellis

M. D.

Spratt

E. G.

Hooper

S. R

. (2011). Neurodevelopmental biology associated with childhood sexual abuse. Journal of Child Sexual Abuse, 20(5), 548–587.

28.

Dillard

Maguire-Jack

Showalter

Wolf

K. G.

Letson

M. M.

(2019). Abuse disclosures of youth with problem sexualized behaviors and trauma symptomology. Child Abuse & Neglect, 88, 201–211.

29.

Dillard

R. L.

(2020). How does that make you feel? Development and pilot testing of two new instruments measuring emotional attributions to experiences of childhood sexual abuse. Manuscript in preparation.

30.

Felitti

(2009). Adverse childhood experiences and adult health. Academic Pediatrics, 9(3), 131–132.

31.

Felitti

Anda

R. F.

Nordenberg

Williamson

D. F.

Sptiz

A. M.

Edwards

… Marks

J. S.

(1998). Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults. The Adverse Childhood Experiences (ACE) study. American Journal of Pediatric Medicine, 14(4), 245–258.

32.

Flora

D. B.

Curran

P. J.

(2004). An empirical evaluation of alternative methods of estimation for confirmatory factor analysis with ordinal data. Psychological Methods, 9(4), 466.

33.

Gao

Huang

(2017). Interaction between prenatal maternal stress and autonomic arousal in predicting conduct problems and psychopathic traits in children. Journal of Psychopathology and Behavioral Assessment, 39(1), 1–14. https://doi.org/10.1007/s10862-016-9556-8

34.

Grady

Yoder

Brown

(2018). Childhood maltreatment experiences, attachment, sexual offending: Testing a theory. Journal of Interpersonal Violence. Advance online publication. https://doi.org/10.1177/0886260518814262

35.

Grady

M. D.

Levenson

J. S.

Bolder

(2017). Linking adverse childhood effects and attachment: A theory of etiology for sexual offending. Trauma, Violence, & Abuse, 18(4), 433–444. https://doi.org/10.1177/1524838015627147

36.

Grady

M. D.

Shields

J. J.

(2018). The relationship between attachment dimensions and emotion regulation in individuals who have committed sexual crimes. Journal of Sexual Aggression, 24(1), 51–65. https://doi.org/10.1080/13552600.2017.1391630

37.

Graif

Matthews

S. A.

(2017). The long arm of poverty: Extended and relational geographies of child victimization and neighborhood violence exposures. Justice Quarterly, 34(6), 1096–1125.

38.

Guy

S. C.

Isquith

P. K.

Gioia

G. A.

(2004). BRIEF-SR: Behavior Rating Inventory of Executive Function–Self-Report version: Professional manual. Psychological Assessment Resources. Psychological Assessment Resources.

39.

Hernandez

D. C.

Marshall

Mineo

(2014). Maternal depression mediates the association between intimate partner violence and food insecurity. Journal of Women’s Health, 23(1), 29–37. https://doi.org/10.1089/jwh.2012.4224

40.

L. T.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55.

41.

Iacobucci

(2010). Structural equations modeling: Fit indices, sample size, and advanced topics. Journal of Consumer Psychology, 20(1), 90–98.

42.

Jaffee

S. R.

Maikovich-Fong

A. K.

(2013). Child maltreatment and risk for psychopathology. In Beauchaine

T. P.

Hinshaw

S. P.

(Eds.), Child and Adolescent Psychopathology (pp. 171–196). John Wiley & Sons, Hoboken, NJ, USA.

43.

Kavanaugh

Holler

(2014). Executive, emotional, and language functioning following childhood maltreatment and the influence of pediatric PTSD. Journal of Child & Adolescent Trauma, 7, 121–130.

44.

Kelly

Richardson

Hunter

Knapp

(2002). Attention and executive function deficits in adolescent sex offenders. Child Neuropsychology, 8(2), 138–143.

45.

Kerig

P. K.

Becker

S. P.

(2010). From internalizing to externalizing: Theoretical models of the processes linking PTSD to juvenile delinquency. In Egan

S. J.

(Ed.), Posttraumatic stress disorder (PTSD): Causes, symptoms and treatment (pp. 33–78). Nova Science Publishers.

46.

King

D. C.

Abram

K. M.

Romero

E. G.

Washburn

J. J.

Welty

L. J.

Teplin

L. A.

(2011). Childhood maltreatment and psychiatric disorders among detained youths. Psychiatric Services, 62(12), 1430–1438. https://doi.org/10.1176/appi.ps.004412010

47.

Knight

R. A.

Sims-Knight

J. E.

(2004). Testing an etiological model for male juvenile sexual offending against females. Journal of Child Sexual Abuse, 13(3–4), 33–55. https://doi.org/10.1300/j070v13n03_03

48.

Lateef

Jenney

(2020). Understanding sexually victimized male adolescents with sexually abusive behaviors: A narrative review and clinical implications. Trauma, Violence, & Abuse. Advance online publication. https://doi.org/10.1177/1524838020906558

49.

Levenson

J. S.

Baglivio

Wolff

K. T.

Epps

Royall

“Chip.,” W.

Gomez

K. C.

Kaplan

(2017). You learn what you live: Prevalence of childhood adversity in the lives of juveniles arrested for sexual offenses. Advances in Social Work, 18(1), 313. https://doi.org/10.18060/21204

50.

Levenson

J. S.

Grady

M. D.

(2016). The influence of childhood trauma on sexual violence and sexual deviance in adulthood. Traumatology, 22(2), 94–103. https://doi.org/10.1037/trm0000067

51.

Levenson

J. S.

Socia

K. M.

(2016). Adverse childhood experiences and arrest patterns in a sample of sexual offenders. Journal of Interpersonal Violence, 31(10), 1883–1911.

52.

Levenson

J. S.

Willis

G. M.

Prescott

D. S.

(2016). Adverse childhood experiences in the lives of male sex offenders: Implications for trauma-informed care. Sexual Abuse, 28(4), 340–359.

53.

Lewis

F. M.

Woods

N. F.

Ellison

(1987). Family impact study (Unpublished report, R01-NUO1000) Division of Nursing, Public Health Service, University of Washington.

54.

Ludy-Dobson

C. R.

Perry

B. D.

(2010). The role of healthy relational interactions in buffering the impact of childhood trauma. In Gil

(Ed.), Working with children to heal interpersonal trauma: The power of play (pp. 26–43). The Guilford Press.

55.

MacCallum

R. C.

Browne

M. W.

Sugawara

H. M.

(1996). Power analysis and determination of sample size for covariance structure modeling. Psychological Methods, 1(2), 130.

56.

Marshall

W. L.

Marshall

L. E.

(2016). An attachment-based theory of the aetiology of affiliative child molestation: Resilience/vulnerability factors across life-span development. In Boer

D. P.

(Ed.), The Wiley handbook on the theories, assessment and treatment of sexual offending (pp. 3–24). Wiley-Blackwell.

57.

McKillop

Smallbone

Wortley

Andjic

(2012). Offenders’ attachment and sexual abuse onset. Sexual Abuse, 24, 591–610. https://doi.org/10.1177/1079063212445571

58.

Menon

S. V.

Cohen

J. R.

Shorey

R. C.

Temple

J. R.

(2018). The impact of intimate partner violence exposure in adolescence and emerging adulthood: A developmental psychopathology approach. Journal of Clinical Child and Adolescent Psychology, 47(Suppl. 1), S497–S508. https://doi.org/10.1080/15374416.2018.1437736

59.

Miner

M. H.

Robinson

B. E.

Knight

R. A.

Berg

Swinburne Romine

Netland

(2010). Understanding sexual perpetration against children: Effects of attachment style, interpersonal involvement, and hypersexuality. Sexual Abuse: A Journal of Research and Treatment, 22(1), 58–77. https://doi.org/10.1177/1079063209353183

60.

Miner

M. H.

Swinburne Romine

Robinson

B. B. E.

Berg

Knight

R. A.

(2016). Anxious attachment, social isolation, and indicators of sex drive and compulsivity: Predictors of child sexual abuse perpetration in adolescent males? Sexual Abuse: A Journal of Research and Treatment, 28(2), 132–153.

61.

Morrell

L. M.

Burton

D. L.

(2014). An exploration of psychopathy in self-report measures among juvenile sex offenders. International Journal of Offender Therapy and Comparative Criminology, 58(5), 550–566.

62.

Muthén

L. K.

Muthén

B. O.

(2018). Mplus (Version 8).

63.

Natarajan

Somasundaram

C. P.

Sundaram

K. R.

(2011). Relationship between attachment security and rejection sensitivity in early adolescence. Psychological Studies, 56, 378. https://doi.org/10.1007/s12646-011-0108-8

64.

Nikulina

Widom

C. S.

(2013). Child maltreatment and executive functioning in middle adulthood: A prospective examination. Neuropsychology, 27(4), 417–427. https://doi.org/10.1037/a0032811

65.

Nooner

K. B.

Leaberry

K. D.

(2013). Trauma symptoms and executive functioning in children: A pilot report on depression and anxiety as mediators. Journal of Traumatic Stress Disorders and Treatment, 2(4), 4–5.

66.

O’Donnell

D. A.

Schwab-Stone

M. E.

Ruchkin

(2006). The mediating role of alienation in the development of maladjustment in youth exposed to community violence. Development and Psychopathology, 18, 215–232.

67.

Pereda

Guilera

Forns

Gómez-Benito

(2009). The prevalence of child sexual abuse in community and student samples: A meta-analysis. Clinical Psychology Review, 29(4), 328–338. https://doi.org/10.1016/j.cpr.2009.02.007

68.

Perry

B. D.

(2009). Examining child maltreatment through a neurodevelopmental lens: Clinical applications of the Neurosequential Model of Therapeutics. Journal of Loss and Trauma, 14(4), 240–255.

69.

Perry

B. D.

Pollard

R. A.

Blakley

T. L.

Baker

W. L.

Vigilante

(1995). Childhood trauma, the neurobiology of adaptation, and “use-dependent” development of the brain: How “states” become “traits.” Infant Mental Health Journal, 16(4), 271–291.

70.

Polak

A. R.

Witteveen

A. B.

Reitsma

J. B.

Olff

(2012). The role of executive function in posttraumatic stress disorder: A systematic review. Journal of Affective Disorders, 141(1), 11–21.

71.

Rosenthal

N. L.

Kobak

(2010). Assessing adolescents’ attachment hierarchies: Differences across developmental periods and associations with individual adaptation. Journal of Research on Adolescence, 20, 678–706.

72.

Salthouse

T. A.

Davis

H. P.

(2006). Organization of cognitive abilities and neuropsychological variables across the lifespan. Developmental Review, 26(1), 31–54.

73.

Sesma

H. W.

Slomine

B. S.

Ding

McCarthy

M. L.

(2008). Executive functioning in the first year after pediatric traumatic brain injury. Pediatrics, 121(6), e1686–e1695.

74.

Seto

M. C.

Lalumière

M. L.

(2010). What is so special about male adolescent sexual offending? A review and test of explanations through meta-analysis. Psychological Bulletin, 136(4), 526–575. https://doi.org/10.1037/a0019700

75.

Steinberg

(2008). A social neuroscience perspective on adolescent risk-taking. Developmental Review, 28(1), 78–106. https://doi.org/10.1016/j.dr.2007.08.002

76.

Tabachnick

B. G.

Fidell

L. S.

(2001). Using multivariate statistics. Allyn & Bacon.

77.

Turner

H. A.

Finkelhor

Ormrod

(2010). Poly-victimization in a national sample of children and youth. American Journal of Preventive Medicine, 38(3), 323–330.

78.

Turner

H. A.

Finkelhor

Ormrod

Hamby

Leeb

R. T.

Mercy

J. A.

Holt

(2012). Family context, victimization, and child trauma symptoms: Variations in safe stable, and nurturing relationships during early and middle childhood. American Journal of Orthopsychiatry, 82(2), 209–219.

79.

van Wijk

Blokland

Duits

Vermeiren

Harkink

. (2007). Relating psychiatric disorders, offender and offence characteristics in a sample of adolescent sex offenders and non-sex offenders. Criminal Behaviour and Mental Health, 17(1), 15–30. https://doi.org/10.1002/cbm.628

80.

van Wijk

Vermeiren

Loeber

Hart-Kerkhoffs

Doreleijers

Bullens

. (2006). Juvenile sex offenders compared to non-sex offenders: A review of the literature 1995-2005. Trauma, Violence, & Abuse, 7(4), 227–243. https://doi.org/10.1177/1524838006292519

81.

Veneziano

LeGrand

Richards

(2004). Neuropsychological executive functions of adolescent sex offenders and nonsex offenders. Perceptual and Motor Skills Research Exchange, 98(2), 661–674.

82.

Wang

Y. A.

Rhemtulla

(2020). Power analysis for parameter estimation in structural equation modeling: A discussion and tutorial. Advances in Methods and Practices in Psychological Science. Advance online publication. https://doi.org/10.31234/osf.io/pj67b

83.

Wanklyn

S. G.

Ward

A. K.

Cormier

N. S.

Day

D. M.

Newman

J. E.

(2012). Can we distinguish juvenile violent sex offenders, violent non-sex offenders, and versatile violent sex offenders based on childhood risk factors?. Journal of Interpersonal Violence, 27(11), 2128-2143. https://doi.org/10.1177/0886260511432153

84.

Widom

C. S.

(2017). Long-term impact of childhood abuse and neglect on crime and violence. Clinical Psychology: Science and Practice, 24(2), 186–202. https://doi.org/10.1111/cpsp.12194

85.

Wilson

H. W.

Berent

Donenberg

G. R.

Emerson

E. M.

Rodriguez

E. M.

Sandesara

(2013). Trauma history and PTSD symptoms in juvenile offenders on probation. Victims and Offenders, 8(4), 465–477.

86.

Wolpaw

J. M.

Ford

J. D.

Briere

(2005). Trauma Symptom Checklist for Children. In Grisso

Vincent

Seagrave

(Eds.), Mental health screening and assessment in juvenile justice (pp. 152–165). Guilford Press.

87.

Yoder

Caserta

D. A.

(2016). Associations between substance use in commission of sexual crimes and offense characteristics among youth: Mitigating effects of substance use treatment. International Journal of Offender Therapy and Comparative Criminology, 62(3), 655–675. https://doi.org/10.1177/0306624X16651903

88.

Yoder

Dillard

Leibowitz

G. S.

(2017). Family experiences and sexual victimization histories: A comparative analysis between youth sexual and nonsexual offenders. International Journal of Offender Therapy and Comparative Criminology, 62(10), 2917–2936. https://doi.org/10.1177/0306624x17738063

89.

Yoder

Grady

M. D.

Brown

Dillard

(2020). Criminogenic needs as intervening factors in the relation between insecure attachments and youth sexual violence. Sexual Abuse, 32(3), 247–272. https://doi.org/10.1177/1079063218821108

90.

Yoder

Grady

M. D.

Precht

(2019). Relationships between early life victimization, antisocial traits, and sexual violence: Executive functioning as a mediator. Journal of Child Sexual Abuse, 28(6), 667–689. https://doi.org/10.1080/10538712.2019.1588819

91.

Yoder

Precht

(2019). Victimization experiences and executive dysfunction as discriminating risk indicators for youth offender typologies. International Journal of Offender Therapy and Comparative Criminology, 64, 63–82. https://doi.org/10.1177/0306624x19865185

92.

Yoder

J. R.

Grady

Dillard

(2018). Maternal caregiving practices and child abuse experiences as developmental antecedents to insecure attachments: Differential pathways between adolescents who commit sexual and non-sexual crimes. Sexual Abuse, 31(7), 837–861. https://doi.org/10.1177/1079063218784557

93.

Yoder

J. R.

Leibowitz

G. S.

Peterson

(2016). Parental and peer attachment characteristics: Differentiating between youth sexual and non-sexual offenders and associations with sexual offense profiles. Journal of Interpersonal Violence, 33, 2643–2663.

94.

Zelazo

P. D.

Anderson

J. E.

Richler

Wallner-Allen

Beaumont

J. L.

Weintraub

(2013). II. NIH Toolbox Cognition Battery (CB): Measuring executive function and attention. Monographs of the Society for Research in Child Development, 78(4), 16–33.