Abstract
Background:
Although predictive associations between childhood executive control (EC) and adolescent substance use have been established in prior research, the developmental pathways involved in these long-term links have not been well understood.
Objective:
The goal of the current study was to investigate the degree to which aggressive behaviors, including both physical and relational aggression, and prosocial behaviors in elementary school operate as developmental pathways between preschool EC and adolescent substance use, while accounting for participants’ age, sex, family history of substance use, and family socioeconomic status.
Method:
Participants were 329 youth (49% male; 63.6% European American) who were recruited to participate in a study between 2006 and 2012 while youth were in preschool and elementary school and followed into adolescence. The sample was recruited from a small Midwestern city in the United States. EC was assessed with performance-based tasks when children were 5 years 3 months. Youth behaviors with peers were reported by teachers when participants were in elementary school. Self-reports of the substance use initiation (e-cigarettes, cigarettes, alcohol, and marijuana) were obtained in adolescence via phone surveys.
Results:
Mediation analyses revealed a statistically significant indirect effect from preschool EC to adolescent substance use through youth’s engagement in relational aggression in elementary school (b = −0.22 [−0.51; −0.08]; β = −0.18).
Conclusions:
Our results suggest that developmental pathways to adolescent substance use may begin in preschool, setting the stage for susceptibility to engagement in relational aggression, which increases, in turn, youth’s likelihood for substance use initiation in adolescence.
Keywords: executive control, peer relations, substance use, adolescence, longitudinal
As a concerning public health issue, substance use among adolescents is prevalent and can lead to morbidity and mortality (Gray & Squeglia, 2018). The results from a national survey on drug use (i.e., Monitoring the Future) indicate that the lifetime prevalence rates in U.S. 8th, 10th, and 12th graders in 2020 were 24.1%, 41%, and 47.2% for vaping, 11.5%, 13.9%, and 24% for cigarette use, 25.6%, 46.4%, and 61.5% for alcohol use, and 14.8%, 33.3%, and 43.7% for marijuana use (non-prescription), respectively (Johnston et al., 2021). Numerous studies have linked deficiencies in executive control (EC; also called executive function) with higher risk of substance use in adolescence concurrently (Pilin et al., 2020) and prospectively (Squeglia et al., 2014; Peeters et al., 2015; Harakeh et al., 2012; Jones et al., 2020; Pentz & Riggs, 2013). EC is commonly defined as a set of higher-order cognitive functions involved in decision-making, problem-solving, and self-regulation (Miyake et al., 2000). EC deficits could increase an individual’s risk of using potentially harmful substances (Peeters et al., 2015). To illustrate, when EC functions are deficient, adolescents may have more difficulties in resisting temptations to use substances (poor inhibitory control), comprehending instructions for resisting peer offers (poor working memory), and selectively shifting attention away from substance-related cues (poor cognitive flexibility), which ultimately can lead to higher rates of substance use (Jones et al., 2020; also see Kim-Spoon et al., 2017 for extensive review). Previous research has emphasized the importance of studying childhood EC, early in development, in associations with adolescent substance use (Riggs et al., 2011). Research on the development of EC has indicated that it first becomes organized and can be measured via performance-based tasks in preschool (Espy et al., 2016); thus, preschool is a critical period of EC development and may represent the earliest opportunity in which to observe EC deficiencies that have implications for later substance use. However, although basic predictive associations between childhood EC and adolescents’ substance use have been documented in prior research, the developmental pathways involved in these long-term links have not been well understood. From a theoretical perspective, the link between EC and substance use has been frequently explained by the slow development of prefrontal cortex (Steinberg, 2008) and imbalance in maturation of the brain regions responsible for approach, avoidance, and control (Casey, 2008), thus resulting in higher rates of risk-taking behaviors during adolescence (Kim-Spoon et al., 2017). Similarly, social information processing theories (Crick & Dodge,2006) have linked EC deficits with deficits in processing and encoding social information, such as misinterpreting others’ intentions or encoding neutral events as potentially threatening, which can generate aggressive behaviors with peers (Ellis et al., 2009). At the same time, social cognitive theories have emphasized the central role of EC in the development of social understanding (Lewis & Carpendale, 2009), characterized by abilities to accurately interpret the perceptions and beliefs of other people, which can generate prosocial behaviors with peers (Razza & Blair, 2009). There have been no empirical studies, however, that have examined associations between EC, aggressive and prosocial behaviors, and substance use in the same model. The current study addresses this gap by testing two potential mediating mechanisms in elementary school, children’s engagement in aggressive and prosocial behaviors, in associations between preschool EC and adolescent substance use. One line of evidence suggests that aggressive behavior (e.g., behavior intended to physically or verbally hurt others) could be a strong risk factor for adolescent substance use (Radliff et al., 2012; Arcadepani, Eskenazi, Fidalgo, & Hong, 2021). The other, less examined, line of evidence suggests that higher levels of prosocial behavior (e.g., behavior intended to help, comfort, or take care of others) could represent a potential protective factor for adolescent substance use (Memmott-Elison et al., 2020; Carlo, Crockett, Wilkinson, & Beal, 2011; Hatta, De Mol, Maurage, & Gabriel, 2018). Poor EC has been linked with higher aggressive behaviors (Poland et al., 2016; Granvald et al., 2016) and lower prosocial behaviors (Williams et al, 2017; Laible et al, 2016); however, the potential mediating roles of these behaviors in the long-term associations between childhood EC and adolescent substance use onset have not been tested.
Physical and Relational Aggression
Aggression can be defined in multiple ways, but research generally has distinguished between physical and relational aggression. Physical aggression refers to behaviors intended to physically (e.g., hitting, pushing, or kicking) or verbally (e.g., through threats) hurt others, whereas relational aggression refers to hurting others through social manipulations, such as social isolation or peer exclusion (Crick et al., 1999). Even though these two types of aggression are distinct, they tend to co-occur in elementary school (Crick & Grotpeter, 1996). One potential explanation of the associations between peer aggression and substance use is that adolescents with aggressive behaviors are more likely to affiliate with deviant peers who frequently use substances (Gaete et al., 2017). It has also been suggested that adolescents demonstrating aggressive behaviors may have poor social skills and are not positively engaged with peers (Quinn et al., 2016); in turn, they may use substances to cope with peer rejection (Carlyle & Steinman, 2007). In fact, evidence has shown that middle and high school youth who were both bullies and victims had the highest prevalence of cigarette, alcohol, and marijuana use (Radliff et al., 2012). In a systematic literature review examining associations between physical and relational aggression and adolescent drinking in various countries, Arcadepani et al. (2021) found that youth who reported bullying others (both physically and verbally) in elementary (grades 1–5) and middle school (grades 6–8) were more likely to use substances in adolescence. Further, in one study, youth engaged in physical or relational aggression in middle school (grades 6–8) were more likely to use substances in the same month compared to students who were less relationally and physically aggressive (Peleg-Oren et al, 2012). In another study, engagement in physical and relational bullying was positively correlated with drinking and smoking onset in high school students (grades 9–11; Quinn et al, 2016). Thus, physical and relational aggression may serve as intervening variables, over the long-term, in the link between childhood EC and adolescent substance use, although this remains to be tested.
Prosocial Behavior
Prosocial behaviors refer to behaviors aimed at comforting, helping, caring, and providing emotional support to others (Williams et al., 2017), and are focused on considerations of the needs and benefits of others (Keltner et al., 2014). One study documented that adolescents in middle and high schools in China who were rated by their peers as engaging in prosocial behaviors (e.g., “Shows care and concern for others”) were less likely to use substances and engage in deviant behaviors that are associated with substance use (Chen et al., 2019). Another longitudinal study conducted by Carlo and colleagues (Carlo et al., 2011) also reported that adolescents who were engaged in prosocial behaviors in grades 10–12 were less likely to engage in substance use seven years later. Further, a longitudinal study by Padilla-Walker and colleagues (2018) examined bidirectional associations between prosocial behavior and delinquent behavior, including substance use (e.g., alcohol, cigarettes, tobacco, and other drugs), in grades 8–12 using multivariate longitudinal growth analysis. Results indicated that increasing prosocial behavior was associated with decreasing aggression and substance use over time, suggesting that prosocial behavior could be a robust protective factor against adolescent substance use (Padilla-Walker et al., 2018). One potential explanation for the protective role of prosocial behavior is that adolescents who are frequently engaged in helping others are less likely to affiliate with peers who use substances; instead, they are more likely to spend time with peers who are involved in healthy behaviors (Chen et al., 2019). Taken together, these findings suggest that prosocial behavior may operate as another intervening variable linking childhood EC with adolescent substance use, in this case reducing risk.
EC and Peer Relations
Elementary school is a critical developmental period during which children begin to establish foundations of social relationships with others (Eisenberg, Fabes, and Spinrad, 2006). This period could be particularly difficult for children with poor EC, represented by lower abilities to inhibit impulses and respond to peers in a positive way (e.g., inhibitory control), remembering social rules (e.g., working memory), and flexibly switching attention away from provoking stimuli (e.g., cognitive flexibility). While frontal lobe deficits have not been linked to aggressive behavior per se, it has been noted that children showing physical aggression tend to have impaired executive control (Seigun, 2009), suggesting EC deficits in children with aggressive behaviors (Ellis et al., 2009). Research has documented that poor EC abilities characterized by deficits in inhibitory control, working memory, and planning were associated with both physical and relational aggression in elementary school children (Granvald et al., 2016; McQuade et al, 2013; Poland et al., 2016; Thomson et al, 2018). Studies have not yet examined the association of EC, as a broader construct, with prosocial behavior; however, studies conducted on related constructs, particularly self-regulation (Eisenberg et al., 2006) are informative. An integrated model proposed that self-regulation includes both (a) temperamental effortful control of arousal and emotions and (b) executive control of behavioral and higher-order cognitive processes (Zhou et al., 2012). It is important to note that EC, per se, tends to focus specifically on cognitive functions in emotionally neutral contexts, whereas temperamental effortful control focuses on emotionally meaningful situations. The two specific constructs, taken together, comprise the broader construct of self-regulation (Korucu et al., 2017). Self-regulation can influence prosocial behavior via its effects on one’s own emotions and behaviors, such that youth who are better self-regulated both emotionally and behaviorally tend to engage in more other-oriented responses rather than focus on personal distress (Eisenberg et al., 2006). For example, research has documented that children who are able to control their negative emotions (aggression and anger) in preschool are better able to cooperate with and help others in elementary school (Laible et al., 2016). Further, emotional regulation in preschool was associated with prosocial behavior in elementary school (Williams et al, 2017. Research suggests that higher levels of self-regulation were associated with higher prosocial behavior at home (e.g., “My child volunteers to help family members”) and with peers (e.g., “My child is kind toward peers”) both concurrently and longitudinally in middle and late adolescence (Carlo et al., 2012). To the extent these findings might generalize from the broad self-regulation construct to EC, specifically, they suggest that EC should be related to children’s ability to engage in prosocial behaviors in elementary school. Some initial evidence suggests that EC contributes to better social understanding (Lewis & Carpendale, 2009; Razza & Blair, 2009); however, studies have not yet been conducted to test the long-term associations leading from preschool EC to childhood peer relations to adolescent substance use onset.
Sex Differences in EC, Peer Relations, and Substance Use
Previous evidence reports that girls have better EC abilities in certain tasks in preschool compared to boys (Matthews et al., 2009). Boys compared to girls also are less likely to use relational aggression to harm others (Crapanzano et al., 2010) and more likely to use physical and verbal aggression (Ostrov & Keating, 2004). Boys compared to girls also show lower levels of empathy and prosocial behavior (Longobardi et al., 2019). Further, sex differences in the prevalence of adolescent alcohol and many other types of substance use have been decreasing over time, with girls reporting higher rates of substance use compared to previous cohorts (Monitoring the Future; Johnston et al., 2021). For these reasons, it is important to examine the longitudinal associations among preschool EC, childhood peer relations, and adolescent substance use while adjusting for sex.
The Current Study
The current study investigates the degree to which aggressive behaviors, including both physical and relational aggression, and prosocial behaviors in elementary school operate as developmental pathways between preschool EC and adolescent substance use initiation, while accounting for participants’ age, sex, family history of substance use, and family socioeconomic status. Our study tested two main hypotheses. First, we hypothesized that relational and physical aggression in elementary school would mediate the association between preschool EC and adolescent substance use, representing potential risk pathways. Second, we hypothesized that prosocial behavior in elementary school would mediate the associations between EC in childhood and substance use in adolescence, representing a potential protective pathway. Bringing light to the potential contributions from aggressive and prosocial behaviors in elementary school to the association between preschool EC and substance use in adolescence can inform the development of prevention and intervention programs, enhancing our understanding of how behaviors with peers as early as elementary school can contribute to future substance use in adolescence. Lastly, in addition to controlling for sex, our analyses controlled for family socioeconomic status, history of alcohol problems, and child age, because these factors have been associated with both child cognitive development (Berthlsen et al., 2017) and adolescent substance use (Griffin & Botvin, 2010).
Method
Participants and Procedures
The participants were 329 youth (49% male) and their mothers, who participated in a longitudinal cohort sequential study on the development of executive control in preschool and its associations with a range of subsequent health outcomes (Espy et al., 2016). Children and their families were recruited from a small Midwestern city between 2006 and 2012, while youth were in preschool and elementary school. The study’s focus was to measure the development of EC in a typically developing sample, therefore the study excluded (prior to enrollment) children diagnosed with speech or language delays, developmental or behavioral disorders, and for whom English was not the primary language spoken in the home. Participants from lower-income families were over-sampled to increase representation of the families with higher sociodemographic risk. At enrollment to the study 53.7% of families received public medical assistance and 37.1% of households were headed by one parent. The analysis sample was 63.6% European American, 3.8% African American, 13.4% Hispanic, 0.3% Asian American, and 18.8% multiracial.
For the current study, the data were drawn from a preschool lab assessment of executive control, elementary teacher’s surveys on peer relations, adolescent phone interviews on substance use, and intake parental interviews (e.g., covariates). All procedures were approved by the Institutional Review Board of the University of Nebraska – Lincoln.
Preschool.
At the preschool phase, when children were 5 years 3 months, they completed a comprehensive battery of laboratory-based tasks measuring the primary domains of EC. Families were requested to fill out a family background questionnaire at their first visit to the lab. Preschool EC data was available for 313 youth.
Elementary school.
During the elementary school phase of the study, when youth were 6–10 years old, teachers filled out a questionnaire on each child’s social behavior in the spring of each grade (e.g. first through fourth). Elementary school data was available for 219 youth in the first grade, 260 youth in the second grade, 269 youth in the third grade, and 248 youth in the fourth grade. The number of individuals who had at least one assessment in elementary school was n = 308.
Adolescence.
During the adolescent phase of the study, when youth were 14–18 years old, 246 participants contributed to one or more phone survey between June 2017 and March 2020. The phone survey included questions about substance use initiation. At the time of the latest phone interview on substance use initiation, the majority of adolescents were 16 years old (37%, n = 91), followed by 15 year olds (20.7%, n = 51), 14 year olds (19.1%, n = 47), 17 year olds (15.9%, n = 39), and 18 year olds (5.5%, n = 18). The number of individuals with complete data from three assessment points (e.g., preschool, elementary school, and adolescence) was n = 227. To maximize use of all available information, the analytic sample consisted of 329 youth (including 16 youth who participated during the elementary phase, but were not among the 313 participants who completed preschool EC assessment at the age of 5 years 3 months). Of the 313 children who completed executive control assessments in preschool, those not participating in elementary or adolescent assessments were primarily from families who could not be located or declined to participate in further assessments. Families with missing data in adolescence had lower mean income-to-needs ratio at baseline (M = 1.76, SD = 1.29) compared to families retained through adolescence (M = 2.43, SD = 1.78), t(309) = −3.08, p = .002, but did not differ from retained families on the rate of family history of alcoholism (30.37% vs. 29.48%) χ2(1) = .023, p = .881. Youth with missing data in adolescence had lower EC scores on only two out of nine EC tasks: Go/No-Go (t(311) = −3.42, p = .001) and Nebraska Barnyard (t(310) = −2.64, p = .01), reflecting no particular pattern to the differences. Missing data on covariates was less than 1%.
Measures
Preschool executive control.
Executive control was assessed in preschool with nine lab-based tasks. Working memory was measured with three tasks: Nine Boxes, adapted from Diamond, Prevor, Callender, and Druin (1997), Delayed Alternation (Espy et al., 1999; Goldman, Rosvold, Vest, & Galkin, 1971), and Nebraska Barnyard adapted from Noisy Book (Hughes, Dunn, & White, 1998). Inhibitory control was measured with four tasks: Big-Little Stroop adapted from Kochanska, Murray, and Harlan (2000), Go/No-Go adapted from Simpson and Riggs (2006), Shape School-Inhibit Condition (Espy, 1997; Espy, Bull, Martin, & Stroup, 2006), and Snack Delay adapted from Kochanska, Murray, Jacques, Koenig, and Vandegeest (1996), and Korkman, Kirk, and Kemp (1998). Flexible shifting was measured with two tasks: Shape School-Switching Condition (Espy, 1997; Espy et al., 2006) and Trails-Switching Condition, modified from Espy and Cwik (2004). A unitary latent EC factor was created using the nine tasks as indicators (Weibe at al., 2011). A higher score on the latent EC factor represents better EC abilities.
Childhood aggression and prosocial behavior.
Child’s physical and relational aggression and prosocial behavior in first, second, third, and fourth grade were reported by teachers using the Children’s Social Behavior Scale (CSBS-T, Crick, 1996). For each item, the teachers were asked to rate the behavior on a 5-point scale ranging from 1 = never true to 5 = almost always true. At each grade, relational aggression was calculated as a mean of five items (e.g. When this child is mad at a peer, s/he gets even by excluding the peer from his or her clique or play group). Physical aggression (e.g., This child initiates or gets into physical fights with peers) and prosocial behavior (e.g., This child says supportive things to peers) were calculated as a mean of four items each at each grade. Cronbach’s alpha reliability coefficients for relational and physical aggression and prosocial behavior across the four grades were .83–.84, .76 –.87, and .89 –.90, correspondingly.
Adolescent substance use initiation.
Substance use initiation in adolescence was based on four dichotomous questions from the adolescent phone interview referring to the lifetime use of four substances, including alcohol, electronic cigarettes, cigarettes, and marijuana (e.g. “Have you ever used alcohol?”).
Covariates.
Adolescent sex was based on the demographic questionnaire at the intake interview (1= male; 0 = female). Adolescent age was the age at which an individual completed their last adolescent survey prior to March 2020. Family history of alcoholism was included to control for substance misuse in the family. At the intake interview, parents indicated if anybody in the child’s family (biological parents or grandparents) had ever been “diagnosed or treated” for an alcohol problem (1 = Yes, 0 = No). Family socioeconomic status was based on family’s income-to-needs ratio, which was calculated by dividing the total household income (adjusted for the family size) by the federal poverty threshold (adjusted for the year of enrollment into a study).
Analyses
Mediation analyses were conducted in Mplus 8.4 (Muthen & Muthen, 1998–2010) in three steps. First, a measurement model that included a latent childhood EC factor (indicated by the nine EC task scores), a latent adolescent substance use factor (indicated by the four substance initiation items), and each latent childhood mediator (each indicated by the four scales at Grades 1–4; e.g., physical aggression in Grade 1, 2, 3, and 4) was estimated to obtain factor loadings for each latent variable, assess overall associations among latent variables, and obtain global model fit prior to examining structural paths. Note that supplemental latent growth curve analyses were conducted using the repeated measures for each mediator from first through the fourth grades to determine the presence of systematic change over time, yet none of the three potential mediators had significant variance in the slope factor. So, a latent factor for each mediator variable was created by fixing a factor loading of the first grade to 1 and allowing the other three loadings from the second, third, and fourth grades to be freely estimated.
Second, a systematic model building strategy was implemented in which a set of structural models was estimated, one at a time, to examine a single mediator (i.e., physical aggression, relational aggression, and prosocial behavior) of the association between preschool EC and adolescent substance use. Third, this culminated in a multiple mediator analysis combining the three mediators in one model to address our primary hypotheses. Mediation was examined based on the statistical significance of the indirect effects, calculated by Mplus using bias-corrected bootstrapped 95% confidence intervals based on 1,000 bootstrap samples. Bias-corrected bootstrapped confidence intervals are estimated using computer intensive resampling from the original sample and are considered to be more accurate compared to bootstrapped confidence intervals (MacKinnon, Lockwood, & Williams, 2004; Valente et al., 2016). Confidence intervals that do not include zero indicate statistical significance at p < .05.
Each model controlled for the effects of adolescent sex, age at the last phone interview, family income-to-needs ratio, and family history of alcohol on childhood EC, adolescent substance use, and each mediator. A model was considered to fit the data adequately if the Comparative Fit Index (CFI) and the Tucker Lewis Index (TLI ) were greater than or equal to .90, and the Root Mean Square Error of Approximation (RMSEA) was less than or equal to .06 (Marsh, Hau, & Wen, 2004). Because of the dichotomous nature of the substance use indicators, all analyses used the weighted least squares mean- and variance (WLSMV) adjusted estimator, recommended for binary variables. The WLSMV estimator in Mplus employs a pairwise missing data approach on the observed exogenous variables in the model. There was no missingness on the sex covariate; therefore, this pairwise missing data approach resulted in retention of the entire sample of 329 cases in all models.
Results
Descriptive Statistics
Table 1 presents descriptive statistics of the analytic variables. Lifetime prevalence of substance initiation among adolescents in the sample was as follows: e-cigarette (38%), cigarette (11%), alcohol (24%), and marijuana (22%). By comparison to national data, the lifetime prevalence for 8th, 10th, and 12th graders combined was 37.2% for any vaping, 16.2% for cigarette smoking, 44% for alcohol use, and 30.2% for marijuana use (Johnston et al, 2021). It is notable that in elementary school youth had lower levels of physical aggression compared to relational aggression, and relatively high levels of prosocial behavior. Further examination of the physical aggression variables revealed positive skewness at each time point (e.g., skewness = 2.27 in the first grade, 2.78 in the second grade; 3.73 in the third grade; 3.05 in the fourth grade), indicating that it was used less frequently compared to relational aggression. The skewness was addressed in the primary analyses with the use of the WLSMV estimator, which is suitable for non-normally distributed and categorical variables (Brown, 2006).
Table 1.
Descriptive Statistics of the Study Variables
Study Variables | Min | Max | M/% | SD |
---|---|---|---|---|
EC tasks (preschool) N=313 | ||||
Nine Boxes | 2 | 9 | 5.67 | 1.84 |
Delayed Alternation | −6 | 16 | 7.86 | 5.91 |
Nebraska Barnyard | 2.33 | 15.00 | 8.65 | 2.43 |
Big Little Stroop | 0.79 | 4.91 | 1.43 | 0.54 |
Go/No-Go | −0.17 | 3.12 | 2.68 | 0.58 |
Shape School – Inhibition | 0.00 | 1.00 | 0.96 | 0.12 |
Shape School – Switching | 0.20 | 1.00 | 0.87 | 0.16 |
Modified Snack Delay | 0.0 | 48.0 | 27.54 | 9.76 |
Trails | 0.35 | 1.00 | 0.90 | 0.11 |
Peer relations (elementary school) N=309 | ||||
G1 Relational Aggression | 5 | 25 | 8.87 | 3.95 |
G2 Relational Aggression | 5 | 24 | 8.75 | 4.21 |
G3 Relational Aggression | 5 | 24 | 8.83 | 3.99 |
G4 Relational Aggression | 5 | 24 | 8.37 | 3.90 |
G1 Physical Aggression | 4 | 17 | 5.33 | 2.52 |
G2 Physical Aggression | 4 | 18 | 5.00 | 2.07 |
G3 Physical Aggression | 3 | 20 | 4.97 | 2.39 |
G4 Physical Aggression | 4 | 16 | 4.85 | 2.09 |
G1 Prosocial Behavior | 6 | 20 | 15.12 | 3.36 |
G2 Prosocial Behavior | 5 | 20 | 15.34 | 3.43 |
G3 Prosocial Behavior | 4 | 20 | 14.98 | 3.41 |
G4 Prosocial Behavior | 5 | 20 | 15.08 | 3.40 |
Substance Use Initiation N=246 | ||||
E-cigarettes (yes) | 0 | 1 | 93 (38%) | |
Cigarettes (yes) | 0 | 1 | 26 (11%) | |
Alcohol (yes) | 0 | 1 | 59 (24%) | |
Marijuana (yes) | 0 | 1 | 55 (22%) | |
Covariates | ||||
Recent Age (N=246) | 14 | 18 | 15.72 | 1.16 |
Sex (male; N=329) | 161 (49%) | |||
Family history of alcohol (yes; N=313) | 0 | 1 | 93 (30%) | |
Income to needs ratio (N=311) | 0 | 10.77 | 2.27 | 1.70 |
Measurement Model
The measurement model that included latent childhood EC, adolescent substance use, childhood relational aggression, childhood physical aggression, and childhood prosocial behavior factors had an acceptable model fit (χ2= 396.01, df = 264, p <.001; CFI= 0.91, TLI = .90, RMSEA =.04). All factor loadings were statistically significant and ranged (in standardized values) from .30 to .61 for preschool EC, from .75 to .97 for substance use, from .57 to .62 for relational aggression, from .59 to .64 for physical aggression, and from .50 to .68 for prosocial behavior factors (See Figure 1). Figure 1 also presents correlations among latent variables. As expected, preschool EC was negatively correlated with both relational aggression (r = −0.37, p <.001) and physical aggression (r = −0.48, p <.001), and positively correlated with prosocial behavior (r = 0.39, p <.001). Substance use was non-significantly associated with both preschool EC (r = −0.17, p =.056) and prosocial behavior (r = −0.14, p =.147), and was significantly positively associated with relational aggression (r = 0.39, p <.001) and physical aggression (r = 0.22, p <.001).
Figure 1. The Measurement Model for the Indicators of the Continuous Latent Variables.
Note. Standardized regression coefficients are present for factor loadings.
*p <.05; **p <.01; ***p <.001.
Single Mediator Analyses
Table 3 presents results from the series of single mediator models. It is notable that preschool EC had significant negative associations with relational (b = −1.33 [95% CI = −2.64; −0.49]; β = −0.37) and physical aggression (b = −0.89 [95% CI = −1.48; −0.42]; β = −0.44), and a significant positive association with prosocial behavior in elementary school (b = 0.75 [95% CI = 0.20; 1.48]; β = 0.31); however, the indirect effect from childhood EC to adolescent substance use was significant only in a model for relational aggression (b = −0.19 [95% CI = −0.42; −0.06]; β = −0.15).
Multiple Mediator Analyses
A model that combined both the relational and physical aggression factors with the prosocial behavior factor was considered first but did not converge, likely due to the large number of parameter estimates combined with the relatively small sample size and the skewed distribution of the physical aggression variable. After taking physical aggression out of the model, the best-fitting model included relational aggression and prosocial behavior. In this model, relational aggression and prosocial behavior were correlated, and time-specific residuals between relational aggression and prosocial behavior were allowed to covary within time. This model had an acceptable model fit (χ2= 339.89, df = 247, p < .001; CFI = 0.92, TLI = .90, RMSEA =.03). The indirect effect through relational aggression remained significant (b = −0.22 [95% CI = −0.51; −0.08]; β = −0.18), and the indirect effect through prosocial behavior remained non-significant (b = 0.05 [95% CI = −0.05; 0.26]; β = 0.04). Thus, our final model was a model capturing associations between preschool EC, adolescent substance use, relational aggression, and prosocial behavior (Figure 2).
Figure 2. Substance Use Mediation Model: Relational Aggression and Prosocial Behavior Mediation.
Note. Solid black lines indicate statistically significant paths. Standardized regression coefficients are reported in parentheses
Model fit: χ2= 339.89, df = 247, p < .001; CFI = 0.92, TLI = .90, RMSEA =.03; N= 329.
As shown in Figure 2, our findings indicate that preschool EC was associated with higher levels of prosocial behavior (b = .78, p = .020, β = .31) and lower levels of relational aggression (b = −1.38, p = .011, β = −.38) in elementary school, indicating that children with stronger EC in preschool were more likely to engage in prosocial behaviors and less likely to engage in relational aggression in elementary school. Furthermore, higher EC in preschool was associated with lower substance use in adolescence by its association with the relational aggression in elementary school, which was positively associated with substance use in adolescence.
Covariates.
Boys compared to girls had lower EC in preschool (b = −.24, p = .014, β = −.19), were less engaged in prosocial behavior (b = −1.02, p = .001, β = −.33) and relational aggression (b = −1.10, p = .012, β = −.24) in elementary school, and did not differ from girls on the substance use in adolescence (b = .23, p =.278, β = .15). The age at which individuals completed their last adolescent survey did not significantly contribute to their engagement in relational aggression (b = −.17, p =.44, β = −.09) or prosocial behavior (b = .06, p =.586, β = .05) in elementary school, but was significantly associated with substance use in adolescence (b = .26, p <.001, β = .40), indicating that youth who completed their last substance use phone interview when they were older, were more likely to have initiated substance use. Youth from families with the higher incomes had higher EC in preschool (b = .07, p =.012, β = .20) and lower substance use in adolescence (b = −.11, p =.008, β = −.24), and did not differ from other youth in their engagement in either relational aggression (b = −.09, p =.43, β = −.07) or prosocial behavior (b = .08, p =.29, β = .08). Family history of alcohol was not significantly associated with preschool EC, youth’s relational aggression or prosocial behavior, or substance use in adolescence. The final model accounted for 34.7% of the variance in the substance use outcome.
Sensitivity Analyses
Sensitivity analyses were conducted after excluding 16 adolescents who joined the study during the elementary school phase and did not have an EC assessment in preschool. The analytic sample for sensitivity analyses thus included 313 youth with no missing data on EC. As before, the model combining relational and physical aggression did not converge, and the best fitting model included relational aggression and prosocial behavior. This model had an acceptable fit (χ2= 334.73, df = 247, p < .001; CFI = 0.92, TLI = .90, RMSEA =.03). The indirect effect through relational aggression remained statistically significant (b = −0.20 [95% CI = −0.47; −0.05]; β = −0.17), and the indirect effect through prosocial behavior remained statistically non-significant (b = 0.06 [95% CI = −0.04; 0.29]; β = 0.05). These results suggest that the findings were not affected by the 16 youth who joined the study during the elementary school phase.
Discussion
The goal of this study was to investigate how relational and physical aggression (a risk pathway) and prosocial behavior (a protective pathway) in elementary school could mediate associations between preschool EC and adolescent substance use, while accounting for participants’ age, sex, family history of substance use, and family socioeconomic status. Our analyses revealed that the indirect effect of preschool EC on adolescent substance use through relational (but not physical) aggression in elementary school was statistically significant, partially confirming the hypothesized risk pathway. The indirect effect through prosocial behavior was non-significant, yet preschool EC had a significant positive association with the prosocial behavior factor in elementary school. Overall, these findings suggest that developmental pathways to adolescent substance use may begin in preschool, setting the stage for susceptibility to engagement in relational aggression, which increases, in turn, youth’s likelihood for substance use in adolescence.
One potential explanation for the association between relational aggression and substance use relates to the fact that both behaviors are fundamentally social activities for youth. Relational aggression, by definition, occurs within a social context in which the perpetrator relies on the verbal and emotional cues of an interpersonal interaction to achieve desired ends (e.g., “threatens to stop being a friend”). Although physical aggression also represents a social interaction, it manifests in behaviors, such as hitting and pushing, that are less reliant on social manipulation through verbal communication. Substance use is predominantly a social activity among adolescents (Fujimoto & Valente, 2012; McCabe, West, Veliz, Frank, & Boyd, 2014). Substance initiation, in particular, typically requires being embedded within a social network that provides access to substances as well as socialization in their use. It is possible that youth who display relational aggression subsequently gravitate toward social contexts that provide opportunities to interact with and learn from like-minded peers who have initiated substance use in adolescence.
Findings from the hypothesized protective pathway indicated that engagment in prosocial behaviors in elementary school did not mediate associations between preschool EC and adolescent substance use, yet preschool EC had positive associations with prosocial behavior in elementary school. Even though it has been previously suggested that adolescents who are frequently engaged in helping others are less likely to engage in substance use (Chen et al., 2019; Padilla-Walker et al., 2018, Carlo et al., 2011), our results did not support this hypothesis. The lack of significant longitudinal associations between prosocial behavior and substance use in our study could be due to the fact that we assessed prosocial behavior in elementary school, compared to middle or high school assessments of prosocial behavior in studies that linked it more proximally with lower likelihood of substance use in adolescence (Chen et al., 2019, Padilla-Walker et al., 2018). In addition, most of the studies that linked prosocial behavior with the likelihood of substance use in adolescence did not focus exclusively on substance use but rather approached it as one of the indicators of deviant behaviors, frequently associated with substance use such as “violating school/home rules” (Hofmann & Muller, 2018), “skipping classes” (Chen et al., 2019), or “staying out all night” (Padilla-Walker et al, 2018). Therefore, it remains unclear whether prosocial behavior has independent predictive associations with substance use, or whether this effect is interdependent on youth engagement in delinquent behavior.
Higher preschool EC was significantly associated with less likelihood of relational and physical aggression and higher likelihood of prosocial behavior in elementary school, indicating that higher levels of preschool EC could enhance positive social relationships with peers in elementary school. These findings are in line with previous research suggesting that children with stronger EC are able to better regulate their behaviors, respond to their peers in socially acceptable ways, and have more flexible thinking, which reduces the likelihood for engagement in aggressive behaviors (McQuade, 2017) and promotes engagement in prosocial behaviors (Riccio et al, 2011).
Consistent with previous research, our findings demonstrated that boys had lower EC in preschool compared to girls (Matthews et al., 2009). In this regard it has been suggested that boys might be prone to more behavior regulation difficulties that can potentially increase the likelihood of aggressive behavior (Ellis et al., 2009). Boys were also less engaged in relational aggression in elementary school compared to girls. In line with our findings, it has been previously suggested that when boys behave aggressively, they are less likely to choose relational aggression, but instead choose to use physical and verbal aggression to harm others (Ostrov & Keating, 2004). But the latter could not be tested in our study because of non-convergence of the model with the physical aggression, which is a limitation. Our study also reported that boys compared to girls were less engaged in prosocial behavior in elementary school. One common way to explain sex differences in prosocial behavior could be looking at the social norms expectations, as it has been previously suggested that girls compared to boys might demonstrate more prosocial behaviors in order to comply with gendered social norms (Longobardi et al., 2019). Finally, boys did not differ from girls on the substance use rates in adolescence, which is consistent with the national trends on substance use (Johnston et al., 2021).
Study Strengths and Limitations.
Our study has a number of strengths, including the long-term longitudinal design, performance-based assessment of EC, and focus on mediating mechanisms. Our study also has a few important limitations. Because our sample was assessed relatively early in adolescence (M (age) = 15.72, SD = 1.16), most of our substance use questions asked about initiation (e.g., Have you ever tried?). Extending assessment later into adolescence will provide opportunities to examine the mediating role of peer relations in elementary school in associations between preschool EC and regular or problematic use of substances. In addition, reliance exclusively on teacher reports of aggressive and prosocial behaviors in school could be a limitation, because teachers see youth’s behaviors in different contexts compared to parents and peers. Combining data on youth aggressive and prosocial behaviors from multiple reporters, or explicitly studying the discrepancies among reporters, could add valuable information on how youth express themselves in across different contexts (De Los Reyes, 2011). Further, a stronger test of the hypothesized multivariate mediator model, including both physical and relational aggression as well as prosocial behavior, could allow for more precise tests of the risk and protective pathways in the associations between preschool EC and substance use in adolescence for boys and girls. In addition, all participants were recruited form a single site in the U.S., so it is not possible to generalize to a broader population.
Conclusion
Findings from our study hold promise for the development of preventive interventions that could mitigate peer difficulties in elementary school and, in turn, substance initiation in adolescence. It has been suggested that social skills training focused on reducing aggressive behaviors and increasing prosocial behaviors should begin early in preschool (Hay et al., 2021). Of note, preventive interventions focused on improving executive functioning in young children already show long-lasting results in improving self-regulatory behaviors and reducing problem behaviors in school (Downer et al., 2018; Tominey & McClelland, 2011). It is possible that such results translate into positive, prosocial relationships that protect, in turn, against affiliation with deviant peers and, ultimately, substance initiation. Future experimental research could extend these findings by investigating if participation in EC-focused interventions early in development can contribute to lower risk of substance use in adolescence.
Table 2.
Results from Single-Mediator Models
Path | B | 95 % CI | p-value | β | Model Fit Statistics | |||||
---|---|---|---|---|---|---|---|---|---|---|
Lower | Upper | Chi Sq | df | CFI | TLI | RMSEA | ||||
Relational Aggression | ||||||||||
EC >>Mediator | −1.33 | −2.64 | −0.49 | 0.02 | −0.37 | 249.855 | 172 | 0.92 | 0.902 | 0.037 |
Mediator>Sub.Use | 0.15 | 0.06 | 0.29 | 0.02 | 0.41 | |||||
EC >> Sub.Use | 0.02 | −0.31 | 0.42 | 0.903 | 0.02 | |||||
indirect effect | −0.19 | −0.42 | −0.06 | 0.07 | −0.15 | |||||
Physical Aggression | 262.215 | 172 | 0.92 | 0.902 | 0.04 | |||||
EC >>Mediator | −0.89 | −1.48 | −0.42 | 0.002 | −0.44 | |||||
Mediator>Sub.Use | 0.16 | −0.02 | 0.46 | 0.27 | 0.26 | |||||
EC >> Sub.Use | −0.01 | −0.42 | 0.31 | 0.938 | −0.01 | |||||
indirect effect | −0.14 | −0.40 | 0.02 | 0.224 | −0.12 | |||||
Prosocial Behavior | 261.243 | 172 | 0.907 | 0.886 | 0.04 | |||||
EC >>Mediator | 0.75 | 0.20 | 1.48 | 0.017 | 0.31 | |||||
Mediator>Sub.Use | −0.05 | −0.21 | 0.06 | 0.441 | −0.11 | |||||
EC >> Sub.Use | −0.11 | −0.48 | 0.16 | 0.493 | −0.09 | |||||
indirect effect | −0.04 | −0.16 | 0.05 | 0.471 | −0.03 |
Note. Sub.Use = Substance Use.
Acknowledgements
This work was supported by the National Institute of Mental Health (R01 MH065668), the National Institute of General Medical Sciences (P20 GM130461), and the National Institute On Drug Abuse (R01 DA041738) of the National Institutes of Health. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or the funding agencies.
Footnotes
The authors have no conflicts of interest to declare.
Contributor Information
Irina Patwardhan, Boys Town.
Ying Guo, University of Tennessee Health Science Center.
Emily R. Hamburger, University of Nebraska-Lincoln
Saira Sarwar, University of Nebraska-Lincoln.
Charles B. Fleming, University of Washington
Tiffany D. James, University of Nebraska-Lincoln
Jennifer Mize Nelson, University of Nebraska-Lincoln.
Kimberly Andrews Espy, University of Texas at San Antonio.
Timothy D. Nelson, University of Nebraska-Lincoln
W. Alex Mason, University of Tennessee Health Science Center.
References
- Arcadepani FB, Eskenazi DY, Fidalgo TM, & Hong JS (2021). An exploration of the link between bullying perpetration and substance use: A review of the literature. Trauma, Violence, & Abuse, 22(1), 207–214. [DOI] [PubMed] [Google Scholar]
- Berthelsen D, Hayes N, White SL, & Williams KE (2017). Executive function in adolescence: Associations with child and family risk factors and self-regulation in early childhood. Frontiers in Psychology, 8, 903. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Brown T (2006). Confirmatory factor analysis for applied research. New York: Guildford. [Google Scholar]
- Carlo G, Crockett LJ, Wilkinson JL, & Beal SJ (2011). The longitudinal relationships between rural adolescents’ prosocial behaviors and young adult substance use. Journal of youth and adolescence, 40(9), 1192–1202. [DOI] [PubMed] [Google Scholar]
- Carlo G, Crockett LJ, Wolff JM, & Beal SJ (2012). The role of emotional reactivity, self-regulation, and puberty in adolescents' prosocial behaviors. Social Development, 21(4), 667–685. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Carlyle KE, & Steinman KJ (2007). Demographic differences in the prevalence, co-occurrence, and correlates of adolescent bullying at school. Journal of School Health, 77(9), 623–629. [DOI] [PubMed] [Google Scholar]
- Casey BJ, Getz S, & Galvan A (2008). The adolescent brain. Developmental review, 28(1), 62–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chen L, Chen X, Zhao S, French DC, Jin S, & Li L (2019). Predicting substance use and deviant behavior from prosociality and sociability in adolescents. Journal of youth and adolescence, 48(4), 744–752. [DOI] [PubMed] [Google Scholar]
- Crick NR (1996). The role of overt aggression, relational aggression, and prosocial behavior in the prediction of children's future social adjustment. Child development, 67(5), 2317–2327. [PubMed] [Google Scholar]
- Crapanzano AM, Frick PJ, & Terranova AM (2010). Patterns of physical and relational aggression in a school-based sample of boys and girls. Journal Of Abnormal Child Psychology, 38(4), 433–445. [DOI] [PubMed] [Google Scholar]
- Crick NR, & Grotpeter JK (1996). Children's treatment by peers: Victims of relational and overt aggression. Development and psychopathology, 8(2), 367–380. [Google Scholar]
- Crick NR, Casas JF, & Ku HC (1999). Relational and physical forms of peer victimization in preschool. Developmental psychology, 35(2), 376. [DOI] [PubMed] [Google Scholar]
- De Los Reyes A (2011). Introduction to the special section: More than measurement error: Discovering meaning behind informant discrepancies in clinical assessments of children and adolescents. Journal of Clinical Child & Adolescent Psychology, 40(1), 1–9. 10.1080/15374416.2011.533405 [DOI] [PubMed] [Google Scholar]
- Diamond A, Prevor MB, Callender G, & Druin DP (1997). Prefrontal cortex cognitive deficits in children treated early and continuously for PKU. Monographs of the society for research in child development, i–206. [PubMed] [Google Scholar]
- Downer JT, Williford AP, Bulotsky-Shearer RJ, Vitiello VE, Bouza J, Reilly S, & Lhospital A (2018). Using data-driven, video-based early childhood consultation with teachers to reduce children’s challenging behaviors and improve engagement in preschool classrooms. School Mental Health, 10(3), 226–242. [Google Scholar]
- Eisenberg N, Fabes RA, & Spinrad TL (2006). Prosocial development. In Damon W & Lerner R (Eds.), Handbook of child psychology, social, emotional, and personality development (Vol. 3, pp. 646–702). Hoboken, NJ: John Wiley & Sons. [Google Scholar]
- Ellis ML, Weiss B, & Lochman JE (2009). Executive functions in children: Associations with aggressive behavior and appraisal processing. Journal Of Abnormal Child Psychology, 37(7), 945–956. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Espy KA (1997). The Shape School: Assessing executive function in preschool children. Developmental Neuropsychology, 13(4), 495–499. [Google Scholar]
- Espy KA, & Cwik MF (2004). The development of a trial making test in young children: the TRAILS-P. The Clinical Neuropsychologist, 18(3), 411–422. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Espy KA, Bull R, Martin J, & Stroup W (2006). Measuring the development of executive control with the shape school. Psychological assessment, 18(4), 373. [DOI] [PubMed] [Google Scholar]
- Espy KA, Clark CAC, Garza JP, Nelson JM, James TD, & Choi HJ (2016). VI. Executive control in preschoolers: New models, new results, new implications. Monographs of the Society for Research in Child Development, 81(4), 111–128. [DOI] [PubMed] [Google Scholar]
- Espy KA, Kaufmann PM, McDiarmid MD, & Glisky ML (1999). Executive functioning in preschool children: Performance on A-not-B and other delayed response format tasks. Brain and cognition, 41(2), 178–199. [DOI] [PubMed] [Google Scholar]
- Fujimoto K, & Valente TW (2012). Social network influences on adolescent substance use: disentangling structural equivalence from cohesion. Social Science & Medicine, 74(12), 1952–1960. 10.1016/j.socscimed.2012.02.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gaete J, Tornero B, Valenzuela D, Rojas-Barahona CA, Salmivalli C, Valenzuela E, & Araya R (2017). Substance use among adolescents involved in bullying: a cross-sectional multilevel study. Frontiers in psychology, 8, 1056. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Goldman PS, Rosvold HE, Vest B, & Galkin TW (1971). Analysis of the delayed-alternation deficit produced by dorsolateral prefrontal lesions in the rhesus monkey. Journal of comparative and physiological psychology, 77(2), 212. [DOI] [PubMed] [Google Scholar]
- Granvald V, & Marciszko C (2016). Relations between key executive functions and aggression in childhood. Child neuropsychology, 22(5), 537–555. [DOI] [PubMed] [Google Scholar]
- Gray KM, & Squeglia LM (2018). Research Review: What have we learned about adolescent substance use? Journal of Child Psychology and Psychiatry, 59(6), 618–627. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Griffin KW, & Botvin GJ (2010). Evidence-based interventions for preventing substance use disorders in adolescents. Child And Adolescent Psychiatric Clinics Of North America, 19(3), 505. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Harakeh Z, de Sonneville L, van den Eijnden RJ, Huizink AC, Reijneveld SA, Ormel J, … & Vollebergh WA (2012). The association between neurocognitive functioning and smoking in adolescence: The TRAILS study. Neuropsychology, 26(5), 541. [DOI] [PubMed] [Google Scholar]
- Hatta O, De Mol J, Maurage P, & Gabriel B (2018). Mediating effect of prosocial behavior between psychological difficulties and consumption of psychoactive substances in Togolese adolescents. Alcoology and Addictology, 40 (2), 131–139. [Google Scholar]
- Hay DF, Paine AL, Perra O, Cook KV, Hashmi S, Robinson C, … & Slade R (2021). Prosocial and aggressive behavior: a longitudinal study. Monographs of The Society For Research In Child Development, 86(2), 7–103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hofmann V, & Müller CM (2018). Avoiding antisocial behavior among adolescents: The positive influence of classmates’ prosocial behavior. Journal of adolescence, 68, 136–145. [DOI] [PubMed] [Google Scholar]
- Hughes C, Dunn J, & White A (1998). Trick or treat?: Uneven understanding of mind and emotion and executive dysfunction in “hard-to-manage” preschoolers. Journal of child psychology and psychiatry, 39(7), 981–994. [PubMed] [Google Scholar]
- Johnston LD, Miech RA, O'Malley PM, Bachman JG, Schulenberg JE, & Patrick ME (2021). Monitoring the Future National Survey Results on Drug Use, 1975–2020: Overview, Key Findings on Adolescent Drug Use. Institute for Social Research. [Google Scholar]
- Jones CB, Meier MH, Corbin WE, & Chassin L (2020). Adolescent executive cognitive functioning and trait impulsivity as predictors of young-adult risky drinking and alcohol-related problems. Psychology of Addictive Behaviors. [DOI] [PubMed] [Google Scholar]
- Keltner D, Kogan A, Piff PK, & Saturn SR (2014). The sociocultural appraisals, values, and emotions (SAVE) framework of prosociality: Core processes from gene to meme. Annual review of psychology, 65, 425–460. [DOI] [PubMed] [Google Scholar]
- Kim-Spoon J, Kahn RE, Lauharatanahirun N, Deater-Deckard K, Bickel WK, Chiu PH, & King-Casas B (2017). Executive functioning and substance use in adolescence: Neurobiological and behavioral perspectives. Neuropsychologia, 100, 79–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kim-Spoon J, Kahn RE, Lauharatanahirun N, Deater-Deckard K, Bickel WK, Chiu PH, & King-Casas B (2017). Executive functioning and substance use in adolescence: Neurobiological and behavioral perspectives. Neuropsychologia, 100, 79–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kochanska G, Murray KT, & Harlan ET (2000). Effortful control in early childhood: continuity and change, antecedents, and implications for social development. Developmental psychology, 36(2), 220. [PubMed] [Google Scholar]
- Kochanska G, Murray K, Jacques TY, Koenig AL, & Vandegeest KA (1996). Inhibitory control in young children and its role in emerging internalization. Child development, 67(2), 490–507. [PubMed] [Google Scholar]
- Korkman M, Kirk U, & Kemp SL (1998). NEPSY—A Developmental Neuropsychological Assessment. San Antonio, TX: Psychological Corporation. [Google Scholar]
- Korucu I, Selcuk B, & Harma M (2017). Self-regulation: Relations with theory of mind and social behaviour. Infant and Child Development, 26(3), 1–23. DOI: 10.1002/icd.1988 [DOI] [Google Scholar]
- Laible D, Carlo G, Davis AN, & Karahuta E (2016). Maternal sensitivity and effortful control in early childhood as predictors of adolescents’ adjustment: The mediating roles of peer group affiliation and social behaviors. Developmental Psychology, 52(6), 922. [DOI] [PubMed] [Google Scholar]
- Lewis C, & Carpendale JI (2009). Introduction: Links between social interaction and executive function. New Directions For Child And Adolescent Development, 2009(123), 1–15. [DOI] [PubMed] [Google Scholar]
- Longobardi E, Spataro P, & Rossi-Arnaud C (2019). Direct and indirect associations of empathy, theory of mind, and language with prosocial behavior: Gender differences in primary school children. The Journal of genetic psychology, 180(6), 266–279. [DOI] [PubMed] [Google Scholar]
- MacKinnon DP, Lockwood CM, & Williams J (2004). Confidence limits for the indirect effect: Distribution of the product and resampling methods. Multivariate behavioral research, 39(1), 99–128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Marsh HW, Hau KT, & Wen Z (2004). In search of golden rules: Comment on hypothesis-testing approaches to setting cutoff values for fit indexes and dangers in overgeneralizing Hu and Bentler's (1999) findings. Structural equation modeling, 11(3), 320–341. [Google Scholar]
- Matthews JS, Ponitz CC, Morrison FJ. Early gender differences in self-regulation and academic achievement. Journal of Educational Psychology 2009;101:689–704. [Google Scholar]
- McCabe SE, West BT, Veliz P, Frank KA, & Boyd CJ (2014). Social contexts of substance use among U.S. high school seniors: a multicohort national study. The Journal Of Adolescent Health, 55(6), 842–844. 10.1016/j.jadohealth.2014.06.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- McQuade JD (2017). Peer victimization and changes in physical and relational aggression: The moderating role of executive functioning abilities. Aggressive behavior, 43(5), 503–512. [DOI] [PubMed] [Google Scholar]
- McQuade JD, Murray-Close D, Shoulberg EK, & Hoza B (2013). Working memory and social functioning in children. Journal of experimental child psychology, 115(3), 422–435. [DOI] [PubMed] [Google Scholar]
- Memmott-Elison MK, Holmgren HG, Padilla-Walker LM, & Hawkins AJ (2020). Associations between prosocial behavior, externalizing behaviors, and internalizing symptoms during adolescence: A meta-analysis. Journal of adolescence, 80, 98–114. [DOI] [PubMed] [Google Scholar]
- Miyake A, Friedman NP, Emerson MJ, Witzki AH, Howerter A, & Wager TD (2000). The unity and diversity of executive functions and their contributions to complex “frontal lobe” tasks: A latent variable analysis. Cognitive psychology, 41(1), 49–100. [DOI] [PubMed] [Google Scholar]
- Muthén LK, & Muthén B (2017). Mplus user's guide: Statistical analysis with latent variables, user's guide. Muthén & Muthén. [Google Scholar]
- Ostrov JM, & Keating CF (2004). Gender differences in preschool aggression during free play and structured interactions: An observational study. Social development, 13(2), 255–277. [Google Scholar]
- Padilla-Walker LM, Memmott-Elison MK, & Coyne SM (2018). Associations between prosocial and problem behavior from early to late adolescence. Journal of youth and adolescence, 47(5), 961–975. [DOI] [PubMed] [Google Scholar]
- Peeters M, Janssen T, Monshouwer K, Boendermaker W, Pronk T, Wiers R, & Vollebergh W (2015). Weaknesses in executive functioning predict the initiating of adolescents’ alcohol use. Developmental cognitive neuroscience, 16, 139–146. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Peleg-Oren N, Cardenas GA, Comerford M, & Galea S (2012). An association between bullying behaviors and alcohol use among middle school students. The Journal of Early Adolescence, 32(6), 761–775. [Google Scholar]
- Pentz MA, & Riggs NR (2013). Longitudinal relationships of executive cognitive function and parent influence to child substance use and physical activity. Prevention Science, 14(3), 229–237. [DOI] [PubMed] [Google Scholar]
- Pilin MA, Pike J, Xie B, & Stacy AW (2020). Working Memory and Response Inhibition Effects on At-Risk Youth’s Willingness to Use Multiple Nicotine and Tobacco Products. Substance Use & Misuse, 55(11), 1867–1874. [DOI] [PubMed] [Google Scholar]
- Poland SE, Monks CP, & Tsermentseli S (2016). Cool and hot executive function as predictors of aggression in early childhood: Differentiating between the function and form of aggression. British journal of developmental psychology, 34(2), 181–197. [DOI] [PubMed] [Google Scholar]
- Quinn CA, Fitzpatrick S, Bussey K, Hides L, & Chan GC (2016). Associations between the group processes of bullying and adolescent substance use. Addictive behaviors, 62, 6–13. [DOI] [PubMed] [Google Scholar]
- Radliff KM, Wheaton JE, Robinson K, & Morris J (2012). Illuminating the relationship between bullying and substance use among middle and high school youth. Addictive behaviors, 37(4), 569–572. [DOI] [PubMed] [Google Scholar]
- Razza RA, & Blair C (2009). Associations among false-belief understanding, executive function, and social competence: A longitudinal analysis. Journal of Applied Developmental Psychology, 30(3), 332–343. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Riccio CA, Hewitt LL, & Blake JJ (2011). Relation of measures of executive function to aggressive behavior in children. Applied neuropsychology, 18(1), 1–10. [DOI] [PubMed] [Google Scholar]
- Riggs NR, Greenberg MT, & Rhoades B (2011). Early risk for problem behavior and substance use: Targeted interventions for the promotion of inhibitory control. In Inhibitory control and drug abuse prevention (pp. 249–262). Springer, New York, NY. [Google Scholar]
- Séguin JR (2009). The frontal lobe and aggression. European Journal of Developmental Psychology. 6, 100–119. doi: 10.1080/17405620701669871 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Simpson A, & Riggs KJ (2006). Conditions under which children experience inhibitory difficulty with a “button-press” go/no-go task. Journal of Experimental Child Psychology, 94(1), 18–26. [DOI] [PubMed] [Google Scholar]
- Squeglia LM, Jacobus J, Nguyen-Louie TT, & Tapert SF (2014). Inhibition during early adolescence predicts alcohol and marijuana use by late adolescence. Neuropsychology, 28(5), 782. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Steinberg L (2008). A social neuroscience perspective on adolescent risk-taking. Developmental Review, 28(1), 78–106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Thomson ND, & Centifanti LC (2018). Proactive and reactive aggression subgroups in typically developing children: The role of executive functioning, psychophysiology, and psychopathy. Child Psychiatry & Human Development, 49(2), 197–208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tominey SL, & McClelland MM (2011). Red light, purple light: Findings from a randomized trial using circle time games to improve behavioral self-regulation in preschool. Early Education & Development, 22(3), 489–519. [Google Scholar]
- Valente MJ, Gonzalez O, Miočević M, & MacKinnon DP (2016). A note on testing mediated effects in structural equation models: Reconciling past and current research on the performance of the test of joint significance. Educational and Psychological Measurement, 76(6), 889–911. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wiebe SA, Sheffield T, Nelson JM, Clark CA, Chevalier N, & Espy KA (2011). The structure of executive function in 3-year-olds. Journal Of Experimental Child Psychology, 108(3), 436–452. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Williams KE, & Berthelsen D (2017). The development of prosocial behaviour in early childhood: Contributions of early parenting and self-regulation. International Journal of Early Childhood, 49(1), 73–94. [Google Scholar]
- Zhou Q, Chen SH, & Main A (2012). Commonalities and differences in the research on children’s effortful control and executive function: A call for an integrated model of self-regulation. Child development perspectives, 6(2), 112–121 DOI: 10.1111/j.1750-8606.2011.00176.x. [DOI] [Google Scholar]