Relational aggression, big five and hostile attribution bias in adolescents

Journal of Applied Developmental Psychology 52 (2017) 101–113

Contents lists available at ScienceDirect

Journal of Applied Developmental Psychology journal homepage: www.elsevier.com/locate/jappdp

Relational aggression, big five and hostile attribution bias in adolescents

MARK

⁎

Constantinos M. Kokkinos , Kyriaki Karagianni, Ioanna Voulgaridou Department of Primary Education, School of Education Sciences, Democritus University of Thrace, Greece

A R T I C L E I N F O

A B S T R A C T

Keywords: Relational aggression Proactive Reactive Hostile attribution bias Five factor model Mediation Moderation Adolescents

Τhe study examined two plausible models of the interplay between hostile attribution bias (HAB) for relational provocations, personality traits (big five), and the two distinct functions of relational aggression (RA) in a sample of 347 Greek adolescents (193 female participants; M = 13.1, SD = 1.20). In the first model the mediating role of HAB in the association between personality and RA was investigated. The moderating effects of personality in the association between HAB and RA were also explored. Results of the mediation analyses revealed that HAB partially mediated the relationship between low extraversion, conscientiousness, agreeableness, openness, high neuroticism, and both proactive and reactive RA. Moderation analyses indicated that low conscientiousness adolescents were more likely to engage in reactive RA when they make more hostile attributions for relational provocations. The findings emphasize the crucial role of personality and social-cognitive factors in understanding RA and are discussed in terms of their practical implications.

Over the recent years there has been an increasing interest in understanding nonphysical and manipulative forms of aggression, including relational aggression (RA; damaging or intent of harming others mainly via the purposeful manipulation or threat of interpersonal relationships, social exclusion, gossiping or rumor spreading; Card, Stucky, Sawalani, & Little, 2008; Crick & Grotpeter, 1995). This aggressive form could be either proactive (i.e., unprovoked, goal-oriented, related to resource control, gaining in popularity and social status) or reactive RA (i.e., displayed in response to threats, related to negative affect; Crapanzano, Frick, & Terranova, 2010; Marsee, Weems, & Taylor, 2008; Marsee & Frick, 2007; Voulgaridou & Kokkinos, 2015). It could also be direct/confrontational (e.g., excluding a peer) or non-direct/ non-confrontational (e.g., spreading rumors) (Archer & Coyne, 2005; Voulgaridou & Kokkinos, 2015). The study of aggressive behavior is guided by several theoretical perspectives that emphasize different aspects ranging from the social context to biological predispositions. The General Aggression Model (GAM), proposed by Anderson and Bushman (2002), is the result of endeavors to integrate existing theories of aggression, such as the Social Learning Theory (Bandura, 1977); Berkowitz (1993) and the Social Interaction Theory of Tedeschi and Felson (1994). According to the GAM, personal (e.g., personality traits, beliefs, gender, attitudes) and situational factors (e.g., frustration, provocation) could influence aggression through their impact on an individual's internal state (i.e., aggressive thoughts, angry feelings and physiological arousal) (Anderson & Bushman, 2002). Repeated interaction with aggressionrelated stimuli (both real and fictitious) and situations, and subsequent ⁎

positively reinforced aggressive behavior, is likely to increase one's aggressive personality through several learned outcomes (e.g., aggressive beliefs, attitudes, and related emotions). Using the GAM as a theoretical framework, this study examines a mediation model of personal and social-cognitive constructs to inform our understanding of RA in adolescence, a period of development where the use of more complex and sophisticated aggressive tactics is evident (Crick, MurrayClose, & Woods, 2006; Voulgaridou & Kokkinos, 2015; Yoon, Barton, & Taiariol, 2004). Although mediating relations between personality and cognitive factors are possible, some researchers have found evidence that personality and cognition might also interact to predict behavior outcomes (e.g., Fite, Goodnight, Bates, Dodge, & Pettit, 2008; Koolen, Poorthuis, & van Aken, 2012; Meece, Mize, Bates, Dodge, & Pettit, 2007). Following this contention, the present study integrates two theoretical models, the relational vulnerability model (Crick, MurrayClose, & Woods, 2004) and the Integrative Cognitive Model (ICM; Wilkowski & Robinson, 2010), in order to explicate the possible moderating role of personality (five factor model; FFM) in the relationship between social-cognitive factors (i.e., hostile attribution bias; HAB) and RA. Empirical evidence provides explanations for the individual factors (e.g., personality, cognition) which are implicated in the manifestation of RA (e.g., Dane & Marini, 2014; Kokkinos, Voulgaridou, & Markos, 2016; Miller, Zeichner, & Wilson, 2012; Tackett, Kushner, Herzhoff, Smack, & Reardon, 2014). A useful contribution to the study of the social-cognitive correlates of relationally

Corresponding author at: Department of Primary Education, School of Education Sciences, Democritus University of Thrace, N. Hili, GR 68100, Alexandroupolis, Greece. E-mail address: [email protected] (C.M. Kokkinos).

http://dx.doi.org/10.1016/j.appdev.2017.07.007 Received 9 March 2016; Received in revised form 31 May 2017; Accepted 22 July 2017 0193-3973/ © 2017 Elsevier Inc. All rights reserved.

Journal of Applied Developmental Psychology 52 (2017) 101–113

C.M. Kokkinos et al.

As far as N is concerned, high levels of this trait and its facets (i.e., negative affect and fear) are linked to RA and mainly to its reactive function among children, adolescents, and adults (Bettencourt, Talley, Benjamin, & Valentine, 2006; Burt et al., 2012; Burton et al., 2007; Seibert et al., 2010; Tackett et al., 2013; Tackett et al., 2014). Moreover, anxiety and anger (N facets) are significantly associated more with reactive than proactive RA (Marsee & Frick, 2007; Marsee et al., 2008). Finally, the relationship between E and RA has not been confirmed in children, adolescents (e.g., Gleason et al., 2004; Tackett et al., 2013), and emerging adults (Burton et al., 2007; Miller et al., 2012). However, some facets of E, such as sociability and shyness were found to be positively linked to RA (Tackett et al., 2014). It has also been found that E, as well as its facet assertiveness, is more strongly, positively related to proactive aggression (Miller & Lynam, 2006; Ramirez, 2009). Relational aggression has been also related to personality pathology (e.g., callous unemotional traits, borderline personality characteristics) in middle childhood and early adolescence (Crick, MurrayClose, & Woods, 2005; Kokkinos et al., 2016; Marsee & Frick, 2007). Additionally, research with emerging adults has shown associations among borderline personality symptoms and RA (Ostrov & Houston, 2008; Werner & Crick, 1999). In terms of the RA functions (i.e., proactive and reactive), Ostrov and Houston (2008) revealed that both were associated with borderline personality even after controlling for physical aggression and gender. Additionally, Crick et al. (2005) showed that both RA and HAB for relational provocations were predictive of borderline personality in middle childhood. It seems also that borderline personality and HAB may have some conceptual links, since hostile world view (which may appear in children lacking social information skills; Nelson, Coyne, Swanson, Hart, & Olsen, 2014) is an indicator of borderline personality (Crick et al., 2005; Ostrov & Houston, 2008). Moreover, it has been empirically supported that the FFM could be viewed as a dimensional model of personality disorder symptomatology (Lynam & Widiger, 2001). Thus, it is suggested that borderline personality disorder includes facets of high N, impulsivity, low C, low A, and interpersonal antagonism (Miller et al., 2010). Hence, the previous literature referring to personality pathology could provide further support for a possible association among some big five traits (i.e., A, N, C), HAB for relational provocations and RA. Altogether, the existing literature shows evidence for robust associations between social-cognition, personality and RA, but the potential interactive contributions of cognitive factors and personality to the explanation of RA have not yet been empirically investigated, therefore pointing to the purpose of this study.

aggressive behavior (Mathieson et al., 2011; Nelson & Coyne, 2009) is made by the Social Information Processing (SIP) Model (Crick & Dodge, 1994), which refers to the encoding and interpretion of a social situation, setting goals and detecting the ways to achieve them and then evaluating, selecting and applying a response. Hostile attribution bias (i.e., the tendency to attribute hostile intent to others' actions even if their real purpose is benign or the circumstances are ambiguous), which relates to the interpretation of social stimuli in the SIP model, was found to be associated to aggression (Dodge et al., 2015; Helfritz-Sinville & Stanford, 2014; Nelson, Mitchell, & Yang, 2008). In terms of RA, empirical evidence is inconclusive (Godleski & Ostrov, 2010; Mathieson et al., 2011; Ostrov & Godleski, 2013) with a number of studies suggesting that relationally aggressive early adolescents demonstrate higher levels of HAB for relational provocations (Crick, 1995; Crick, Grotpeter, & Bigbee, 2002; Ostrov & Godleski, 2013; Yeung & Leadbeater, 2007). Other studies, however, with primary school children did not confirm this association or even support a negative link (Crain, Finch, & Foster, 2005; Mikami, Lee, Hinshaw, & Mullin, 2008; Nelson et al., 2008). A possible explanation for the lack of an association between HAB and RA may be that some children use proactive instead of reactive RA, which seems more likely to be linked to HAB (Mathieson et al., 2011). Although recent evidence with emerging adults (Bailey & Ostrov, 2008) has shown that HAB for relational provocations is associated only with reactive RA (Murray-Close, Ostrov, Nelson, Crick, & Coccaro, 2010), these findings were not confirmed for adolescents (Kunimatsu, 2010; Marsee & Frick, 2007). Adolescent personality – particularly traits reflecting tendencies toward negative emotions (e.g., neuroticism) and poor self-regulatory control (e.g., low trait agreeableness and conscientiousness) – has been strongly associated with aggressive behaviors, including RA (e.g., Tackett, Daoud, De Bolle, & Burt, 2013; Tackett et al., 2014). However, the examination of the links between personality and RA using different personality constructs, such as the FFM has only recently received empirical investigation (e.g., Burt, Donnellan, & Tackett, 2012; Burton, Hafetz, & Henninger, 2007; Gleason, Jensen-Campbell, & South Richardson, 2004;Miller et al., 2012 ; Tackett et al., 2014). The FFM is the most contemporary and widely accepted comprehensive model of personality structure that describes personality as a set of five broad trait dimensions or domains (DeYoung, Quilty, & Peterson, 2007), that is Agreeableness (A; the tendency to care about others' needs and the readiness to keep positive interpersonal relationships), Conscientiousness (C; trustworthiness/reliability, accuracy and commitment fulfillment), Neuroticism/Emotional Instability (N/EI; negative thoughts, feelings of anxiety, depression and anger), Extraversion (E; levels of energy, excitement, assertiveness, self-confidence), and Openness to Experience (O; creativity, imagination, intellect and depicts the levels of social and cultural interest). Regarding the relationship between RA and FFM traits, it has been found that children and adolescents who tend to engage in such behaviors are more likely to show lower A, C (Gleason et al., 2004; Tackett et al., 2013; Tackett et al., 2014), and in some cases O (Tackett et al., 2014). It seems also that from these traits, A may have stronger predictive utility compared to the other FFM traits (Tackett et al., 2014), while the connections of RA with C (Miller et al., 2012; Tackett et al., 2014) and O (e.g., Gleason et al., 2004; Miller et al., 2012) in children, adolescents or emerging adults are not always confirmed. It has also been indicated that A is negatively linked to both proactive and reactive aggression (Koolen et al., 2012; Miller et al., 2012; Seibert, Miller, Pryor, Reidy, & Zeichner, 2010). Regarding C, a negative association only with reactive aggression has been confirmed, suggesting that it is related to the lack of behavioral reactivity regulation in case of experiencing frustration (Koolen et al., 2012); nevertheless, other studies did not evidence such differences (Miller & Lynam, 2006; Miller et al., 2012). Similarly, a facet of O (i.e., openness to actions) was found to be significantly correlated with reactive aggression (Miller & Lynam, 2006).

1. FFM and RA: the mediating role of HAB Prior research has mainly demonstrated direct links of HAB (e.g., Ostrov & Godleski, 2013) and FFM traits (e.g., Tackett et al., 2014) with RA. It has been argued that someone's personality might contribute to more biased interpretations, which in turn could lead to aggressive behaviors (Bettencourt et al., 2006; Miller, Lynam, & Jones, 2008). In an endeavor to understand aggressive behavior based on the GAM, recently published studies (Barlett & Anderson, 2012; Cavalcanti & Pimentel, 2016) showed that FFM traits were directly and indirectly related to aggression, through aggressive attitudes. Specifically, Barlett and Anderson (2012) suggested that in case of facing hostile provocations, individuals characterized by certain personality traits (e.g., low A), are more likely to display HAB, as well as form hostile patterns of social information processing, and thus be led to aggressive behavior. Depending on certain FFM traits, the GAM would suggest that these traits are related to aggressive behavior because they may either enhance or inhibit the development and chronic accessibility of aggressive emotions and attitudes (Barlett & Anderson, 2012; Cavalcanti & Pimentel, 2016). Thus, based on the GAM, the present study examined a mediation model with five personality factors as predictors of RA and HAB as the mediator. 102

Journal of Applied Developmental Psychology 52 (2017) 101–113

C.M. Kokkinos et al.

2. HAB and RA: the moderating role of FFM

As far as RA is concerned, some FFM traits are linked to tendencies to attribute hostile intents (Koolen et al., 2012; Miller et al., 2008), which in turn could put individuals at risk for RA (Werner, 2012), particularly its reactive subtype (Murray-Close et al., 2010). Recent work has also provided support for the interactive links among personality, social-cognition, and RA suggesting that personality traits could affect RA through social-cognitive mechanisms (Kokkinos, Voulgaridou, Mandrali, & Parousidou, 2016). In accordance with this evidence, the present study examines the mediating role of HAB in the association between FFM traits and RA (Model 1). Compelling reasons hold that A is negatively associated with tendencies to assume the worst scenario (relevant to hostile attributions) among children (Koolen et al., 2012) and HAB among emerging adults (Miller et al., 2008). Seibert et al. (2010) claimed that antagonistic individuals make more hostile attributions and rely on more aggressive actions, instead of employing prosocial strategies. Additionally, Gleason et al. (2004) suggested that social-cognitive factors could be the principal path through which A as well as tendencies to keep positive relationships are linked to specific behaviors. For instance, it has been suggested that antagonism (one of the poles of trait A; Caspi, Roberts, & Shiner, 2005) and disagreeableness may predispose individuals to HAB, making more possible aggressive responses (e.g., reactive RA) instead of concession or prosocial behavior (Barlett & Anderson, 2012; Jones, Miller, & Lynam, 2011; Miller et al., 2012). This association could be attributed to the propensity of individuals high in HAB to focus more on hostile cues and consequently form hostile patterns of social information processing (Barlett & Anderson, 2012). Based on these findings and the reasoning that A is the FFM dimension most closely related to peer relations (Miller et al., 2008), it appears theoretically relevant to explore the mediating role of HAB in the link between A and RA. As far as C is concerned, its links with HAB are less explicit (Koolen et al., 2012; Miller et al., 2008). However, a significant overlap of C with self-control was found (i.e., referring to skills such as delaying gratification, effortful control, determination and self-regulation; Duckworth, 2011; Koolen et al., 2012), which is negatively associated with HAB (Nelson & Perry, 2015). Conscientiousness (Gleason et al., 2004; Tackett et al., 2013) and HAB (Ostrov & Godleski, 2013) have been also linked to RA with some findings suggesting associations of these factors (i.e., C, HAB) with reactive aggression (Koolen et al., 2012; Murray-Close et al., 2010). Thus, it is likely that HAB could serve as a mediator between C and RA. Fewer links have been documented between N and HAB. Specifically, the impact of N on the interpretation of cues is based on the theoretical framework of “reactive interaction pattern” (Caspi & Roberts, 2001), which portrays the different ways in which individuals interpret and react in specific situations (Hessels, van den Hanenberg, de Castro, & van Aken, 2014). This pattern reveals that high N adolescents may interpret a situation as more frustrating (Hessels et al., 2014) or make more irrational thoughts (Bettencourt et al., 2006). Based on this evidence and the suggestions that individuals high in N are more likely to perceive their environment as threatening (Caspi et al., 2005) and tend to react aggressively in provocations (Seibert et al., 2010), it seems possible that high N adolescents may perceive a provocation in ambiguous situations and thus exhibit reactive RA. In this direction, Marsee et al. (2008) found that social-cognitive errors mediated the link between anxiety (a facet of N; DeYoung et al., 2007) and reactive RA. The role of E and O in the emergence of HAB has not received considerable research attention. Moreover, no significant associations have been found between E and RA, except for some of its facets (i.e., socialibility, shyness, positive emotions; Tackett et al., 2014), while limited studies found relations between O and RA (Tackett et al., 2014). Therefore, because these relations have not held up consistently in the studies reviewed, E and O will be included and analyzed in the models in an exploratory manner.

Theoretical and empirical evidence suggests that the associations between cognitive distortions (including tendencies to attribute hostile intentions; Pornari & Wood, 2010) and aggression are differentiated depending on personality traits (Fite et al., 2008; Koolen et al., 2012; Meier & Robinson, 2004). It has been recommended that the modest relationships, often found between SIP or specifically HAB and aggressive behavior (Fite et al., 2008; Orobio de Castro, Veerman, Koops, Bosch, & Monshouwer, 2002), could be attributed to the moderating effects of other factors on this association. Although Fite et al. (2008) claimed that the relationship between SIP and aggression might be evident only in some children, limited studies have examined these relationships. Thus, the present research addressed this gap in the literature by examining the moderating role of adolescents' personality traits in the association between HAB and RA. Research studying the moderating role of personality traits in the relationship between social-cognitive processes and aggressive behavior focuses on certain characteristics associated with self-regulation and effortful control (e.g., A, C) (Fite et al., 2008; Koolen et al., 2012). Traits A and C are related to effortful control, which refers to abilities to plan behavior, form and shift attention and suppress a dominant behavioral response to endorse a subdominant (Ahadi & Rothbart, 1994; Koolen et al., 2012). Specifically, A is related to self-inhibition and control (Ahadi & Rothbart, 1994), while C to self-regulatory skills such as behavioral inhibition and perseverance (Koolen et al., 2012). The moderating role of these personality traits in the relationship between HAB and reactive RA could be interpreted by the Integrative Cognitive Model (ICM; Wilkowski & Robinson, 2010). In general, the ICM suggests that cognitive processing tendencies are involved in the relationship between hostile situational input (e.g., insult, provocation) and responses, such as anger and reactive aggression. The model proposes that the first process related to understanding trait anger and reactive aggression is the interpretation of situational input and this may include biased hostile interpretations. Additionally, the model claims that ruminative processes could strengthen interpretation-related biases, and thus reinforce anger and reactive aggression. The ICM further suggests that effortful control processes, related to emotion regulation, could buffer against initial tendencies toward anger and reactive aggression. The effortful control processes may lead to reappraisal of situational input in order to promote a non-hostile interpretation. It could also inhibit ruminative processes and hostile thoughts, as well as constrain the tendency to react aggressively. Consequently, based on the ICM, hostile attributions, ruminative attention, and effortful control could interactively reinforce or inhibit trait anger and reactive aggression (Wilkowski & Robinson, 2010). Several hypotheses could be formulated, based on the ICM. Specifically, individuals who make hostile interpretations and display low levels of effortful control (i.e., low A and C) are more likely to engage in reactively aggressive behaviors, such as reactive RA (Wilkowski & Robinson, 2010). On the contrary, individuals who attribute hostile intentions to others and score high in effortful control (i.e., high A and C) may be more protected against reactive aggression (Wilkowski & Robinson, 2010). It seems therefore possible that selfregulatory skills, high self-control (high A and C), and children's abilities to calm themselves down could protect them against aggression in case of exhibiting HAB (Koolen et al., 2012; Meece et al., 2007). In terms of N, there is only indirect evidence indicating that negative emotions may affect the association between HAB and RA. The moderating role of N in the relationship between HAB and RA seems theoretical relevant to the relational vulnerability model. The model was proposed by Crick et al. (2004) to shed light on the paths and risk factors through which HAB for relational provocations could be associated with RA. Specifically, the model proposes that emotional sensitivity to relational provocations (i.e., feelings of distress in response to perceived provocations) may be a possible moderator between HAB and 103

Journal of Applied Developmental Psychology 52 (2017) 101–113

C.M. Kokkinos et al.

RA. As RA could be a response to perceived provocations in the peer context, both cognitive and emotion processes in response to such provocations, could predispose to RA (Mathieson et al., 2011). Consequently, the model proposes that HAB would lead to RA only among children who exhibit vulnerabilities such as emotional sensitivity (Crick, Geiger, & Zimmer-Gembeck, 2003; Crick et al., 2004). Indeed, Mathieson et al. (2011) provided evidence that emotionally distressed children may be more vulnerable to RA, in case of attributing hostile intentions to others. Considering that high N individuals may be more likely to perceive life events as stressful and that N has been positively associated with negative emotional sensitivity (i.e., tendency to experience a self-oriented negative emotionality; Guarino, Roger, & Olason, 2007), the hypothesis that individuals high in N may be more likely to engage in RA, in case of displaying HAB, seems theoretically possible. In the same vein, it has been shown that increased anxiety and anger, experienced by high N individuals (DeYoung et al., 2007), are likely to strengthen the association between HAB and aggression (Fite et al., 2008; Runions & Keating, 2010). Hence, it would be expected that high N children are more likely to engage in RA when they experience hostile provocations, and especially in reactive RA, since there is evidence supporting that anxiety (e.g., Marsee et al., 2008) and HAB (e.g., Murray-Close et al., 2010) are uniquely linked to this dimension. To understand the relationship between personality traits, cognition and aggression, Koolen et al. (2012) hypothesized that low C and A children are more likely to become reactively aggressive in case of assuming the worst scenario, which pertains to tendencies to make hostile attributions (Pornari & Wood, 2010). Furthermore, high A individuals may possess interpersonal skills which could buffer against the results of biases in social-cognitive processes (Hessels et al., 2014) and they may be more able to activate helpful thoughts in hostile situations, which may contribute to the avoidance of aggressive behavior (Meier, Wilkowski, & Robinson, 2008). Thus, it seems theoretically relevant that children high in A and C may be less engaged in aggression (overt or relational), especially in its reactive functional subtype, even when they perceive hostile cues in ambiguous situations. In terms of E, its role in the association between HAB and RA seems rather unclear. For example, impulsiveness which is associated with a facet of E (i.e., excitement seeking; Mullins-Sweat & Widiger, 2006), was found to strengthen the link between HAB and aggression, suggesting that individuals who demonstrate HAB are likely to react aggressively, when they are highly impulsive, since they may not evaluate different responses or the consequences of their actions (Chen, Coccaro, & Jacobson, 2012). Adversely, positive emotions (i.e., facet of E) were found to negatively predict RA (Tackett et al., 2014). This trait, which has been negatively linked to negative emotional sensitivity (Guarino et al., 2007), would seem also more likely to weaken the relationship between HAB for relational provocations and RA, based on the relational vulnerability model. Finally, no significant evidence exists that O could affect the association between HAB and RA. This trait did not fall into the category of FFM traits examined as moderators (Koolen et al., 2012) nor seems to be related to factors (i.e., impulsiveness, self-control skills, negative feelings) implicated in this association (Chen et al., 2012; Mathieson et al., 2011).

some studies claim that girls may be more involved in RA (e.g., Crick & Grotpeter, 1995), especially reactive, compared to boys (Marsee et al., 2011), others indicated higher levels of RA among boys (e.g., Kokkinos et al., 2016; Yeung & Leadbeater, 2007) for both of its functions (Kokkinos & Voulgaridou, 2017; Vagos, Rijo, Santos, & Marsee, 2014) or no gender differences at all (e.g., Putallaz et al., 2007; Willoughby, Kupersmidt, & Bryant, 2001) for both proactive and reactive RA (Archer, 2004; Card et al., 2008; Marsee et al., 2014). The present study will also examine the within gender differences regarding the aggressive forms (i.e., relational and overt). According to the gender-linked model of aggression proposed by Ostrov and Godleski (2010), children are theorized to prefer to use gender-consistent aggressive behaviors and avoid the display of gender-inconsistent aggressive acts. From this perspective, this model posits that girls will more frequently select relationally than overt aggressive behaviors, whereas boys the opposite (Murray-Close et al., 2016). Indeed, this prediction has been partially supported in prior research (Crapanzano et al., 2010; Murray-Close et al., 2016; Putallaz et al., 2007). In the case of the within gender differences regarding the functions of overt and relational aggression prior findings are not clear. Thus, there is limited evidence supporting that female adolescents are more likely to report more both overt and relational proactive compared to reactive aggression (Marsee & Frick, 2007; Vitaro, Brendgen, & Tremblay, 2002), whereas male adolescents seem to be more likely to engage in reactive than proactive aggression as they are more impulsive and reactive to provocations in their social relations and prefer direct retaliation (Knyazev, 2004). In another study, Fung, Raine, and Gao (2009) reported significant gender differences on proactive aggression, with female adolescents being less proactively aggressive than male adolescents, while no gender differences were found in the case of reactive aggression. Finally, other studies (i.e., Baş & Yurdabakan, 2012; Connor, Steingard, Anderson, & Melloni, 2003) found no gender differences in both reactive and proactive aggressive behavior. As far as HAB for relational provocations is concerned, empirical findings attest that girls demonstrate significantly higher HAB than boys (Godleski & Ostrov, 2010; Mathieson et al., 2011) and consider relational provocations as more distressing (Crick, 1995; Crick et al., 2002; Godleski & Ostrov, 2010). Several longitudinal studies among adolescents have demonstrated gender differences in FFM traits. Thus, according to some findings, girls show higher levels of A (Klimstra, Hale, Raaijmakers, Branje, & Meeus, 2009; McCrae et al., 2002) and C (Klimstra et al., 2009). However, these differences are, in some cases, detected only in early to middle adolescence (Klimstra et al., 2009) or are not generally confirmed (Branje, Van Lieshout, & Gerris, 2007). As far as N is concerned, boys often demonstrate a more advanced pattern, since they score lower in this dimension (Klimstra et al., 2009; McCrae et al., 2002; Pullmann, Raudsepp, & Allik, 2006), nonetheless, in some cases such differences are not confirmed (Branje et al., 2007). In terms of O and E, findings seem controversial as some of them reveal higher levels in girls (e.g., McCrae et al., 2002), other in boys (e.g., Branje et al., 2007) or do not detect any gender differences at all (e.g., Klimstra et al., 2009). However, regarding O, although the latter study did not confirm differentiations in middle to late adolescence, it indicated girls' prevalence in early to middle adolescence.

3. Gender differences

4. The present study

Currently, research has shown inconsistent evidence regarding gender differences in the function of aggression (i.e., proactive or reactive) as well as in the form of aggression (i.e., relational or overt) that each gender selects (Murray-Close, Nelson, Ostrov, Casas, & Crick, 2016; Voulgaridou & Kokkinos, 2015). Thus, the present study sought to investigate both between and within gender differences in the functions of both relational and overt aggression. As far as the between gender differences are concerned, although

Given the conflicting findings regarding the relationship between HAB for relational provocations and RA and the limited number of studies which have examined the association between big five traits and RA in children and adolescents, this study aims at further exploring the aforementioned linkages, in order to provide insights in the intrapersonal factors involved in the manifestation of such behaviors. As far as the relationship between HAB and RA is concerned, although the empirical evidence seems contradictory, a positive association between 104

Journal of Applied Developmental Psychology 52 (2017) 101–113

C.M. Kokkinos et al.

Fig. 1. Mediation Model (HAB mediates the association between big five and RA).

HAB

FFM traits Low A, C, E, O High N

RA

(193 female and 154 male), aged between 11 and 16 years (Μ = 13.1, SD = 1.20), selected from urban public primary and junior high schools from Northern Greece, participated in the study. Ninety-five of the participants (27%) were attending the last grade of the primary school, whereas the rest 252 (73%) the three grades of the junior high school. That is, 105 (30.3%) the first, 81 (23.3%) the second and 66 (18.7%) the third grade. The sample was recruited mostly from urban public schools, while most of the participants came from middle socio-economic circumstances (parents were primarily civil servants).

these factors is expected, particularly for reactive RA (e.g., MurrayClose et al., 2010; Ostrov & Godleski, 2013; Werner, 2012; Yeung & Leadbeater, 2007). Moreover, it could be hypothesized that A, C and O will be negatively associated with RA (e.g., Gleason et al., 2004; Tackett et al., 2014), N positively (Tackett et al., 2013; Tackett et al., 2014), while E will not be significantly linked to the manifestation of this aggressive form (e.g., Gleason et al., 2004; Tackett et al., 2013). No prior research, to our knowledge, has examined the linkages between big five factors and the distinct functions of RA. However, considering the literature regarding proactive and reactive aggression (e.g., Bettencourt et al., 2006; Koolen et al., 2012; Miller & Lynam, 2006), it is hypothesized that A would be negatively correlated to both proactive and reactive RA, C and O would be more strongly and negatively associated with reactive RA, and lastly, N would be positively linked to reactive RA. Finally, to further clarify the underlining processes which are implicated in the associations among HAB, big five factors and RA, two models will be tested (Figs. 1 & 2). The first examines whether HAB constitutes the mechanism through which some personality traits predict different types of RA (Fig. 1). Given that some FFM traits might contribute to hostile attribution tendencies (Bettencourt et al., 2006; Miller et al., 2008), which are likely to be associated with RA (Werner, 2012), it is expected that some personality characteristics (i.e., low A, low C, high N) indicative of low self-control, as well as high anxiety and distress could predispose to HAB, which in turn may be linked to RA, especially to its reactive subtype. On the other hand, given that the relationship between specific cognitive processes or distortions and aggression may be evident only in specific groups of children (e.g., Fite et al., 2008; Koolen et al., 2012; Meier & Robinson, 2004) the second model extends previous research by investigating whether the link between HAB and RA is more significant for those who are characterized by specific personality traits (Fig. 2). To develop a moderation hypothesis, it is predicted that low levels of A and C, and high N will enhance the impact of HAB on RA, which, according to the literature, is primarily expected for its reactive function.

5.2. Measuring instruments The Greek translations of the scales were developed using the front and back translation method with the assistance of two bilingual psychologists, who further resolved issues regarding differences in translation through consensus. All translated versions maintained the original scoring system. Finally, the scales were pilot tested with a small sample bearing the same characteristics as the main study's sample for item comprehensibility. 5.3. Aggression The Peer Conflict Scale (PCS; Marsee & Frick, 2007), a 40-item selfreport measure developed to assess the various types of aggressive behaviors was used. It includes four 10-item subscales. The two reactive subscales, Reactive-Overt (e.g., “If others make me mad, I hurt them”) and Reactive-Relational (e.g., “If others make me mad, I tell their secrets”) include items worded as such that the individual was clearly provoked, and the reaction is either to hurt or fight the other person (overt) or to harm their social relationships (relational). In contrast the Proactive-Overt subscale (e.g., “I carefully plan out how to hurt others”) also involves hurting others or fighting, but in a way that is clearly planned or for gain. Similarly, the Proactive-Relational subscale (e.g., “I deliberately exclude others from my group, even if they haven't done anything to me”) involves hurting others socially but again in a way that is clearly not in reaction to a perceived provocation. Each item was scored either 0 (Not at all true), 1 (Somewhat true), 2 (Very true), or 3 (Definitely true). A factor analysis of the PCS with adolescents has supported the conceptual differences (i.e., overt and relational forms, proactive and reactive functions of aggression) of the PCS content (Marsee et al., 2011).

5. Method 5.1. Participants A convenience sample of 347 predominantly adolescents students

HAB

RA

FFM traits (A, C, N)

105

Fig. 2. Moderation Model (FFM traits moderate the relationship between HAB and RA).

Journal of Applied Developmental Psychology 52 (2017) 101–113

C.M. Kokkinos et al.

of the Greek Ministry of Education, Research and Religious Affairs. A combination of active assent by the students and passive consent by their parents and/or guardians was obtained as it was requested by the terms of the research permit. Thus, school principals were contacted and students were provided with a letter that briefly explained the purpose of the study, emphasizing that participation was voluntary and anonymous, that participants were free to withdraw from the study at any time, and that the information collected was confidential. Students were instructed to hand in the letter to their parents and/or guardians and to return it only if they did not want them to participate. Passive consent from the participants' parents and/or guardians was deemed sufficient because no sensitive personally identifying data was collected, implying that consent was given by answering the questionnaire. Sixteen parents of a total of 363 contacted denied participation. On the day of the administration, students were informed about the purpose of the study and were asked in case they did not want to participate (i.e., active assent). None of them refuted participation. The questionnaires were group administered during a class period by trained researchers (i.e., the second and third authors). Teachers were present during the questionnaire completion time which took approximately 40 min.

The construct validity of the measure was assessed with Confirmatory Factor Analysis (CFA) using a robust Maximum Likelihood estimation method with the Satorra-Bentler chi-square (SBχ2) correction implemented in Mplus version 6.1 (Muthén & Muthén, 2015). Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), Root Mean Square Error of Approximation (RMSEA) and the Standardized Root Mean Square Residual (SRMR) were used to determine model fit. A CFI > 0.90, TLI > 0.90, RMSEA < 0.05 and SRMR < 0.08 were considered acceptable (Hu & Bentler, 1999). For the present sample, CFA revealed a 4-factor model which provided a good fit to the data, SB-χ2 = 74.60, df = 32, CFI = 0.93, TLI = 0.90, SRMR = 0.058, RMSEA = 0.043 (0.024–0.071). The internal consistency coefficient for this sample was 0.89 for the proactive and 0.86 for the reactive relational aggression; 0.91 and 0.87 for the proactive and reactive overt aggression subscales respectively. 5.4. Hostile attribution bias Hostile attribution biases regarding relationally-toned provocations were based on participants' responses to the Intent Attribution Measure (Crick, 1995; Crick et al., 2002). The instrument includes five hypothetical situations that depict relational peer conflicts (e.g., not getting invited to a friend's birthday party), all of which describe the intent of the peer in ambiguous terms. For each situation, participants were presented with two questions assessing intent attributions. In the first question, they were presented with four possible reasons for provocation and were asked to choose the most possible one. Two of the reasons portrayed benign (e.g., the friend didn't invite me to the birthday party because s/he was planning to invite me later), and two hostile intent (e.g., the friend didn't invite me to the birthday party because s/he was trying to get back at me for something). In the second question, participants indicated whether they thought the provocateur was trying to be mean (i.e., hostile intent) or not (i.e., benign intent). Responses to the two questions were summed within each story, and these scores were summed together to produce a total intent attribution score for relational provocations (possible range from 0 to 10). Confirmatory factor analysis replicated the uni-dimensionality of the scale, SBχ2 = 71.843, df = 31, CFI = 0.95, TLI = 0.93, SRMR = 0.050, RMSEA = 0.032 (0.021–0.050). The instrument has been shown to be highly reliable in previous research (Mathieson et al., 2011; Yeung & Leadbeater, 2007), although in the present study a fairly acceptable α = 0.69 was obtained.

7. Analysis plan Descriptive statistics were calculated for all measures. A series of independent t-tests examined potential gender differences in RA, HAB and personality traits. Further, paired samples t-test explored withingroup gender differences for proactive and reactive RA. Pearson correlation coefficients evaluated the associations among these variables. For the first conceptual model, ten mediation analyses were run to examine the mediating role of HAB in the relationships between big five personality traits and both proactive and reactive RA. For the second conceptual model, ten moderation analyses tested whether big five factors moderated the links between HAB and proactive and reactive RA. Prior research (Card & Little, 2006) suggests that reactive and proactive functions of aggressive behavior (relational and overt form) may be difficult to distinguish empirically when assessed with measures that do not separate the functions served by these two aggressive forms, as very high correlations are typically observed between these aggression functions. Because of this, in the analyses predicting reactive aggression, proactive aggression was included as a covariate and vice versa. In addition, moderation and mediation analyses were conducted for residualized reactive and proactive RA scores (e.g., Raine et al., 2006) controlling for residualized proactive and reactive RA respectively. Specifically, reactive RA was regressed on proactive scores and Pearson standardized residuals were saved to index purely proactive RA, while the standardized residuals of proactive RA on reactive RA were saved to index purely reactive RA. Residualized measures of proactive and reactive overt aggression were also created. Given the high correlations between the subtypes of relational and overt aggression (Marsee & Frick, 2007; Marsee et al., 2011), residualized reactive and proactive overt aggression scores were controlled in mediation and moderation analyses. All analyses were performed using IBM Statistics SPSS 21.

5.5. Personality The five broad traits of personality were assessed with the Short Form of the Big Five Questionnaire for Children (BFQ-C SF; Markos & Kokkinos, 2017), a 30-item self-report questionnaire for children and adolescents. Each of the factors (i.e., Energy/Extraversion, Agreeableness, Conscientiousness, Emotional Instability/Neuroticism, and Intellect/Openness) was measured with 6 items. Responses are given on a 5-point scale (1 = Almost never to 5 = Almost always). The CFA confirmed the correlated five-factor structure for the 30-item version, SB-χ2 = 671.459, df = 381, CFI = 0.91, TLI = 0.90, SRMR = 0.062, RMSEA = 0.048 (0.042–0.054). Alpha reliability coefficients were: 0.75 for E, 0.76 for A, 0.80 for C, 0.74 for EI/N, and 0.75 for O.

8. Preliminary analyses

5.6. Student characteristics

Means, standard deviations and correlations for the study's variables are reported in Table 1. Preliminary analyses regarding the between gender differences revealed that male participants reported higher proactive t(345) = 3.67, p < 0.001, Cohen's d = 0.30 and reactive RA t(345) = 2.72, p < 0.001, Cohen's d = 0.13 as well as proactive t (345) = 3.67, p < 0.001, Cohen's d = 0.47, and reactive overt aggression t(345) = 3.67, p < 0.001, Cohen's d = 0.22 than female participants. Further, female students reported higher levels of C t(344)

Participants also reported their gender, age, and grade level. 6. Procedure The study was conducted upon permission by the Institute of Education Policy, a scientific and consulting body under the authority 106

Journal of Applied Developmental Psychology 52 (2017) 101–113

C.M. Kokkinos et al.

Table 1 Descriptive statistics, Cronbach alphas (diagonal) and correlations between relational aggression, hostile attribution bias and personality.

1. RRA 2. PRA 3. RPA 4. PPA 5. HAB 6. E 7. A 8. C 9. N 10. O

M

SD

Range

1

2

3

4

5

6

7

8

9

10

0.47 0.32 0.61 0.31 4.52 3.76 3.81 3.57 2.68 3.49

0.49 0.48 0.55 0.51 2.21 0.69 0.68 0.82 0.75 0.77

2.60 2.50 2.80 2.90 10 3.50 3.33 3.67 4.00 3.50

(0.86) 0.89⁎⁎ 0.76⁎⁎ 0.85⁎⁎ 0.28⁎⁎ − 0.26⁎⁎ − 0.45⁎⁎ − 0.32⁎⁎ 0.32⁎⁎ − 0.23⁎⁎

(0.89) 0.72⁎⁎ 0.89⁎⁎ 0.24⁎⁎ −0.26⁎⁎ −0.45⁎⁎ −0.36⁎⁎ 0.29⁎⁎ −0.24⁎⁎

(0.87) 0.78⁎⁎ 0.27⁎⁎ − 0.18⁎⁎ − 0.42⁎⁎ − 0.33⁎⁎ 0.38⁎⁎ − 0.23⁎⁎

(0.91) 0.23⁎⁎ − 0.26⁎⁎ − 0.45⁎⁎ − 0.37⁎⁎ 0.33⁎⁎ − 0.27⁎⁎

(0.69) − 0.12⁎ − 0.13⁎ − 0.12⁎ 0.25⁎⁎ − 0.13⁎

(0.75) 0.47⁎⁎ 0.26⁎⁎ 0.01 0.41⁎⁎

(0.76) 0.67⁎⁎ − 0.28⁎⁎ 0.50⁎⁎

(0.80) − 0.24⁎⁎ 0.62⁎⁎

(0.74) −14⁎

(0.75)

Note 1. RRA = Reactive Relational Aggression, PRA = Proactive Relational Aggression, RPA = Reactive Physical Aggression, PPA = Proactive Physical Aggression, HAB = Hostile Attribution Bias, E = Extraversion, A = Agreeableness, C = Conscientiousness, N = Neuroticism, O = Openness to Experience. Note 2. M, SDs are presented for reactive and proactive aggression raw scores. Correlations with the other variables are presented for standardized residual aggression scores. ⁎ p < 0.05. ⁎⁎ p < 0.01.

bootstrap samples, and effect sizes was assessed using kappa squared (k2; Preacher & Kelley, 2011). The mediation analysis demonstrated that there was significant indirect effects of N[b = 0.041, SE = 0.021, 95% CI(0.019, 0.069)] with k2 = 0.042, E[b = −0.032, SE = 0.018, 95% CI(− 0.062, −0.010)] with k2 = 0.024, C[b = − 0.023, SE = 0.016, 95% CI (−0.053, − 0.011)] with k2 = 0.025, Α[b = − 0.031, SE = 0.019, 95% CI (− 0.034, − 0.010)] with k2 = 0.034, and O[b = − 0.026, SE = 0.018, 95% CI (−0.051, − 0.009)] with k2 = 0.029 on reactive RA through HAB. In all cases, results indicated that the direct effects of personality traits on reactive RA remained significant. Specifically, results showed that HAB mediated the links between high N, low E, C, A, and O on the one side and RA on the other. All the mediation effects of big five factors on reactive RA through HAB are summarized in Table 2. The analysis also demonstrated that there was a significant direct effect of N[b = 0.21, SE = 0.12, 95% CI(0.15, 0.32)] and HAB [b = 0.42, SE = 0.12, 95% CI (0.08, 0.14)] on proactive RA. However, there was no significant direct effect of N on HAB. According to Hayes (2009) it is possible to interpret a significant indirect effect even when the independent variable to mediator link is non-significant, as error variance or suppression effects can degrade the test of the effect of the independent variable (N) on the mediator (HAB), masking a real association. A small indirect effect of N on RA through HAB k2 = 0.09, 95% CI (0.04, 0.12) was observed. Therefore, these results suggest that the positive association between N and proactive RA is partially explained by HAB. A significant direct effect of E[b = −0.24, SE = 0.10, 95% CI (− 0.33, − 0.14)] and ΗΑΒ[b = 0.12, SE = 0.08, 95% CI(0.05, 0.11)] on proactive RA was also found. There was no significant direct effect of E on HAB. The indirect effect of E on proactive RA through HAB was small k2 = 0.06, 95% CI (− 0.14, − 0.13), supporting that HAB partially mediated the abovementioned link. Furthermore, C[b = −0.25, SE = 0.09, 95% CI (−0.36, − 0.23)] and ΗΑΒ[b = 0.13, SE = 0.11, 95% CI (0.08, 0.15)] significantly directly predicted proactive RA, while C was not found to predict HAB. The indirect effect of C on proactive RA through HAB was small k2 = 0.12, 95% CI (−0.12, − 0.009), supporting that HAB partially mediated this link. A significant direct effect of A[b = −0.35, SE = 0.09, 95% CI (−0.44, − 0.18)] and ΗΑΒ[b = 0.09, SE = 0.07, 95% CI (0.06, 0.14)] on proactive RA was found. There was no significant direct effect of A on HAB. The indirect effect of Α on proactive RA through HAB was small k2 = 0.08, 95% CI (− 0.10, − 0.008), supporting that HAB partially mediated the link between A and proactive RA. Finally, O[b = −0.23, SE = 0.11, 95% CI (−0.34, − 0.11)] and ΗΑΒ [b = 0.14, SE = 0.12, 95% CI (0.09, 0.21)] significantly directly predicted proactive RA, while O was not found to predict HAB. The indirect effect of O on proactive RA through HAB was small k2 = 0.12, 95% CI (−0.15,

Table 2 Mediator effects of big five factors on reactive relational aggression through HAB.

Neuroticism Extraversion Openness Agreeableness Conscientiousness

Big five factors as predictors of HAB

HAB as predictor of RRA

Big five factors as predictors of RRA

0.16⁎⁎ −0.17⁎⁎ −0.18⁎⁎ −0.17⁎⁎ −0.16⁎⁎

0.03⁎ − 0.02⁎ − 0.02⁎ − 0.02⁎ − 0.03⁎

0.08⁎ −0.09⁎⁎ −0.09⁎⁎ −0.08⁎⁎ −0.07⁎

Note 1. HAB = Hostile Attribution Bias, RRA = Reactive Relational Aggression. ⁎ p < 0.05. ⁎⁎ p < 0.01.

= − 4.27, p < 0.001,Cohen's d = 0.46, A t(344) = − 5.38, p < 0.001, Cohen's d = 0.58, and O t(344) = − 3.04, p < 0.01, Cohen's d = 0.33, compared to male ones. However, no other between gender differences were found. To examine the within gender differences, one-tailed paired samples t-tests were used. t-tests revealed that female students use more reactive (M = 0.41, SD = 0.42) compared to proactive RA (M = 0.24, SD = 0.38), t(193) = −10.22, p < 0.001, Cohen's d = 0.37. Male students were also found to use more reactive (M = 0.55, SD = 0.56) than proactive RA (M = 0.43, SD = 0.56), t(154) = − 6.72, p < 0.001, Cohen's d = 0.71, as well as more reactive overt (M = 0.74, SD = 0.62) than reactive RA (M = 0.55, SD = 0.74), t(153) = − 0.5, 94, Cohen's d = 0.65. However, no other significant within gender differences were found. Pearson correlations indicated that proactive and reactive RA were significantly and positively associated. As expected, greater proactive and reactive RA scores were related to higher HAB. Extraversion, A, C, and O were negatively associated to both proactive and reactive RA, whereas N positively. In addition, HAB was negatively correlated with Ε, A, C, and O while positively with N. 9. Model testing 9.1. Mediation The mediation model (Fig. 1) was tested using model 4 from the PROCESS macro for SPSS (Hayes, 2013). With this method, significance is tested by calculating bias-corrected 95% bootstrap confidence intervals (CI) for the direct and indirect effects, while significance is inferred if the resultant interval does not contain zero (Hayes, 2013). In the present study, the direct effects of personality traits on RA were estimated, as well as the indirect effect of these traits on RA through the mediator (i.e., HAB). Ninety-five percent CI were based on 5000 107

Journal of Applied Developmental Psychology 52 (2017) 101–113

C.M. Kokkinos et al.

were examined by rearranging the order of the variables. However, none of the indirect effects in these models were statistically significant. Specifically, the indirect effects were not significant when: a) HAB served as the predictor variable, personality traits were the mediating variables and RA was the dependent variable, and b) personality traits served as the predictor variables, HAB was the moderator and RA was the dependent variable (output available from the authors upon request).

Table 3 Mediator effects of big five factors on proactive relational aggression through HAB.

Neuroticism Extraversion Openness Agreeableness Conscientiousness

Big five factors as predictors of HAB

HAB as predictor of PRA

Big five factors as predictors of PRA

0.16⁎⁎ −0.17⁎⁎ −0.17⁎⁎ −0.18⁎⁎ −0.16⁎⁎

0.03⁎ − 0.03⁎ − 0.03⁎ − 0.03⁎ − 0.03⁎

0.08 − 0.08 − 0.07 − 0.09 − 0.07

10. Discussion

Note 1. HAB = Hostile Attribution Bias, PRA = Proactive Relational Aggression. ⁎ p < 0.05. ⁎⁎ p < 0.01.

The current study addresses a gap in the literature by examining the underlying mechanisms implicated in the association between HAB, personality and the distinct functions of RA (i.e., proactive and reactive). As expected, the results indicated that both reactive and proactive RA were positively correlated with HAB for relational provocations (Bailey & Ostrov, 2008; Murray-Close et al., 2010). Relationally aggressive students, irrespective of their goals (i.e., instrumental or retaliatory), seem to be more interested in their interpersonal relationships and demonstrate higher HAB in response to perceived relational provocations (Mathieson et al., 2011). Regarding the relationship between big five factors and RA, results suggested that A, C, E, and O were negatively correlated with both proactive and reactive RA, whereas N positively. It seems that students with low A, C, and O, as well as high N are more likely to engage in RA (Gleason et al., 2004; Tackett et al., 2013; Tackett et al., 2014). The observed correlation between E and RA was inconsistent with the initial hypothesis, nevertheless, it agrees with prior research supporting that adolescents characterized by positive emotions (i.e., facet of E; Tackett et al., 2014), sociability and extraversion are less likely to manifest RA (Kokkinos et al., 2016). These findings also support that the same traits, which predispose students to RA, may be associated in the same way with both its proactive and reactive functions (Bettencourt et al., 2006; Miller et al., 2012). In accordance with prior research, proactive and reactive RA were highly correlated (Crapanzano et al., 2010) suggesting that many aggressive individuals may simultaneously engage in both aggressive functions (Crapanzano et al., 2010; Ramirez, 2009). This high correlation between the two functions of aggression has driven some researchers to dispute whether these dimensions refer to distinct behaviors and have different sources (Marsee et al., 2014). Results also indicated that A, C, E and O were negatively correlated with HAB, while N positively. In line with previous studies, it is suggested that antagonistic individuals tend to attribute hostile intents to others' actions (Koolen et al., 2012; Miller et al., 2008), which could be due to their past negative experiences (e.g., trauma, victimization; Miller et al., 2012). Additionally, students scoring low in C, indicative of low self-regulation (Koolen et al., 2012) which is linked to higher HAB (Nelson & Perry, 2015), may be more susceptible to hostile interpretations. On the contrary, those high in N, who tend to perceive threats (Caspi et al., 2005) and interpret facts as frustrating (Hessels et al., 2014), were more likely to demonstrate HAB (Simoncic, Kuhlman, Vargas, Houchins, & Lopez-Duran, 2014). Although none of the prior studies has examined the relationship between HAB and E or O, the current findings indicate that high O students, who are more likely to process abstract and perceptual information in a versatile and efficient way (DeYoung et al., 2010), may demonstrate lower HAB. In terms of E, a negative association with HAB was found, in line with evidence supporting that these attributions are more common among withdrawn children (characterized by lower E; Krueger & Tackett, 2006; Salmivalli, Ojanen, Haanpää, & Peets, 2005). Results also provided evidence supporting that the links between high N, low A, C, E, O and reactive RA were mediated by HAB. Antagonistic individuals may be more likely to make hostile attributions (Miller et al., 2008), which in turn are linked to aggressive responses (Barlett & Anderson, 2012; Jones et al., 2011; Miller et al.,

− 0.009), providing support that HAB partially mediated the association between O and proactive RA. The mediation effects of big five factors on proactive RA through HAB are summarized in Table 3.

9.2. Moderation Regarding the second model (Fig. 2) moderation analyses were conducted (model 1 of the PROCESS macro) to explore whether personality traits moderated the association between HAB and proactive and reactive RA. Results showed that only C from the big five factors moderated the positive link between HAB and reactive RA. To interpret the significant interaction effects of C, Fig. 3 shows the line graph that was plotted based on the values of the 10th, 25th, 50th, 75th and 90th percentile of the moderator. Simple slopes analysis demonstrated that at one standard deviation (SD) below the M of C, increased reactive RA was significantly related to HAB[b = − 0.12, SE = 0.08 95% CI (0.08, 0.19), t = 5.83, p < 001].In contrast, at one SD above the M of C, the positive effects of HAB on reactive RA were smaller and insignificant. Other personality traits were not found to moderate either this link or the link between HAB and proactive RA. To further establish the validity of the models, competing models

Fig. 3. Hostile Attribution Bias as a function of Conscientiousness. The interaction predicts reactive relational aggression. The slope indicates that as Hostile Attribution Bias increases, reactive relational aggression also increases. However, the slope is increasingly reduced as the level of Conscientiousness increases. Note. RRA = Reactive Relational Aggression.

108

Journal of Applied Developmental Psychology 52 (2017) 101–113

C.M. Kokkinos et al.

for relational versus physical aggression. In the same vein, male students were found more likely to engage in more reactive aggression (both relational and overt) compared to proactive as they seem to be more impulsive in their social relations and prefer direct retaliation (Knyazev, 2004). However, there were no indications of gender differences in HAB for relational provocations possibly reflecting that interpersonal relationships become important for male students, since the manipulation of relationships could threaten their group inclusion or self-esteem (Yeung & Leadbeater, 2007). With respect to big five factors, results revealed gender differences in C, A and O indicating that female participants have higher levels of these traits (Klimstra et al., 2009; Kokkinos, Antoniadou, Dalara, Koufogazou, & Papatziki, 2013). Klimstra et al. (2009) noted that girls are likely to mature earlier during adolescence, which may be linked to higher levels of some big five factors related to social maturity, peer prestige, leadership, and selfesteem. From an evolutionary perspective, the higher levels of positive traits such as C among female students could also be attributed to the fact that smooth interpersonal relationships are stressed by being female (Klimstra et al., 2009). In terms of O, it has been suggested that girls may precede in intellectual and cognitive functioning during early adolescence (De Bolle et al., 2015). However, De Bolle et al. (2015) showed that adolescent girls score higher only in aesthetics and feelings which are two of the facets of O. Overall, the present study stresses the role of personality traits and social-cognitive factors in the emergence of relationally aggressive behavior. However, it goes beyond examining the main effects of personal and cognitive mechanisms on RA by investigating more complex interactive relationships among these factors. Based on the GAM (Anderson & Bushman, 2002), it was suggested that personality traits (i.e., FFM traits) could be associated to RA through cognitive processes (i.e., HAB; Barlett & Anderson, 2012). Additionally, this study integrates two theoretical models (Integrative Cognitive Model; Wilkowski & Robinson, 2010; relational vulnerability model; Crick et al., 2004) proposing that some personality traits, reflecting self-regulatory/effortful control skills (i.e., A and C) and high emotional distress (i.e., N), could affect the relationship between HAB and RA. On the one hand, the findings indicated that personality traits could contribute to RA through HAB. Specifically, the results revealed that some personality traits (low A, C, E, O and high N) may predispose to HAB which in turn could be associated with RA, thus providing support for the GAM. On the other hand, according to the present findings, C may moderate the association between HAB and reactive RA. It seems therefore that not only may personality traits precede hostile attributions which are likely to be associated with RA (GAM) but also some traits (e.g., low C) may enhance the impact of HAB on reactive RA. Hence, low C (indicative of lower effortful control) adolescents, may be more vulnerable to reactive RA in case of making hostile interpretations. Adversely, those high in C (indicative of higher effortful control) are more likely to be protected against reactive RA when they perceive hostile provocations. The finding of the interactive effects of person factors (i.e., personality) and individual's internal state (i.e., cognitive factors) in the development of reactive RA seems not to be currently considered by the GAM (Bushman, 2016). However, these relationships are in line with the ICM referring that individuals characterized by higher effortful control (e.g., high C) may be less likely to react aggressively when they make hostile interpretations (Wilkowski & Robinson, 2010). There are several theoretical and practical implications derived from the results of the current study. The findings support the proposed models where a cluster of individual factors (i.e., personality traits, social cognition) place adolescents at risk for RA. In other words, it is those adolescents who are characterized by low A, C, E, O, high N, and demonstrate HAB in response to relational provocation who are likely to be relationally aggressive. Prevention programs targeting RA could directly target these risk factors. In this line, both parents and adults

2012) and specifically to reactive RA (Murray-Close et al., 2010). Specifically, low C students characterized by lower self-control (found to be linked to higher HAB; Koolen et al., 2012; Nelson & Perry, 2015) are likely to engage through HAB in reactive RA. Furthermore, those low in E, who appear to be more withdrawn (Krueger & Tackett, 2006), may demonstrate higher HAB (Salmivalli et al., 2005) and thus react via RA. High O students who are more likely to process information flexibly and efficiently (DeYoung et al., 2010), may avoid making hostile interpretations and thus demonstrate lower reactive RA. Regarding N, students scoring high in this trait may also make more hostile interpretations (Simoncic et al., 2014), which are linked to reactive RA (Bailey & Ostrov, 2008). According to Mathieson et al. (2011) when children show more hostile attributions and emotional distress in response to hypothetical relational provocations are more likely to act in a relationally aggressive manner. Additionally, results indicated that HAB could partially mediate the association between big five factors and proactive RA, implying that students low in A, C, O, E and high in N are also likely to select proactive relationally aggressive strategies to achieve their instrumental goals (e.g., popularity, high social status) partly due to their tendency to make hostile interpretations of relational provocations. Regarding the second model, results showed that C moderated the relationship between HAB and reactive RA. It seems therefore that low C students, who possibly have lower self-control and selfregulatory skills (Duckworth, 2011; Jensen-Campbell, Knack, Waldrip, & Campbell, 2007), are more likely to get involved in reactive RA when they demonstrate HAB. These findings are in line with previous research supporting this relationship for reactive aggression (Koolen et al., 2012). On the contrary, high C students are more likely to resist to reactive RA when they demonstrate HAB, which is in agreement with the proposition that high self-control could protect against reactive aggression, even in case of making hostile interpretations (Koolen et al., 2012). As far as gender differences are concerned, male participants reported higher levels of both proactive and reactive RA which is consistent with prior evidence (Kokkinos et al., 2016; Vagos et al., 2014) and opposes the notion that RA is a female type of aggression (Voulgaridou & Kokkinos, 2015). The desire to dominate over peers with weaker social status, alternation of mood, jealousy, boredom and insecurity in a friendship relationship are some possible explanations for the obtained difference (Kokkinos & Voulgaridou, 2017; Pronk & Zimmer-Gembeck, 2010). According to Murray-Close et al. (2010), the ability of male students to use relationally aggressive strategies could be attributed to the increased levels of interaction between boys and girls during adolescence. Additionally, since the motivation to engage in RA is about domination, popularity and the need of belongingness, male students are more likely to try to exclude from the group those who exhibit less ‘masculine’ behaviors or those not participating in several sports. According to the gender-linked model of aggression such preferences of males for relationally aggressive behaviors may be related to cultural differences. For instance, when youth from collectivistic societies are aggressive, they are likely to adopt relationally aggressive rather than physically aggressive behavioral strategies (Murray-Close et al., 2016). Further, female students were found to show more reactive compared to proactive RA, providing evidence for prior studies (Marsee & Frick, 2007; Vitaro et al., 2002). However, female participants paradoxically were not found to score higher on the RA functions compared to overt aggression contrary to recent research evidence showing that across different samples (i.e., either community or detained), female adolescents are more likely to report more relational than physical aggression (Marsee et al., 2014). This could be attributed to collectivistic cultures that may discourage the use of various types of aggression because they may interfere with a harmonious group functioning (Murray-Close et al., 2016). These findings point to distinct cultural norms for aggression and underscore the importance of considering cultural differences in within-group preferences 109

Journal of Applied Developmental Psychology 52 (2017) 101–113

C.M. Kokkinos et al.

causal model or determine a timeline for which factors developed first. Maxwell, Cole, and Mitchell (2011) claimed that the results from cross-sectional data can be biased as the temporal sequence of the independent, mediator, and dependent variables cannot be verified. The onset of the mediating variable must precede the onset of the independent to be considered a true mediator (Collins, Graham, & Flaherty, 1998). However, although it has been argued that cross-sectional data are appropriate for mediation research (Hayes, 2013, p. 89), the findings should be cautiously interpreted. The cross-sectional design, as well as the study's correlational nature may also have confounded the direction of the effect. Research on instrumental HAB and physical aggression suggests that HAB precede aggressive responses, both immediately and over time (Mathieson et al., 2011). In terms of RA, similar effects may emerge although to date the empirical evidence is scarce. Prior research which explored the direction of the effects in the association between HAB and RA revealed that adolescents hostile attributions for relational provocations predict their engagement in RA (Hoglund & Leadbeater, 2007; Yeung & Leadbeater, 2007). Regarding the association between personality traits and RA, there are strong theoretical reasons to support that a host of individual factors (e.g., personality) predict adolescents' engagement in this aggressive behavior (Kuppens, Grietens, Onghena, Michiels, & Subramanian, 2008). Nevertheless, longitudinal designs are necessary to address the direction of these associations as well as to illuminate whether this study's proposed models predict increases in RA over time. Another limitation of the study is that all the measures were based on self-report, providing a potential source of measurement error due to rater bias, as well as shared method variance using a single method at a single time (cross-sectional). Although this approach has been used in past research considering that adolescents themselves seem to have more direct knowledge of their own behavior and personality characteristics (Lansford et al., 2012), future research could benefit from combining self-report measures with reports from other informants, such as peers, parents, and teachers. Further, while the self-report Peer Conflict Scale was used in this study to assess both overt and relational forms of aggression, some of the items in the overt aggression scales use the word “hurt” to describe the aggressive act (e.g., “If others make me mad, I hurt them”), which could be interpreted as either an overt or relationally aggressive response. While these items have been used in previous studies showing good psychometric properties (Gázquez, Pérez-Fuentes, Molero, Martín, & Sánchez-Marchán, 2016; Kokkinos & Voulgaridou, 2017; Loflin & Barry, 2016; Marsee et al., 2011, 2014), minor revisions regarding the wording of some items may be warranted in future research with the PCS to create items that unambiguously assess overt and relational aggression. Lastly, as the sample included mainly adolescents, the proposed pathways remain to be replicated in other age groups. The fact that the present sample was convenient may limit the ability to generalize results from this study to a wider population. Hence, future studies are necessary to replicate these findings in more targeted samples. In all, the goal of this study was to investigate how hostile attribution bias for relational provocations is related to the two distinct functions of RA, given the mixed findings from prior research. The study supported the hypothesis that it is the combination of HAB and personality traits that put adolescents at risk for engaging in RA. However, future research is needed to replicate these models using longitudinal study designs, as well as multiple measures of the constructs. Thus, this study has laid the groundwork for future ones to explore these associations across development.

working with children (e.g., teachers, counselors) should be well-informed about RA to reflect on or even modify their normative beliefs (i.e., that this aggressive form is not harmful; Werner, Senich, & Przepyszny, 2006), and recognize the negative consequences associated with such behaviors (e.g., serious adjustment difficulties; Fraser et al., 2005; Leff, Waasdorp, & Crick, 2010). Proactive steps to decrease RA and simultaneously increase the use of prosocial strategies as well as empathy, problem solving and social-cognitive skills should be taken in both school and family settings (e.g., Cappella & Weinstein, 2006; Fraser et al., 2005).With respect to the educational settings, both teachers and school psychologists should be aware of the individual factors (e.g., social cognition) implicated in this aggressive form in order to take measures to prevent RA through social-cognitive reframing interventions (Kokkinos et al., 2016; Leff et al., 2010). For example, interventions could focus on reducing HAB in response to ambiguous relational situations by encouraging adolescents to cautiously process not only verbal but also nonverbal information and implicate strategies, including slowdown of reactions as well as teaching of effective problem solving, conflict resolution and information processing skills (Fraser et al., 2005; Yeung & Leadbeater, 2007). Considering that positive personality traits are likely to be associated with lower RA, educators could also use teaching strategies which promote behaviors consistent with high C (e.g., divide schoolwork into sections), A (e.g., encourage cooperation), and O (e.g., cultivate imagination, Komarraju, Karau, Schmeck, & Avdic, 2011). These interventions could be considered irrespective of gender, as relationally aggressive strategies could be equally employed by both male and female students (Leff et al., 2010). Additionally, personality traits should be taken into consideration when designing and implementing intervention programs. The empirical evidence for the role of personality characteristics (low E, C, A, O, high N) in the association between HAB and RA provides support for the existing interventions for relational aggressors that promote social skills and prosocial behavior. With regards to the protective role of C in the association between HAB and RA, interventions that enhance C in adolescents are effective, including certain teaching strategies, such as fragmentation of the responsibilities of students or cooperative learning (Komarraju et al., 2011). The current findings are also in line with research evidence indicating that personality traits could enhance the effectiveness of social-cognitive interventions targeting RA, suggesting that these interventions should be tailored to participants' specific needs. For instance, a more intensive intervention may be necessary for low C students (Stoltz et al., 2013). On the other side, the positive behavioral outcomes of interventions targeting aggression may be mediated by an individual's SIP skills (Fraser et al., 2005), thus providing to those skills a central role in interventions for reducing RA. Although personality traits have been conceptualized as rather immutable, interventions should not aspire to change students' stable personality characteristics; instead to equip them with the behavioral tools to cope more effectively with externalizing problems (McNiel, Lowman, & Fleeson, 2010). Understanding the personality profile of students involved in RA could also provide a basis for designing interventions including anger management skills (e.g., relaxation techniques, role playing), detecting biases toward provocations of anger, and thus regulating the conditions related to aggression (Lochman & Wells, 2004). The findings from this study should be considered in light of various strengths and limitations. Strengths include focusing on adolescence, a life period characterized by important changes in RA, and accounting for both personality (big five factors) and social cognitive factors (i.e., HAB) in understanding RA. Additionally, the present research examined differential relations between these factors and proactive and reactive RA. However, although this study has provided insights regarding the complex nature of social-information-processing, personal characteristics and RA, it is not without limitations. Firstly, given the cross-sectional research design we cannot discern a

References Ahadi, S. A., & Rothbart, M. K. (1994). Temperament, development and the big five. In C. F. Halverson, G. A. Kohnstamm, & R. P. Martin (Eds.), The developing structure of temperament and personality from infancy to adulthood (pp. 186–207). Hillsdale, NJ:

110

Journal of Applied Developmental Psychology 52 (2017) 101–113

C.M. Kokkinos et al.

Crick, N. R., Grotpeter, J. K., & Bigbee, M. A. (2002). Relationally and physically aggressive children's intent attributions and feelings of distress for relational and instrumental peer provocations. Child Development, 73, 1134–1142. http://dx.doi.org/ 10.1111/1467-8624.00462. Crick, N. R., Murray-Close, D., & Woods, K. (2006). Borderline personality features in childhood: A short-term longitudinal study. Development and Psychopathology, 17, 1051–1070. http://dx.doi.org/10.1017/S0954579405050492. Crick, N. R., Murray-Close, D., & Woods, K. E. (2004, March). The role of hostile intent attributions, peer victimization, and gender in predicting physical and relational aggression. Paper presented at the biennial meeting of the Society for Research on adolescence. MD: Baltimore. Dane, A. V., & Marini, Z. A. (2014). Overt and relational forms of reactive aggression in adolescents: Relations with temperamental reactivity and self-regulation. Personality and Individual Differences, 60, 60–66. http://dx.doi.org/10.1016/j.paid.2013.12.021. De Bolle, M., De Fruyt, F., McCrae, R. R., Löckenhoff, C. E., Costa, P. T., Jr, Aguilar-Vafaie, M. E., ... Avdeyeva, T. V. (2015). The emergence of sex differences in personality traits in early adolescence: A cross-sectional, cross-cultural study. Journal of Personality and Social Psychology, 108, 171–185. DeYoung, C. G., Hirsh, J. B., Shane, M. S., Papademetris, X., Rajeevan, N., & Gray, J. R. (2010). Testing predictions from personality neuroscience brain structure and the big five. Psychological Science, 21, 820–828. http://dx.doi.org/10.1177/ 0956797610370159. DeYoung, C. G., Quilty, L. C., & Peterson, J. B. (2007). Between facets and domains: 10 aspects of the big five. Journal of Personality and Social Psychology, 93, 880–896. http://dx.doi.org/10.1037/0022-3514.93.5.880. Dodge, K. A., Malone, P. S., Lansford, J. E., Sorbring, E., Skinner, A. T., Tapanya, S., ... Pastorelli, C. (2015). Hostile attributional bias and aggressive behavior in global context. Proceedings of the National Academy of Sciences, 112, 9310–9315. http://dx. doi.org/10.1073/pnas.1418572112. Duckworth, A. L. (2011). The significance of self-control. Proceedings of the National Academy of Sciences, 108, 2639–2640. http://dx.doi.org/10.1073/pnas.1019725108. Fite, J. E., Goodnight, J. A., Bates, J. E., Dodge, K. A., & Pettit, G. S. (2008). Adolescent aggression and social cognition in the context of personality: Impulsivity as a moderator of predictions from social information processing. Aggressive Behavior, 34, 511–520. http://dx.doi.org/10.1002/ab.20263. Fraser, M. W., Galinsky, M. J., Smokowski, P. R., Day, S. H., Terzian, M. A., Rose, R. A., & Guo, S. (2005). Social information-processing skills training to promote social competence and prevent aggressive behavior in the third grades. Journal of Consulting and Clinical Psychology, 73, 1045–1055. http://dx.doi.org/10.1037/0022-006X.73.6. 1045. Fung, A. L., Raine, A., & Gao, Y. (2009). Cross-cultural generalizability of the reactiveproactive aggression questionnaire (RPQ). Journal of Personality Assessment, 91, 473–479. http://dx.doi.org/10.1080/00223890903088420. Gázquez, J. J., Pérez-Fuentes, M. D. C., Molero, M. D. M., Martín, A. B. B., & SánchezMarchán, C. (2016). Drug use in adolescents in relation to social support and reactive and proactive aggressive behavior. Psicothema, 28, 318–322. http://dx.doi.org/10. 7334/psicothema2015.327. Gleason, K. A., Jensen-Campbell, L. A., & South Richardson, D. (2004). Agreeableness as a predictor of aggression in adolescence. Aggressive Behavior, 30, 43–61. http://dx.doi. org/10.1002/ab.20002. Godleski, S. A., & Ostrov, J. M. (2010). Relational aggression and hostile attribution biases: Testing multiple statistical methods and models. Journal of Abnormal Child Psychology, 38, 447–458. http://dx.doi.org/10.1007/s10802-010-9391-4. Guarino, L., Roger, D., & Olason, D. T. (2007). Reconstructing N: A new approach to measuring emotional sensitivity. Current Psychology, 26, 37–45. http://dx.doi.org/10. 1007/s12144-007-9004-8. Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical mediation analysis in the new millennium. Communication Monographs, 76, 408–420. http://dx.doi.org/10.1080/ 03637750903310360. Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. Guilford Press. Helfritz-Sinville, L. E., & Stanford, M. S. (2014). Hostile attribution bias in impulsive and premeditated aggression. Personality and Individual Differences, 56, 45–50. http://dx. doi.org/10.1016/j.paid.2013.08.017. Hessels, C., van den Hanenberg, D., de Castro, B. O., & van Aken, M. A. (2014). Relationships: Empirical contribution. Understanding personality pathology in adolescents: The five factor model of personality and social information processing. Journal of Personality Disorders, 28, 121–142. http://dx.doi.org/10.1521/pedi.2014. 28.1.121. Hoglund, W. L., & Leadbeater, B. J. (2007). Managing threat: Do social-cognitive processes mediate the link between peer victimization and adjustment problems in early adolescence? Journal of Research on Adolescence, 17, 525–540. http://dx.doi.org/10. 1111/j.1532-7795.2007.00533.x. Hu, 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, 1–55. http://dx.doi.org/10.1080/ 10705519909540118. Jensen-Campbell, L. A., Knack, J. M., Waldrip, A. M., & Campbell, S. D. (2007). Do big five personality traits associated with self-control influence the regulation of anger and aggression? Journal of Research in Personality, 41, 403–424. http://dx.doi.org/10. 1016/j.jrp.2006.05.001. Jones, S. E., Miller, J. D., & Lynam, D. R. (2011). Personality, antisocial behavior, and aggression: A meta-analytic review. Journal of Criminal Justice, 39, 329–337. http:// dx.doi.org/10.1016/j.jcrimjus.2011.03.004. Klimstra, T. A., Hale, W. W., Raaijmakers, Q. A., Branje, S. J., & Meeus, W. H. (2009). Maturation of personality in adolescence. Journal of Personality and Social Psychology,

Erlbaum. Anderson, C. A., & Bushman, B. J. (2002). Human aggression. Psychology, 53, 27–51. http://dx.doi.org/10.1146/annurev.psych.53.100901.135231. Archer, J. (2004). Sex differences in aggression in real-world settings: A meta-analytic review. Review of General Psychology, 8, 291–322. http://dx.doi.org/10.1037/10892680.8.4.291. Archer, J., & Coyne, S. M. (2005). An integrated review of indirect, relational, and social aggression. Personality and Social Psychology Review, 9, 212–230. http://dx.doi.org/ 10.1207/s15327957pspr0903_2. Bailey, C. A., & Ostrov, J. M. (2008). Differentiating forms and functions of aggression in emerging adults: Associations with hostile attribution biases and normative beliefs. Journal of Youth and Adolescence, 37, 713–722. http://dx.doi.org/10.1007/s10964007-9211-5. Bandura, A. (1977). Social learning theory. New York: Prentice Hall. Barlett, C. P., & Anderson, C. A. (2012). Direct and indirect relations between the big 5 personality traits and aggressive and violent behavior. Personality and Individual Differences, 52, 870–875. http://dx.doi.org/10.1016/j.paid.2012.01.029. Baş, A. U., & Yurdabakan, İ. (2012). Factor structure of the Reactive-Proactive Aggression Questionnaire in Turkish children and gender, grade-level, and socioeconomic status differences in reactive and proactive aggression. Journal of Psychoeducational Assessment, 30, 284–297. Berkowitz, L. (1993). Pain and aggression: Some findings and implications. Motivation and Emotion, 17, 277–293. http://dx.doi.org/10.1007/BF00992223. Bettencourt, B., Talley, A., Benjamin, A. J., & Valentine, J. (2006). Personality and aggressive behavior under provoking and neutral conditions: A meta-analytic review. Psychological Bulletin, 132, 751–777. http://dx.doi.org/10.1037/0033-2909.132.5. 751. Branje, S. J., Van Lieshout, C. F., & Gerris, J. R. (2007). Big five personality development in adolescence and adulthood. European Journal of Personality, 21, 45–62. http://dx. doi.org/10.1002/per.596. Burt, S. A., Donnellan, M. B., & Tackett, J. L. (2012). Should social aggression be considered “antisocial”? Journal of Psychopathology and Behavioral Assessment, 34, 153–163. http://dx.doi.org/10.1007/s10862-011-9267-0. Burton, L. A., Hafetz, J., & Henninger, D. (2007). Gender differences in relational and physical aggression. Social Behavior and Personality: An International Journal, 35, 41–50. http://dx.doi.org/10.2224/sbp.2007.35.1.41. Bushman, B. J. (2016). Violent media and hostile appraisals: A meta-analytic review. Aggressive Behavior, 42, 605–613. http://dx.doi.org/10.1002/ab.21655. Cappella, E., & Weinstein, R. (2006). The prevention of social aggression among girls. Social Development, 15, 434–462. http://dx.doi.org/10.1111/j.1467-9507.2006. 00350.x. Card, N. A., & Little, T. D. (2006). Proactive and reactive aggression in childhood and adolescence: A meta-analysis of differential relations with psychosocial adjustment. International Journal of Behavioral Development, 30, 466–480. http://dx.doi.org/10. 1177/0165025406071904. Card, N. A., Stucky, B. D., Sawalani, G. M., & Little, T. D. (2008). Direct and indirect aggression during childhood and adolescence: A meta-analytic review of gender differences, intercorrelations, and relations to maladjustment. Child Development, 79, 1185–1229. http://dx.doi.org/10.1111/j.1467-8624.2008.01184.x. Caspi, A., & Roberts, B. W. (2001). Personality development across the life course: The argument for change and continuity. Psychological Inquiry, 12, 49–66. http://dx.doi. org/10.1207/S15327965PLI1202_01. Caspi, A., Roberts, B. W., & Shiner, R. L. (2005). Personality development: Stability and change. Annual Review of Psychology, 56, 453–484. http://dx.doi.org/10.1146/ annurev.psych.55.090902.141913. Cavalcanti, J. G., & Pimentel, C. E. (2016). Personality and aggression: A contribution of the general aggression model. Estudos de Psicologia (Campinas), 33, 443–451. http:// dx.doi.org/10.1590/1982-02752016000300008. Chen, P., Coccaro, E. F., & Jacobson, K. C. (2012). Hostile attributional bias, negative emotional responding, and aggression in adults: Moderating effects of gender and impulsivity. Aggressive Behavior, 38, 47–63. http://dx.doi.org/10.1002/ab.21407. Collins, L. M., Graham, J. J., & Flaherty, B. P. (1998). An alternative framework for defining mediation. Multivariate Behavioral Research, 33, 295–312. http://dx.doi.org/ 10.1207/s15327906mbr3302_5. Connor, D. F., Steingard, R. J., Anderson, J. J., & Melloni, R. H. (2003). Gender differences in reactive and proactive aggression. Child Psychiatry and Human Development, 33, 279–294. http://dx.doi.org/10.1023/A:1023084112561. Crain, M. M., Finch, C. L., & Foster, S. L. (2005). The relevance of the social information processing model for understanding relational aggression in girls. Merrill-Palmer Quarterly, 51, 213–249. http://dx.doi.org/10.1353/mpq.2005.0010. Crapanzano, A. M., Frick, P. J., & Terranova, A. M. (2010). Patterns of physical and relational aggression in a school-based sample of boys and girls. Journal of Abnormal Child Psychology, 38, 433–445. http://dx.doi.org/10.1007/s10802-009-9376-3. Crick, N. R. (1995). Relational aggression: The role of intent attribution, feelings of distress, and provocation type. Development and Psychopathology, 7, 313–322. http://dx. doi.org/10.1017/S0954579400006520. Crick, N. R., & Dodge, K. A. (1994). A review and reformulation of social informationprocessing mechanisms in children's social adjustment. Psychological Bulletin, 115, 74–101. http://dx.doi.org/10.1037/0033-2909.115.1.74. Crick, N. R., Geiger, T. C., & Zimmer-Gembeck, M. J. (2003, March). Relational vulnerability: A model for understanding girls, aggression, and depressive symptoms. Paper presented at the biennial meeting of the Society for Research on child development. FL: Tampa. Crick, N. R., & Grotpeter, J. K. (1995). Relational aggression, gender, and social-psychological adjustment. Child Development, 66, 710–722. http://dx.doi.org/10.2307/ 1131945.

111

Journal of Applied Developmental Psychology 52 (2017) 101–113

C.M. Kokkinos et al.

1002/per.738. Meece, D. W., Mize, J., Bates, J. E., Dodge, K. A., & Pettit, G. S. (2007, March). Temperament moderates the association between young children's hostile attributions and aggression. Poster session presented at the Biennial Meeting of the Society of Research on Child Development, Boston MA. Meier, B. P., & Robinson, M. D. (2004). Does quick to blame mean quick to anger? The role of agreeableness in dissociating blame and anger. Personality and Social Psychology Bulletin, 30, 856–867. http://dx.doi.org/10.1177/0146167204264764. Meier, B. P., Wilkowski, B. M., & Robinson, M. D. (2008). Bringing out the agreeableness in everyone: Using a cognitive self-regulation model to reduce aggression. Journal of Experimental Social Psychology, 44, 1383–1387. http://dx.doi.org/10.1016/j.jesp. 2008.05.005. Mikami, A. Y., Lee, S. S., Hinshaw, S. P., & Mullin, B. C. (2008). Relationships between social information processing and aggression among adolescent girls with and without ADHD. Journal of Youth and Adolescence, 37, 761–771. http://dx.doi.org/10. 1007/s10964-007-9237-8. Miller, J. D., Dir, A., Gentile, B., Wilson, L., Pryor, L. R., & Campbell, W. K. (2010). Searching for a vulnerable dark triad: Comparing factor 2 psychopathy, vulnerable narcissism, and borderline personality disorder. Journal of Personality, 78, 1529–1564. http://dx.doi.org/10.1111/j.1467-6494.2010.00660.x. Miller, J. D., & Lynam, D. R. (2006). Reactive and proactive aggression: Similarities and differences. Personality and Individual Differences, 41, 1469–1480. http://dx.doi.org/ 10.1016/j.paid.2006.06.004. Miller, J. D., Lynam, D. R., & Jones, S. (2008). Externalizing behavior through the lens of the five-factor model: A focus on agreeableness and conscientiousness. Journal of Personality Assessment, 90, 158–164. http://dx.doi.org/10.1080/ 00223890701845245. Miller, J. D., Zeichner, A., & Wilson, L. F. (2012). Personality correlates of aggression: Evidence from measures of the five-factor model, UPPS model of impulsivity, and BIS/BAS. Journal of Interpersonal Violence, 27, 2903–2919. http://dx.doi.org/10. 1177/0886260512438279. Mullins-Sweat, N. S., & Widiger, A. T. (2006). The five factor model of personality disorder: A translation across science and practice. In R. F. Krueger, & J. L. Tackett (Eds.), Personality and psychopathology (pp. 39–70). New York: Guilford Press. Murray-Close, D., Nelson, D. A., Ostrov, J. M., Casas, J. F., & Crick, N. R. (2016). Relational aggression: A developmental psychopathology perspective. In D. Cohen, & D. Cicchetti (Eds.), Developmental psychopathology, genes and environment (pp. 660– 723). Hoboken, New Jersey: John Wiley & Sons, Inc. http://dx.doi.org/10.1002/ 9781119125556.devpsy413. Murray-Close, D., Ostrov, J. M., Nelson, D. A., Crick, N. R., & Coccaro, E. F. (2010). Proactive, reactive, and romantic relational aggression in adulthood: Measurement, predictive validity, gender differences, and association with intermittent explosive disorder. Journal of Psychiatric Research, 44, 393–404. http://dx.doi.org/10.1016/j. jpsychires.2009.09.005. Muthén, L. K., & Muthén, B. O (2015). Mplus User’s Guide. Los Angeles, CA: Muthén & Muthén. Nelson, D. A., & Coyne, S. M. (2009). Children's intent attributions and feelings of distress: Associations with maternal and paternal parenting practices. Journal of Abnormal Child Psychology, 37, 223–237. http://dx.doi.org/10.1007/s10802-008-9271-3. Nelson, D. A., Coyne, S. M., Swanson, S. M., Hart, C. H., & Olsen, J. A. (2014). Parenting, relational aggression, and borderline personality features: Associations over time in a Russian longitudinal sample. Development and Psychopathology, 26, 773–787. http:// dx.doi.org/10.1017/S0954579414000388. Nelson, D. A., Mitchell, C., & Yang, C. (2008). Intent attributions and aggression: A study of children and their parents. Journal of Abnormal Child Psychology, 36, 793–806. http://dx.doi.org/10.1007/s10802-007-9211-7. Nelson, J. A., & Perry, N. B. (2015). Emotional reactivity, self-control and children's hostile attributions over middle childhood. Cognition and Emotion, 29, 592–603. http://dx.doi.org/10.1080/02699931.2014.924906. Orobio de Castro, B., Veerman, J. W., Koops, W., Bosch, J. D., & Monshouwer, H. J. (2002). Hostile attribution of intent and aggressive behavior: A meta-analysis. Child Development, 73, 916–934. http://dx.doi.org/10.1111/1467-8624.00447. Ostrov, J. M., & Godleski, S. A. (2010). Toward an integrated gender-linked model of aggression subtypes in early and middle childhood. Psychological Review, 117, 233–242. http://dx.doi.org/10.1037/a0018070. Ostrov, J. M., & Godleski, S. A. (2013). Relational aggression, victimization, and adjustment during middle childhood. Development and Psychopathology, 25, 801–815. http://dx.doi.org/10.1017/S0954579413000187. Ostrov, J. M., & Houston, R. J. (2008). The utility of forms and functions of aggression in emerging adulthood: Association with personality disorder symptomatology. Journal of Youth and Adolescence, 37, 1147–1158. http://dx.doi.org/10.1007/s10964-0089289-4. Pornari, C. D., & Wood, J. (2010). Peer and cyber aggression in secondary school students: The role of moral disengagement, hostile attribution bias, and outcome expectancies. Aggressive Behavior, 36, 81–94. http://dx.doi.org/10.1002/ab.20336. Preacher, K. J., & Kelley, K. (2011). Effect size measures for mediation models: Quantitative strategies for communicating indirect effects. Psychological Methods, 16, 93–115. http://dx.doi.org/10.1037/a0022658. Pronk, R. E., & Zimmer-Gembeck, M. J. (2010). It's “mean,” but what does it mean to adolescents? Relational aggression described by victims, aggressors, and their peers. Journal of Adolescent Research, 25, 175–204. http://dx.doi.org/10.1177/ 0743558409350504. Pullmann, H., Raudsepp, L., & Allik, J. (2006). Stability and change in adolescents' personality: A longitudinal study. European Journal of Personality, 20, 447–459. http:// dx.doi.org/10.1002/per.611. Putallaz, M., Grimes, C. L., Foster, K. J., Kupersmidt, J. B., Coie, J. D., & Dearing, K.

96, 898–912. http://dx.doi.org/10.1037/a0014746. Knyazev, G. G. (2004). Behavioural activation as predictor of substance use: Mediating and moderating role of attitudes and social relationships. Drug and Alcohol Dependence, 75, 309–321. http://dx.doi.org/10.1016/j.drugalcdep.2004.03.007. Kokkinos, C. M., Antoniadou, N., Dalara, E., Koufogazou, A., & Papatziki, A. (2013). Cyber-bullying, personality and coping among pre-adolescents. International Journal of Cyber Behavior, Psychology and Learning, 3, 55–69. http://dx.doi.org/10.4018/ ijcbpl.2013100104. Kokkinos, C. M., & Voulgaridou, I. (2017). Relational and cyber aggression among adolescents: Personality and emotion regulation as moderators. Computers in Human Behavior, 68, 528–537. http://dx.doi.org/10.1016/j.chb.2016.11.046. Kokkinos, C. M., Voulgaridou, I., Mandrali, M., & Parousidou, C. (2016). Interactive links between relational aggression, theory of mind, and moral disengagement among early adolescents. Psychology in the Schools, 53, 253–269. http://dx.doi.org/10.1002/ pits.21902. Kokkinos, C. M., Voulgaridou, I., & Markos, A. (2016). Personality and relational aggression: Moral disengagement and friendship quality as mediators. Personality and Individual Differences, 95, 74–79. http://dx.doi.org/10.1016/j.paid.2016.02.028. Komarraju, M., Karau, S. J., Schmeck, R. R., & Avdic, A. (2011). The big five personality traits, learning styles, and academic achievement. Personality and Individual Differences, 51, 472–477. http://dx.doi.org/10.1016/j.paid.2011.04.019. Koolen, S., Poorthuis, A., & van Aken, M. A. (2012). Cognitive distortions and self-regulatory personality traits associated with proactive and reactive aggression in early adolescence. Cognitive Therapy and Research, 36, 776–787. http://dx.doi.org/10. 1007/s10608-011-9407-6. Krueger, R. F., & Tackett, J. L. (Eds.), (2006). Personality and psychopathology. New York: Guilford Press. Kunimatsu, M. M. (2010). The facets of hostile attributional bias: The importance of aggression subtypes and provocateur motivation. ((Master's Thesis). Retrieved from the University of New Orleans Theses and Dissertations. (Paper1286)). Kuppens, S., Grietens, H., Onghena, P., Michiels, D., & Subramanian, S. V. (2008). Individual and classroom variables associated with relational aggression in elementary-school aged children: A multilevel analysis. Journal of School Psychology, 46, 639–660. http://dx.doi.org/10.1016/j.jsp.2008.06.005. Lansford, J. E., Skinner, A. T., Sorbring, E., Giunta, L. D., Deater-Deckard, K., Dodge, K. A., ... Uribe Tirado, L. M. (2012). Boys' and girls' relational and physical aggression in nine countries. Aggressive Behavior, 38, 298–308. http://dx.doi.org/10.1002/ab. 21433. Leff, S. S., Waasdorp, T. E., & Crick, N. R. (2010). A review of existing relational aggression programs: Strengths, limitations, and future directions. School Psychology Review, 39, 508–535. (Retrieved from) http://www.ncbi.nlm.nih.gov/pmc/articles/ PMC3111222/. Lochman, J. E., & Wells, K. C. (2004). The coping power program for preadolescent aggressive boys and their parents: Outcome effects at the 1-year follow-up. Journal of Consulting and Clinical Psychology, 72, 571–578. http://dx.doi.org/10.1037/0022006X.72.4.571. Loflin, D. C., & Barry, C. T. (2016). ‘you can't sit with us’: Gender and the differential roles of social intelligence and peer status in adolescent relational aggression. Personality and Individual Differences, 91, 22–26. http://dx.doi.org/10.1016/j.paid.2015.11.048. Lynam, D. R., & Widiger, T. A. (2001). Using the five-factor model to represent the DSMIV personality disorders: An expert consensus approach. Journal of Abnormal Psychology, 110, 401–412. http://dx.doi.org/10.1037//0021-843X.110.3.401. Markos, A., & Kokkinos, C. M. (2017). Development of a short form of the big five questionnaire for children: Validation in preadolescent samples. Personality and Individual Differences, 112, 12–17. http://dx.doi.org/10.1016/j.paid.2017.02.045. Marsee, M. A., Barry, C. T., Childs, K. K., Frick, P. J., Kimonis, E. R., Munoz, L. C., ... Lau, K. S. (2011). Assessing the forms and functions of aggression using self-report: Factor structure and invariance of the peer conflict scale in youths. Psychological Assessment, 23, 792–804. http://dx.doi.org/10.1037/a0023369. Marsee, M. A., & Frick, P. J. (2007). Exploring the cognitive and emotional correlates to proactive and reactive aggression in a sample of detained girls. Journal of Abnormal Child Psychology, 35, 969–981. http://dx.doi.org/10.1007/s10802-007-9147-y. Marsee, M. A., Frick, P. J., Barry, C. T., Kimonis, E. R., Centifanti, L. C. M., & Aucoin, K. J. (2014). Profiles of the forms and functions of self-reported aggression in three adolescent samples. Development and Psychopathology, 26, 705–720. http://dx.doi.org/ 10.1017/S0954579414000339. Marsee, M. A., Weems, C. F., & Taylor, L. K. (2008). Exploring the association between aggression and anxiety in youth: A look at aggressive subtypes, gender, and social cognition. Journal of Child and Family Studies, 17, 154–168. http://dx.doi.org/10. 1007/s10826-007-9154-1. Mathieson, L. C., Murray-Close, D., Crick, N. R., Woods, K. E., Zimmer-Gembeck, M., Geiger, T. C., & Morales, J. R. (2011). Hostile intent attributions and relational aggression: The moderating roles of emotional sensitivity, gender, and victimization. Journal of Abnormal Child Psychology, 39, 977–987. http://dx.doi.org/10.1007/ s10802-011-9515-5. Maxwell, S. E., Cole, D. A., & Mitchell, M. A. (2011). Bias in cross-sectional analyses of longitudinal mediation: Partial and complete mediation under an autoregressive model. Multivariate Behavioral Research, 46, 816–841. http://dx.doi.org/10.1080/ 00273171.2011.606716. McCrae, R. R., Costa, P. T., Jr., Terracciano, A., Parker, W. D., Mills, C. J., De Fruyt, F., & Mervielde, I. (2002). Personality trait development from age 12 to age 18: Longitudinal, cross-sectional and cross-cultural analyses. Journal of Personality and Social Psychology, 83, 1456–1468. http://dx.doi.org/10.1037//0022-3514.83.6. 1456. McNiel, J. M., Lowman, J. C., & Fleeson, W. (2010). The effect of state extraversion on four types of affect. European Journal of Personality, 24, 18–35. http://dx.doi.org/10.

112

Journal of Applied Developmental Psychology 52 (2017) 101–113

C.M. Kokkinos et al.

org/10.1017/S0954579414000443. Tedeschi, J. T., & Felson, R. B. (1994). Violence, aggression, & coercive actions. Washington, DC: American Psychological Association. Vagos, P., Rijo, D., Santos, I. M., & Marsee, M. A. (2014). Forms and functions of aggression in adolescents: Validation of the Portuguese version of the peer conflict scale. Journal of Psychopathology and Behavioral Assessment, 36, 570–579. http://dx. doi.org/10.1007/s10862-014-9421-6. Vitaro, F., Brendgen, M., & Tremblay, R. E. (2002). Reactively and proactively aggressive children: Antecedent and subsequent characteristics. Journal of Child Psychology and Psychiatry, 43, 495–505. http://dx.doi.org/10.1111/1469-7610.00040. Voulgaridou, I., & Kokkinos, C. M. (2015). Relational aggression in adolescents: A review of theoretical and empirical research. Aggression and Violent Behavior, 23, 87–97. http://dx.doi.org/10.1016/j.avb.2015.05.006. Werner, N. E. (2012). Do hostile attribution biases in children and parents predict relationally aggressive behavior? The Journal of Genetic Psychology, 173, 221–245. http://dx.doi.org/10.1080/00221325.2011.600357. Werner, N. E., & Crick, N. R. (1999). Relational aggression and social–psychological adjustment in a college sample. Journal of Abnormal Psychology, 108, 615–623. http://dx.doi.org/10.1037/0021-843X.108.4.615. Werner, N. E., Senich, S., & Przepyszny, K. A. (2006). Mothers' responses to preschoolers' relational and physical aggression. Journal of Applied Developmental Psychology, 27, 193–208. http://dx.doi.org/10.1016/j.appdev.2006.02.002. Wilkowski, B. M., & Robinson, M. D. (2010). The anatomy of anger: An integrative cognitive model of trait anger and reactive aggression. Journal of Personality, 78, 9–38. http://dx.doi.org/10.1111/j.1467-6494.2009.00607.x. Willoughby, M., Kupersmidt, J., & Bryant, D. (2001). Overt and covert dimensions of antisocial behavior in early childhood. Journal of Abnormal Child Psychology, 29, 177–187. http://dx.doi.org/10.1023/A:1010377329840. Yeung, R. S., & Leadbeater, B. J. (2007). Does hostile attributional bias for relational provocations mediate the short-term association between relational victimization and aggression in preadolescence? Journal of Youth and Adolescence, 36, 973–983. http:// dx.doi.org/10.1007/s10964-006-9162-2. Yoon, J. S., Barton, E., & Taiariol, J. (2004). Relational aggression in middle school: Educational implications of developmental research. The Journal of Early Adolescence, 24, 303–318. http://dx.doi.org/10.1177/0272431604265681.

(2007). Overt and relational aggression and victimization: Multiple perspectives within the school setting. Journal of School Psychology, 45, 523–547. http://dx.doi. org/10.1016/j.jsp.2007.05.003. Raine, A., Dodge, K., Loeber, R., Gatzke-Kopp, L., Lynam, D., & Reynolds, & Liu, J. (2006). The reactive-proactive aggression questionnaire: Differential correlates of reactive and proactive aggression in adolescent boys. Aggressive Behavior, 32, 159–171. http:// dx.doi.org/10.1002/ab.20115. Ramirez, J. M. (2009). Some dichotomous classifications of aggression according to its function. Journal of Organisational Transformation & Social Change, 6, 85–101. http:// dx.doi.org/10.1386/jots.6.2.85/1. Runions, K. C., & Keating, D. P. (2010). Anger and inhibitory control as moderators of children's hostile attributions and aggression. Journal of Applied Developmental Psychology, 31, 370–378. http://dx.doi.org/10.1016/j.appdev.2010.05.006. Salmivalli, C., Ojanen, T., Haanpää, J., & Peets, K. (2005). ‘I'm OK but you're not’ and other peer-relational schemas: Explaining individual differences in children's social goals. Developmental Psychology, 41, 363–375. http://dx.doi.org/10.1037/00121649.41.2.363. Seibert, L. A., Miller, J. D., Pryor, L. R., Reidy, D. E., & Zeichner, A. (2010). Personality and laboratory-based aggression: Comparing the predictive power of the five-factor model, BIS/BAS, and impulsivity across context. Journal of Research in Personality, 44, 13–21. http://dx.doi.org/10.1016/j.jrp.2009.09.003. Simoncic, T. E., Kuhlman, K. R., Vargas, I., Houchins, S., & Lopez-Duran, N. L. (2014). Facebook use and depressive symptomatology: Investigating the role of neuroticism and extraversion in youth. Computers in Human Behavior, 40, 1–5. http://dx.doi.org/ 10.1016/j.chb.2014.07.039. Stoltz, S., Prinzie, P., De Haan, A., Londen, M., De Castro, B. O., & Deković, M. (2013). Child personality as moderator of outcome in a school-based intervention for preventing externalizing behaviour. European Journal of Personality, 27, 271–279. http:// dx.doi.org/10.1002/per.1892. Tackett, J. L., Daoud, S. L., De Bolle, M., & Burt, S. A. (2013). Is relational aggression part of the externalizing spectrum? A bifactor model of youth antisocial behavior. Aggressive Behavior, 39, 149–159. http://dx.doi.org/10.1002/ab.21466. Tackett, J. L., Kushner, S. C., Herzhoff, K., Smack, A. J., & Reardon, K. W. (2014). Viewing relational aggression through multiple lenses: Temperament, personality, and personality pathology. Development and Psychopathology, 26, 863–877. http://dx.doi.

113