Student: Jesse Karemakera,*
Advisors: Geert-Jan J.M. Stamsa, Lucres M.C. Jansenb, Tijs Jambroesb,
a University of Amsterdam, Department of Social and Behavioral Sciences: Forensic Orthopedagogics, Nieuwe Achtergracht 127, 1018 WS Amsterdam, The Netherlands
b VU University Medical Center, Department of Child and Adolescent Psychiatry, c/o De Bascule, P.O. Box 303, 1115 ZG Duivendrecht, The Netherlands
*corresponding author. Tel.: +31 6 24409255
Abstract
This study investigates whether hypothalamus-pituitary-adrenal (HPA) axis activity influences the relation between callous-unemotional (CU) traits and proactive (cold, emotionless, goal-focused) and reactive (explosive, emotional, not goal-oriented) aggression. The study hypothesized that the HPA axis acts as a moderator of the relation between CU traits and proactive aggression, but not for reactive aggression. CU traits, proactive and reactive aggression, and HPA axis activity were examined in a sample of 145 justice-involved adolescents. Using hierarchical multiple regression analyses, results showed an influence of CU traits on both types of aggression, but the HPA axis had no significant influence on either types. Future research needs to confirm these findings and examine alternative methods for assessing influential factors on aggression, specified in the discussion section.
Keywords: callous-unemotional traits, HPA axis, proactive aggression, reactive aggression, hierarchical multiple regression, adolescents.
Contents Introduction……….5 This study……….8 Method……….9 Participants……….9 Measures……….9 Statistical Analysis………10 Results………...10 Discussion……….13 References……….17
Introduction
Callous-unemotional (CU) traits constitute the affective component of psychopathy (Guelker, Barry, Barry, & Malkin, 2014). Adolescents with CU traits are known to have a lack of feeling guilt and remorse, a lack of concern for the feelings of others, a superficial and shallow
expression of emotions, and a lack of concern to perform well during important activities (Frick, 2009). CU traits are negatively associated with the effects of psychological and behavioral treatment and positively associated with severe problem behavior and aggressive behavior (Guelker et al., 2014). CU traits can fairly predict which violent youths persist in showing aggressive behavior into adulthood (Frick, Cornell, Barry, Bodin, & Dane, 2003). The present study examines the association between CU traits and aggressive behavior in justice-involved adolescents and the role the HPA axis plays in this relation.
Aggression can be defined as deliberate behavior aimed at physically and/or
psychologically inflicting damage on persons or property (Van Goozen, Fairchild, Snoek, & Harold, 2007) and is classified by looking at the function it serves (Hubbard, McAuliffe, Morrow, & Romano, 2010). Aggression can be divided into two types, the reactive type and the proactive/instrumental type (Blais, Solodukhin, & Forth, 2014). Reactive aggression can be described as an extreme emotional response to a perceived threat or provocation (Blais et al., 2014; Van Goozen, 2007). It is defensive, retaliatory (Hubbard et al., 2010), hostile (Van Goozen, 2007), and has no foreseeable external goal other than a direct emotional response (Blais et al., 2014). Proactive aggression is on the other side of the scale, and has an external goal that can be materialistic and/or socially beneficial to the offender (Hubbard et al., 2010). The aggression is controlled, instrumental, deliberate (Blais et al., 2014), predatory (Van Goozen, 2007) and seems emotionless (Blais et al., 2014).
Current research links CU traits to aggressive behavior on several other accounts, but the evidence for the association seems equivocal (Pardini, 2011). Pardini, Lochman and Frick (2003) found that youths with CU traits organize and interpret their predictions about events in their lives in a way that serves to see aggression as a reasonable solution to achieve positive results (Pardini et al., 2003). Adolescents with high levels of CU traits even do not seem to concern themselves with the negative influences their aggressive behavior has on others (Pardini & Byrd, 2012). Pardini (2011) links the elevated CU traits of juveniles with behavior that endorses social conflicts leading to violence by escalating conflict with peers to assert dominance and other social benefits. These associations were found even after controlling for prior violent behavior (Pardini, 2011). Blais and colleagues (2014) found a moderate and
significant relationship between CU traits and proactive and reactive aggression in their meta-analysis of 53 studies. Their conclusion was that a higher level of CU traits was associated with higher rates of proactive and reactive aggression (Blais et al., 2014). Fanti, Frick and Georgiou (2009) found that adolescents who have higher levels of CU show a combined form of reactive and proactive aggression instead of just proactive or reactive aggression. Frick and White (2008) actually went so far as saying that youths with CU traits designate a subgroup within antisocial adolescents that show more severe aggression and violence. The association between CU traits and aggressive behavior was even stronger for adolescents who used both proactive and reactive aggression (Frick & White, 2008). Overall, there seems to be a
consensus that higher levels of CU traits predict more aggressive behavior, but the strength of the influence seems to change from study to study.
Differences in the relation between CU traits and aggression may be explained by differences in the activity of the hypothalamus-pituitary-adrenal (HPA) axis. The HPA axis is an endocrine system that helps to adapt a human to bodily and contextual challenges by creating behavioral and psychological changes (Fries, Dettenborn, & Kirschbaum, 2009). These changes improve the person’s chances of adapting to changes in the environment (Fries et al., 2009). A reduced HPA axis may lead to adolescents seeking stimulation to seek arousal and therefore involve themselves in stressful and dangerous situations (Van Goozen et al., 2007). It also seems that the amygdala, a distinct part of the limbic system, plays a major role in a reduced HPA axis of adolescents with high levels of CU traits. The amydala coordinates emotional responses (Fries et al., 2009) and plays a key role in the perception of threat signals (Alink et al., 2008). Adolescents with high CU traits have a hypo-aroused amygdala that impairs their emotional responses, especially under stress (Gotisha et al., 2014). The amygdala of adolescents with high levels of CU traits inhibits the HPA axis activity during negative situations (Fries et al., 2009), which in turn lowers the chances of adapting to changes in the environment.
A reduced HPA axis may explain the difference between normally developing children and children who are at risk of showing aggressive behavior. While normally developing children have a wide range of responses to bodily and contextual challenges, the HPA axis of at-risk children develops in such a way that they constantly fail to respond to the same kind of challenges (Shirtcliff, Granger, Booth, & Johnson, 2005). This difference in the HPA axis may lead to children who are at risk in seeing aggression and other disruptive behavior as a good answer to bodily and contextual challenges (Shirtcliff et al., 2005). This is supported by the fact that delinquent and non-delinquent male adolescents with a disruptive behavior disorder
show an altered HPA axis that has significantly less activity than found in normally
developing adolescents (Popma et al., 2006; Popma et al., 2007b). In this paper, the HPA axis will be examined to explain the relation between CU traits and aggression.
To fully understand the association between the HPA axis on one side and aggression and callous-unemotional traits on the other, more needs to be said about how to investigate the HPA axis. To investigate the influence of the HPA axis, cortisol levels are usually examined to index HPA axis activity (Van Goozen, 2007). Cortisol is a product of the HPA axis, is
involved in stress reactivity (Dadds & Rhodes, 2008) and is crucial to help an adolescent adapt to visible challenges (Fries et al., 2009). The basal HPA axis activity follows a circadian rhythm with distinct pattern of several secretory episodes during which it produces cortisol. An episode in particular occurs after awakening, where there is a short, steep rise of cortisol levels that lasts until 20-30 minutes after awakening (Wilhelm, Born, Kudielka, Schlotz, & Wüst, 2007). This episode shows the highest cortisol levels during the entire day (Goozen, 2007) and is coined by the researchers as the cortisol awakening response (CAR) (Wilhelm et al., 2007). The CAR has been extensively used in several studies to reliably examine the activity of the HPA axis (Clow, Hucklebrigde, Stalder, Evans, & Thorn, 2010; Fries et al., 2009). Therefore, multiple studies successfully use the assessed cortisol levels to investigate the relation between HPA activity and behavior (Fries et al., 2009).
Low cortisol levels have been associated with externalizing problem behavior in adolescents (De Vries-Bouw et al., 2011; Platje et al., 2013a; Popma et al., 2007a; Popma et al., 2007b; Shirtcliff, Granger, Booth, & Johnson, 2005). Behavioral problems often occur during this time frame during middle childhood and adolescence (Walker, Walder, &
Reynolds, 2001). This period also shows maturational changes in the HPA axis activity, which have been shown to be related to externalizing problem behavior (Platje et al., 2013; Shirtcliff et al., 2005). Low cortisol seems to hold important clues to a reason why some youths express themselves in a more aggressive manner than other adolescents (Platje et al., 2013a; Platje et al., 2013b; Shirtcliff et al., 2005). The influence of cortisol on aggressive behavior is that adolescents with low cortisol may have a lack of fear and anxiety when bodily or
psychological challenges arise. This lack of fear and anxiety may lead to adolescents finding aggressive behavior more reliable in challenging situations, because the execution of such behavior requires a psychological state of low arousal (Raine, 2002). This finding is supported by the fact that adolescents who show chronic disruptive behavior have lower HPA axis activity than other adolescents (Popma et al., 2007a; Popma et al., 2007b) and show persistent aggressive behavior (Platje et al., 2013a; Shirtcliff et al., 2005). However, the association
between low cortisol levels and aggressive behavior has also been found in children and adolescents without chronic disruptive behavior (Platje, et al., 2013b; Spear, 2000). The association between relatively low cortisol levels and aggression is not always found in both sexes, as the study of Sondeijker and colleagues (2008) only found a relation between low cortisol levels and disruptive behavior in male adolescents. In her meta-analysis, Alink and collegues (2008) even found no significant relation between the HPA axis and externalizing behavior. Findings on the relation between the HPA axis and externalizing behavior are equivocal, but more research favors the influence of low HPA axis activity on aggressive behavior.
In the research on aggressive behavior among adolescents, youths with low HPA axis activity constitute a severe subgroup who develop antisocial behavior if they have high levels of CU traits (Hawes, Brennan, & Dadds, 2009). This is confirmed in the fact that adolescents with low cortisol levels show high and chronic levels of aggressive behavior (Platje et al., 2013), especially in adolescents with a high level of CU traits (Dadds & Rhodes, 2008). Biological studies show that the influence of low cortisol leads to adolescents showing more proactive aggression, but that reactive aggression might be influenced by another hormone (Dadds & Rhodes, 2008). Low HPA activity might also act as a moderator, which in turn facilitates the influence of CU traits on antisocial behavior (Shirtcliff et al., 2009).
This study
The present study investigates whether the activity of the HPA axis influences the relation between CU traits on the one hand and proactive and reactive aggression on the other. On several accounts CU traits were linked to aggressive behavior found in adolescents. Overall, higher CU traits predict more aggression, although differences exist in the collected evidence. This might be explained by the HPA axis activity, as low cortisol is especially found in adolescents with high levels of CU traits (Shirtcliff et al., 2009). In adolescence, the HPA axis rapidly develops along with the rest of the body. During this time frame, the HPA axis matures and influences externalizing problems and aggressive behavior (Platje, et al., 2013; Shirtcliff et al., 2005). It is possible that the difference in high and low activity of the HPA axis can explain the equivocal evidence found in the current literature.
The present study hypothesizes that the HPA axis moderates the relation between CU traits and aggressive behavior (Shirtcliff et al., 2009). Low activity of the HPA axis enables the influence of CU traits on more aggression in adolescents (Dadds & Rhodes, 2008; Hawes, et al., 2009; Shirtcliff et al., 2009). Adolescents with CU traits tend to show more proactive
than reactive aggression (Blais et al., 2014). Along with that, low cortisol seems mostly associated with proactive instead of reactive aggression (Dadds & Rhodes, 2008). Therefore, the present study further hypothesizes that the HPA axis only influences the relation between CU traits and proactive aggression. Reactive aggression is a form of aggression that is highly emotional (Blais et al., 2014), while superficial and shallow expression of emotions are found in adolescents with CU traits (Frick, 2009). Because of the differences in emotional
influences, the present study hypothesizes that the influence of the HPA axis does not affect the relation between CU traits and reactive aggression. This study also examines whether the relation between CU traits, the HPA axis and aggression is dependent on gender, since there are gender differences in expressing aggressive behavior during adolescence ((Loeber, Slot, Van der Laan, & Hoeve, 2008; Shirtcliff, et al., 2005).
Method Participants
The participants of this study were selected from adolescents who were admitted to a facility for compulsory residential youth care in the Netherlands between 2007 and 2013. After admittance into the facility, diagnostic information was collected within the first six weeks. All participants and their parents did have knowledge of the diagnostic evaluations and consented in its use for research. This study uses part of the diagnostic information that was collected with two self-report questionnaires and one neurobiological test. All 145 adolescents in the sample were included in this study. The sample consisted of 49,1% males and all participants were between 12 and 17 years old (mean age = 15.1, SD = 1.3).
Measures
The measures described here were part of an admittance requirement to collect diagnostic information before the start of the treatment. Only the measures that are relevant for this study are described here.
Youth Psychopathic traits Inventory. The Youth Psychopathic traits Inventory (YPI) is
a self-report questionnaire that contains 50 items that asses psychopathic characteristics that are divided among three domains. The domain that is used in this study is the affective domain, also referred to as Callous-Unemotional (CU). This instrument uses a 4-point Likert scale (1 = does not apply at all, 2 = does not apply well, 3 = applies fairly well, 4 = applies
very well). The Cronbach's Alpha for Callous-Unemotional was .81 in an adolescent
community sample (Andershed, Hodgins, & Tengström, 2007) and .77 in a delinquent sample (Skeem & Cauffman, 2003). In the present study Cronbach's Alpha for Callous-unemotional was .81.
Reactive–Proactive Aggression Questionnaire. The Reactive–Proactive Aggression
Questionnaire (RPQ) is a self-report questionnaire that contains 26 items that asses
aggression, which is divided in two aggression types, namely proactive aggression (13 items) and reactive aggression (13 items). This instrument uses simple instructions that help to facilitate non-defensive responses from a participant by acknowledging that every person feels angry from time to time. The items were rated by using a 3-point Likert scale (0 = never, 1 = sometimes, 2 = often) to assess the frequency of different aggressive behavioral outcomes. The Cronbach’s Alpha in an adolescent male sample was .84 for proactive aggression and .89 for reactive aggression (Raine et al., 2006). In this study Cronbach's Alpha for reactive aggression was .90 and for proactive aggression is .90.
Cortisol Awakening Response. The Cortisol Awakening Response (CAR) indicates the
activity of the HPA axis (Hellhammer et al., 2007). The CAR is used to reliably measure the HPA axis activity by collecting saliva to determine the amount of cortisol (Fries et al., 2009). In total, three saliva samples are taken to determine the CAR. The first saliva sample is taken at the moment an adolescent awakens. The second and third sample are taken thirty minutes and sixty minutes after awakening. Each sample consisted of 0.1 ml saliva that was contained in a Sarstedt Salivette. Thereafter, the samples were stored in a freezer at -20 degrees Celsius until analysis. At the Endocrinology Laboratories of the University Medical Centre Utrecht, the cortisol in the saliva was measured using an in-house competitive radio-immunoassay employing a polyclonal anticortisol-antibody (K7348). [1,2-3H(N)]-Hydrocortisone (PerkinElmer NET396250UC) was used as a tracer with a lower limit of detection of 1.0 nmol/l. Two measures of the CAR were computed. First, the area under the curve for the increase of cortisol after awakening (AUCi) was computed to determine the reactivity of the
HPA axis (Hellhammer et al., 2007). Second, the Area under the response curve in respect to the ground (AUCg) to determine the basal activity of the HPA axis (Clow et al., 2010).
Statistical Analysis
The present study hypothesizes that the HPA axis plays a moderating role in the relation between CU traits and aggressive behavior. To analyze this question, firstly the correlations between all the variables will be determined to analyze if there exists a relationship between the dependent and independent variables (Brace, Kemp, & Snelgar, 2012). After these
analyses, hierarchical multiple regression (HMR) will be used to answer the main hypothesis. A four stage HMR will be conducted with reactive and proactive aggression as dependent variables. In the first stage, gender will be entered as a control variable. The level of CU traits in adolescents will be entered at the second stage, followed by the AUCg or the AUCi and the
interaction between CU traits and the AUCg/i. This analysis is conducted to determine if the
HPA axis moderates the relation between CU traits and the two types of aggression, controlling for gender (Brace et al., 2012).
Results
Prior to the main analyses, statistical analyses revealed no differences in callous-unemotional traits, aggression and cortisol levels between gender or age. In Table 1 the descriptives of the independent and dependent variables are shown. Prior to the hierarchical multiple regression, the necessary assumptions were tested. The sample consisted of 145 adolescents, which seems adequate for the examination of three independent variables. The assumption of singularity was also met, since no independent variables were a combination of other independent variables.
Table 1
Descriptives
Factors N Mean Std. Deviation
Callous-unemotional traits 145 15.72 7.65
AUCg 79 1033.33 334.16
AUCi 79 186.13 214.99
Proactive aggression 132 6.43 5.67
Reactive Aggression 132 12.26 5.69
Table 2 shows the Pearson correlations between all variables, and clearly shows that only CU traits and the two aggression types were significantly correlated with each other. The
assumption of multicollinearity was met, since independent variables were not significantly correlated with each other. Scatter plots showed that the assumptions of normality, linearity and homoscedasticity, were all met.
Table 2
Pearson Correlations between factors
1. 2. 3. 4. 5. 1. Callous-unemotional traits - -.14 -.23 .50** .45** 2. AUCg - -.62** -.11 .05 3. AUCi - -.10 -.08 4. Proactive Aggression - .75** 5. Reactive Aggression -** P<0.001, *p<0.05
A four stage hierarchical multiple regression was conducted with proactive and reactive aggression separately as a dependent variable. For both regressions, gender was entered in the first step to act as a control variable, so that differences between adolescents were not dependent on gender. CU traits were added in the second step to see if the variable significantly influences both independent variables. The CAR was added into the third step, followed in the fourth step by the independent variable that shows the interaction between CU
traits and either the AUCg or the AUCi. The regression statistics for both regressions are
shown in Table 3 and Table 4. Table 3
Hierarchical Multiple Regression for proactive aggression
Variable β T sr2 R R2 ∆R2 AUCg/AUCi Step 1 0.11 0.01 .01 Gender -.07 -.66 .01 Step 2 0.50 0.25 .24** Gender -.07 -.61 .01 Callous-unemotional Traits .50 4.73 .24** AUCg Step 3 0.50 0.25 .00 Gender -.07 -.61 .00 Callous-unemotional Traits .49 4.54 .23** AUCg -.03 -.24 .00 Step 4 0.50 0.25 .00 Gender -.06 -.56 .00 Callous-unemotional Traits .49 4.46 .23** AUCg -.03 -.25 .00 CU traits* AUCg .02 .13 .00 AUCi Step 3 .50 .25 .00 Gender -.07 -.66 .01 Callous-unemotional Traits .50 4.53 .23** AUCi .01 .13 00 Step 4 .51 .26 .00 Gender -.07 -.69 .01 Callous-unemotional Traits .49 4.33 .22** AUCi -.00 -.03 .00 CU traits* AUCi -.06 -.49 .00
** P<0.001, *p<0.05, The best model to ascertain the influence on proactive aggression is in bold.
The hierarchical multiple regression analyses revealed that gender differences did not contribute significantly to the regression model. In step two, CU traits contributed
significantly to the regression model for proactive aggression, F (2,67) = 11.32, p < .05) and accounted for 24% of the variation in proactive aggression. CU traits also contributed to the regression model for reactive aggression, F (2,67) = 9.11, p < .05) and accounted for 21% of the variation in reactive aggression. For both aggression types, both the HPA axis and its interaction with CU traits did not contribute significantly to the regression model and
accounted for almost no variation in both aggression types. Only CU traits were a significant predictor of both types of aggression.
Table 4
Hierarchical Multiple Regression for reactive aggression
Variable β T sr2 R R2 ∆R2 AUCg/AUCi Step 1 .09 .01 .01 Gender .09 .71 .01 Step 2 .46 .21 .21** Gender .12 1.11 .01 Callous-unemotional Traits .46 4.20 .21** AUCg Step 3 .47 .22 .01 Gender .10 .95 .01 Callous-unemotional Traits .47 4.26 .21** AUCg .09 .84 .01 Step 4 .47 .22 .00 Gender .11 .95 .01 Callous-unemotional Traits .47 4.12 .21** AUCg .09 .81 .01 CU traits* AUCg .02 .16 .00 AUCi Step 3 .46 .21 .00 Gender .12 1.11 .01 Callous-unemotional Traits .46 4.10 .20** AUCi .02 .02 .00 Step 4 .47 .22 .00 Gender .12 1.11 .02 Callous-unemotional Traits .47 4.07 .20** AUCi .03 .29 .00 CU traits* AUCi .04 .35 .00
Discussion
The present paper investigated whether the activity of the hypothalamus-pituitary-adrenal (HPA) axis moderated the relation between callous-unemotional (CU) traits and aggression. This was tested among justice-involved adolescents. During this research, aggression was divided in two different types. The first type is called proactive aggression and is cold, emotionless and goal-focused (Blais et al., 2014; Hubbard et al., 2010; Van Goozen, 2007). Reactive aggression is the second type and is rash, explosive, extremely emotional and not goal-oriented (Blais et al., 2014; Hubbard et al., 2010; Van Goozen, 2007). The current study hypothesized that the HPA axis acts as a moderator of the relation between CU traits and proactive aggression, but not for reactive aggression. It can be derived from the results that, although the influence of CU traits on both aggression types is clear, the HPA axis plays no observable role. This conclusion was reached while controlling for gender, which also showed no significant influence.
According to the hypothesis about the non-influence of the HPA axis on the change in reactive aggression, the results seem to confirm the hypothesis. The level of CU traits of adolescents influences their level of reactive aggression, but the HPA axis has no influence on this association. Although previous studies did find an influence on aggression, their results showed only effects that could not be solely pinned on just one type of aggression (Dadds & Rhodes, 2008; Hawes, et al., 2009; Shirtcliff et al., 2009). Dadds and Rhodes (2008) believe that the relation between CU traits and reactive aggression is not influenced by cortisol, but by another hormone, called serotonin. Therefore, the results in this study in this sample do not come as a surprise. The fact that the HPA axis activity has no influence on the relation
between CU traits and proactive aggression does.
According to Dadds and Rhodes (2008), cortisol is a hormone that influences stress reactivity in adolescents. Low levels of stress in negative situations, such as conflicts and confrontations, are associated with more proactive aggression (Dadds & Rhodes, 2008). It was therefore hypothesized that it is possible that the HPA axis activity has a negative relation with proactive aggression. This means that low HPA axis activity should reinforce the
influence of high levels of CU traits on a high level of proactive aggression among adolescents. In this sample, however, this was not the case.
The results of this study show no influence of the HPA axis. This finding contradicts other studies. As said earlier in response to the findings on the relation between CU traits and reactive aggression, other studies only found effects that were not solely based on just one
type of aggression (Dadds & Rhodes, 2008; Hawes, et al., 2009; Shirtcliff et al., 2009). For reactive aggression, these studies, along with the influence of serotonin (Dadds & Rhodes, 2008), provide ample possibility that the hypothesis concerning reactive aggression might not be confirmed. A problem would arise if the HPA axis had no influence on the relation between CU traits and proactive aggression, since this was a possibility (Shirtcliff et al., 2009). Since this is the case in this sample, a more thorough explanation is required.
This study first tried to justify these findings by providing evidence that gender
differences accounted for the result that the HPA axis had no influence on the relation between CU traits and both types of aggression. Since Sondeijker and colleagues (2008) only found influence of the HPA axis on aggressive behavior in male adolescents, it could provide a possibility. However, these results were found in neither genders when analyzed separately. The HPA axis did still not have an effect on the relation between CU traits and both types of aggression.
Another reason why this study did not find the hypothesized results, might be because of the sample. This study’s sample consists of justice-involved adolescents who all show high levels of aggression and external problem behavior. Since all adolescents have serious
behavioral problems, almost no significant correlations between variables were found. This leads to almost no variation between the adolescents of this sample. It might be possible that no link between the HPA axis, CU traits and aggression was found, since all adolescents might have an underdeveloped HPA axis.
A third reason why the HPA had no significant influence might be because of sleep-related factors that influence the CAR. For instance, the cortisol secretion during the CAR is light-sensitive (Clow et al., 2010). This means that the amount of cortisol in an adolescent not only depends on stress, but also on the amount of light during the morning hours (Leproult, Colecchia, L’Hermite-Baleriaux, & Van Cauter, 2001). The light during the awakening phase of an adolescent enhances the dynamic of the CAR (Clow et al., 2010; Leproult et al., 2001) and in turn has an influence on the results. However, all youths were in a facility for
compulsory residential youth care that provides identical sleeping conditions. The possibility of the influence of light in the morning hours on the CAR seem low. There is a second sleep related influence. It is possible that some youths woke up earlier than expected before taking the saliva samples to determine the CAR. This means that some samples might have been collected too late and do not represent the CAR accurately. These samples might therefore have altered the data in such a way that no relations between CU traits, HPA axis activity and aggression was found. However, this does not seem likely. All results from the saliva samples
were examined so that individual samples did not fall too far from the overall mean. If one sample seemed inaccurate, it was thrown out of the analyses.
A fourth and last reason why no significant results were found might be due to the reasoning behind the answers given in the self-report questionnaires. In the research of Boonmann and colleagues (2013), delinquent and non-delinquent adolescents were compared on psychopathic traits, including CU traits, measured with the self-report questionnaire YPI. In this study, it appeared that delinquent adolescents with psychopathic traits score lower on the YPI than the general population. This finding gives rise to the conclusion that delinquent youths might answer questions regarding negative traits in a socially desirable way to avoid negative impressions from people that influence the length of their sentences (Boonmann et al., 2013). It seems possible that the adolescents in this sample answered favorably to avoid negative impacts on the length of their stay in the facility. Therefore, this possibility might have influenced the results, as only two self-report questionnaires and one neurobiological test were used in this study. Boonmann and colleagues (2013) did report that although the scores on the questionnaires were lower, there were significant correlations between their variables. In this study, significant correlations between CU traits and both types of aggression were found. This leads to a conclusion that although the scores might be lower than they actually should be, the questionnaires did measure what they were supposed to measure. Future research concerning CU traits, the HPA axis and aggression does need to take into account that using different informants may lead to different results than were found in this study. For instance, using different informants to measure CU traits of adolescents may show clearer results as to a more realistic level of CU traits. A possible solution is to use the Antisocial
Process Screening Device, which allows using different informants to measure CU traits
(Falkenbach, Poythress, & Heide, 2003). Another solution might be to follow these
adolescents over time to separate youths who consequently show high levels of CU traits and those who show a different developmental trajectory. These different kind of adolescents can be compared with each other to better determine the influence of the HPA axis. By using a longitudinal design, there is also less influence of one single measurement by having a bigger scope than with a cross-sectional design (Bijleveld, 2009). Altogether, this means that a longitudinal design gives a higher chance of finding a relation between traits, the HPA axis and aggression.
Although several reasons were named, the conclusion in the present study still stands. This study found that the HPA axis activity had no significant influence on the relation between CU traits and aggression. This conclusion indicates that the HPA axis may not have
clinical relevance for adolescents with high levels of CU traits and proactive and reactive aggression. Since CU traits only accounted for 20-25% of the difference in aggression found in adolescents, there are still many pieces of the puzzle that might help understand the causes of the aggressive behavior in adolescents. This study contributed to confirm one influential factor and disconfirm another. The puzzle might seem easier, but it will take a lot more research to map out the direct influences on aggressive behavior in adolescents.
References
Alink, L. R., van IJzendoorn, M. H., Bakermans‐Kranenburg, M. J., Mesman, J., Juffer, F., & Koot, H. M. (2008). Cortisol and externalizing behavior in children and adolescents:
Mixed meta‐analytic evidence for the inverse relation of basal cortisol and cortisol reactivity with externalizing behavior. Developmental psychobiology, 50, 427-450. Andershed, H., Hodgins, S., & Tengström, A. (2007). Convergent validity of the youth
psychopathic traits inventory (YPI) association with the psychopathy checklist: youth version (PCL: YV). Assessment, 14, 144-154.
Blais, J., Solodukhin, E., & Forth, A. E. (2014). A Meta-Analysis Exploring the Relationship Between Psychopathy and Instrumental Versus Reactive Violence. Criminal Justice
and Behavior, 41, 797-821. doi:10.1177/0093854813519629.
Bijleveld, C. C. J. H. (2009). Methoden en technieken van onderzoek in de criminologie. Boom Koninklijke Uitgevers: Den Haag.
Boonmann, C., Jansen, L. M., Lisette, A., Vahl, P., Hillege, S. L., Doreleijers, T. A., & Vermeiren, R. R. (2013). Self-reported psychopathic traits in sexually offending juveniles compared with generally offending juveniles and general population youth.
International journal of offender therapy and comparative criminology, XX(X), 1-11.
Brace, N., Kemp, R., & Snelgar, R. (2012). SPSS for Psychologists. 5th edition. Palgrave
Macmillan: London.
Clow, A., Hucklebridge, F., Stalder, T., Evans, P., & Thorn, L. (2010). The cortisol awakening response: more than a measure of HPA axis function. Neuroscience & Biobehavioral
Reviews, 35, 97-103.
Dadds, M. R., & Rhodes, T. (2008). Aggression in young children with concurrent callous– unemotional traits: can the neurosciences inform progress and innovation in treatment approaches?. Philosophical Transactions of the Royal Society B: Biological Sciences,
363, 2567-2576.
De Vries-Bouw, M., Popma, A., Vermeiren, R., Doreleijers, T. A., Van De Ven, P. M., & Jansen, L. (2011). The predictive value of low heart rate and heart rate variability during stress for reoffending in delinquent male adolescents. Psychophysiology, 48, 1597-1604.
Duits, N., & Bartels, J.A.C. (2011). Jeugdpsychiatrie en recht; wetgeving, zorg en praktijk. Assen: Van Gorcum.
Falkenbach, D. M., Poythress, N. G., & Heide, K. M. (2003). Psychopathic features in a juvenile diversion population: reliability and predictive validity of two self‐report
measures. Behavioral sciences & the law, 21, 787-805.
Fanti, K. A., Frick, P. J., & Georgiou, S. (2009). Linking callous-unemotional traits to instrumental and non-instrumental forms of aggression. Journal of Psychopathology
and Behavioral Assessment, 31, 285-298.
Frick, P. J. (2009). Extending the construct of psychopathy to youth: implications for
understanding, diagnosing, and treating antisocial children and adolescents. Canadian
Journal of Psychiatry/Revue canadienne de psychiatrie, 31, 803-812.
Frick, P. J., Cornell, A. H., Barry, C. T., Bodin, S. D., & Dane, H. E. (2003). Callous- unemotional traits and conduct problems in the prediction of conduct problem severity, aggression, and self-report of delinquency. Journal of abnormal child
psychology, 31, 457-470.
Frick, P. J., & White, S. F. (2008). Research review: The importance of callous‐unemotional traits for developmental models of aggressive and antisocial behavior. Journal of
Child Psychology and Psychiatry, 49, 359-375.
Fries, E., Dettenborn, L., & Kirschbaum, C. (2009). The cortisol awakening response (CAR): facts and future directions. International Journal of Psychophysiology, 72, 67-73. Gostisha, A. J., Vitacco, M. J., Dismukes, A. R., Brieman, C., Merz, J., & Shirtcliff, E. A.
(2014). Beyond physiological hypoarousal: The role of life stress and
callous-unemotional traits in incarcerated adolescent males. Hormones and behavior, 65, 469-479.
Guelker, M. D., Barry, C. T., Barry, T. D., & Malkin, M. L. (2014). Perceived positive outcomes as a mediator between adolescent callous-unemotional traits and antisocial behavior. Personality and Individual Differences, 69, 129-134.
Hawes, D. J., Brennan, J., & Dadds, M. R. (2009). Cortisol, callous-unemotional traits, and pathways to antisocial behavior. Current opinion in psychiatry, 22, 357-362.
Hellhammer, J., Fries, E., Schweisthal, O. W., Schlotz, W., Stone, A. A., & Hagemann, D. (2007). Several daily measurements are necessary to reliably assess the cortisol rise after awakening: state-and trait components. Psychoneuroendocrinology, 32, 80-86. Hubbard, J. A., McAuliffe, M. D., Morrow, M. T., & Romano, L. J. (2010). Reactive and
proactive aggression in childhood and adolescence: Precursors, outcomes, processes, experiences, and measurement. Journal of personality, 78, 95-118.
Leproult, R., Colecchia, E. F., L’Hermite-Balériaux, M., & Van Cauter, E. (2001). Transition from Dim to Bright Light in the Morning Induces an Immediate Elevation of Cortisol Levels. The Journal of Clinical Endocrinology & Metabolism, 86, 151-157.
Loeber, R., Slot, N. W., Van der Laan, P. H., & Hoeve, M. (2008). Conclusions and
recommendations.In R. Loeber, N. W. Slot, P. H. Van der Laan & M. Hoeve (Eds.),
Tomorrow's criminals: The development of child delinquency and effective interventions (pp.261-284). Hampshire: Ashgate Publishing Ltd.
Pardini, D. A. (2011). Perceptions of social conflicts among incarcerated adolescents with callous‐unemotional traits: ‘You’re going to pay. It’s going to hurt, but I don’t care.’.
Journal of child psychology and psychiatry, 52, 248-255.
Pardini, D. A., & Byrd, A. L. (2012). Perceptions of aggressive conflicts and others’ distress in children with callous‐unemotional traits:‘I’ll show you who’s boss, even if you suffer and I get in trouble’. Journal of child psychology and psychiatry, 53, 283-291. Pardini, D.A., Lochman, J.E., & Frick, P.J. (2003). Callous unemotional traits and
social-cognitive processes in adjudicated youths. Journal of the American Academy of Child
and Adolescent Psychiatry, 42, 364–371.
Platje, E., Jansen, L., Raine, A., Branje, S. J., Doreleijers, T. A., de Vries-Bouw, M., Popma, A., Van Lier, P.A.C., Koot, H.M., Meeus, W.H.J., & Vermeiren, R. R. (2013a).
Longitudinal associations in adolescence between cortisol and persistent aggressive or rule-breaking behavior. Biological psychology, 93, 132-137.
Platje, E., Vermeiren, R. R. J. M., Raine, A., Doreleijers, T. A. H., Keijsers, L. G. M. T., Branje, S. J. T., Popma, A., Van Lier, P.A.C., Koot, H.M., Meeus, W.H.J., & Jansen, L. M. C. (2013b). A longitudinal biosocial study of cortisol and peer influence on the development of adolescent antisocial behavior. Psychoneuroendocrinology, 38, 2770-2779.
Popma, A., Jansen, L., Vermeiren, R., Steiner, H., Raine, A., Van Goozen, S. H., Van Engeland, H., & Doreleijers, T. A. (2006). Hypothalamus pituitary adrenal axis and autonomic activity during stress in delinquent male adolescents and controls. Psychoneuroendocrinology, 31, 948-957.
Popma, A., Vermeiren, R., Geluk, C. A., Rinne, T., van den Brink, W., Knol, D. L., Jansen, L.M.C., Van Engeland, H., & Doreleijers, T. A. (2007a). Cortisol moderates the relationship between testosterone and aggression in delinquent male adolescents. Biological psychiatry, 61, 405-411.
Popma, A., Doreleijers, T. A., Jansen, L. M., Van Goozen, S. H., Van Engeland, H., & Vermeiren, R. (2007b). The diurnal cortisol cycle in delinquent male adolescents and normal controls. Neuropsychopharmacology, 32, 1622-1628.
Raine, A. (2002). Biosocial studies of antisocial and violent behavior in children and adults: A review. Journal of abnormal child psychology, 30, 311-326.
Raine, A., Dodge, K., Loeber, R., Gatzke-Kopp, L., Lynam, D., Reynolds, C., Stouthamer-Loeber, M., & Liu, J. (2006). The Reactive-Proactive Aggression Questionnaire: Differential correlates of reactive and proactive aggression in adolescent boys.
Aggressive behavior, 32, 159-171.
Shirtcliff, E. A., Granger, D. A., Booth, A., & Johnson, D. (2005). Low salivary cortisol levels and externalizing behavior problems in youth. Development and Psychopathology, 17, 167-184.
Shirtcliff, E. A., Vitacco, M. J., Graf, A. R., Gostisha, A. J., Merz, J. L., & Zahn‐Waxler, C. (2009). Neurobiology of empathy and callousness: implications for the development of antisocial behavior. Behavioral sciences & the law, 27, 137-171.
Skeem, J. L., & Cauffman, E. (2003). Views of the downward extension: Comparing the youth version of the Psychopathy Checklist with the Youth Psychopathic traits Inventory. Behavioral Sciences and the Law, 21, 737-770.
Sondeijker, F. E., Ferdinand, R. F., Oldehinkel, A. J., Tiemeier, H., Ormel, J., & Verhulst, F. C. (2008). HPA‐axis activity as a predictor of future disruptive behaviors in young adolescents. Psychophysiology, 45, 398-404.
Spear, L. P. (2000). The adolescent brain and age-related behavioral manifestations.
Neuroscience & Biobehavioral Reviews, 24, 417-463.
Van Goozen, S. H., Fairchild, G., Snoek, H., & Harold, G. T. (2007). The evidence for a neurobiological model of childhood antisocial behavior. Psychological Bulletin, 133, 149-182.
Walker, E. F., Walder, D. J., & Reynolds, F. (2001). Developmental changes in cortisol secretion in normal and at-risk youth. Development and Psychopathology, 13, 721-732.
Wilhelm, I., Born, J., Kudielka, B. M., Schlotz, W., & Wüst, S. (2007). Is the cortisol awakening rise a response to awakening?. Psychoneuroendocrinology, 32, 358-366.
