Graduate School of Childhood Development and Education
Running head: AGGRESSION AND ANS RESPONSES
Autonomic Nerve System Responses to
Emotion-Eliciting Film Clips Portraying Sadness in
Multiproblem Young Male Adolescents:
Associations with Proactive and Reactive
Aggressive Behavior
Research Master Child Development and Education Thesis 2
Marika Kouvelis, 10080910 September, 2015
Supervisors:
mw. dr. Machteld Hoeve
Research Institute of Child Development and Education, University of Amsterdam Josjan Zijlmans, MSc. & mw. dr. Reshmi Marhe
Abstract
Low autonomic nervous system (ANS) activity has been claimed to be associated with aggressive behavior. This study examined the association between responses of the ANS to sadness, and proactive aggression (PA) and reactive aggression (RA) in multiproblem young male adults. Heart rate (HR) and skin conductance (SC) were monitored at rest and in response to film clips eliciting sadness. PA and RA were measured with a self-report
questionnaire: Dutch Reactive-Proactive Aggression Questionnaire (RPQ). Results suggested that HR is associated with PA in multiproblem young adults; a higher score on PA predicted lower HR. Examining the difference in ANS responses between baseline and sadness in association with PA and RA, the results showed that aggression is not a predictor for ANS responses to sadness in this sample of multiproblem young adults.
Keywords: heart rate, skin conductance, ANS responses, proactive aggression, reactive aggression, multiproblem, young adults, sadness
Multiproblem Young Adults
Adolescence is a developmental period characterized by an increase in risk behaviors. For most adolescents, engagement in some level of risk behavior is age-appropriate and part of the normal development process (Mun, Windle, Schainker, 2008). In this developmental period major changes in psychological, societal (Blokland, Palmen, & Van San, 2012), and neurobiological functioning (Crone, 2009) occur that are critical to healthy development towards adulthood. Arnett (2000) proposed that young adults (in the general population) are undergoing a separate developmental stage, called ‘emerging adulthood’. Emerging adulthood is a period between 18 and 25 years of age where adolescents make the transition from young adulthood into adulthood. This period is featured by the transition into adulthood because of an increase in independence and an exploration of various life possibilities (Arnett, 2000). How well an adolescent makes the transition depends in large part on the right balance of the adolescent pushing for independence and whether parents and society give the adequate amount of support (Arnett, 2000). In addition, limitations in educational and occupational opportunities influence the extent to which young people can experience their late teens and twenties as a volitional period.
Opportunities tend to be less widely available for multiproblem young adults, who are characterized by severe psychosocial problems in comparison with their peers in the general population (Hammink & Schrijvers, 2013). Therefore, this group may be less likely to experience ages 18-25 as a period of independent exploration of possible life directions (Arnett, 2000). The majority of multiproblem young adolescents demonstrate antisocial behavior, including aggressive behavior (AW-DNK, 2015). As a result, there has been considerable interest in biological processes that are partly heritable and may act as markers for antisocial behavior including aggression. There is ample evidence that several biological mechanisms are associated with aggression. Interest in neurobiological markers has increased
substantially, which has resulted in, for example, examination of the main human stress-regulating systems such as the autonomic nervous system (ANS; Popma, 2011; Vries-Bouw, Popma, Vermeiren, Doreleijers, Van de Ven, & Jansen, 2011).
The objective of the present study is to better understand the biology underlying aggression and the emotional deficits related with aggression in a specific group of multiproblem young adults. Although there is ample evidence that several biological
mechanisms underlie aggression, there is limited knowledge on the association in this specific group. With the knowledge gained in this study, it will be possible to examine risk profiles of multiproblem young adults, based on neurobiological characteristics. The results might ultimately contribute to the knowledge base needed to expand for treatment programs to be more specific and thus more effective. The present study will likely lead to a better
understanding of biological mechanisms that are associated with aggression and related emotional deficits. Answering the research question can help inform intervention and prevention research on aggressive behavior in multiproblem young adults.
Multiproblem Young Adults and Aggression
Multiproblem young adults may face extra challenges during the transition from childhood to adulthood, as they may not have been able to attain this developmental stage (Doreleijers & Fokkens, 2010). The problems of these multiproblem young adults can comprise for example psychiatric disorders, substance abuse, unemployment, homelessness, low educational attainment, a poor social network, and crime. Most of them have a criminal history and often this antisocial behavior persists into later adulthood (Van Domburgh, Vermeieren, Blokland, & Doreleijers, 2009). This occurrence and persistence of antisocial behavior is one of the most prominent issues in multiproblem young adults, and its societal impact can be large (Hammink & Schrijvers, 2013). A recently conducted pilot study by The Academic Workplace at De Nieuwe Kans (AW-DNK, 2015) indicated that 52% of the
multiproblem male young adults meet the criteria for the psychiatric condition called antisocial personality disorder (ASPD). In the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR, 4TH ed; American Psychiatric Association, 2000) ASPD is
characterized by a pervasive disregard for, and violation of, other people’s rights and encompasses many heterogeneous behaviors and traits, including aggression (APA, 2000).
Motives for aggression have traditionally been divided into two general categories: proactive and reactive aggression (Crick & Dodge, 1996). Proactive aggression (PA)
represents goal-directed behavior motivated by an external reward (Card & Little, 2007). This type of aggression has been characterized as instrumental, premeditated, and ‘cold-blooded’, with little evidence of autonomic arousal (Card & Little, 2007; Crick & Dodge, 1996; Hoeve et al., 2015). In comparison, reactive aggression (RA) has been designated as a combative response to a perceived threat. It is performed in response to provocation and is characterized as defensive, emotional, ‘warm-blooded’, and impulsive (Card & Little, 2007; Hoeve et al., 2015). One person can show both PA and RA (Raine et al., 2006). Two primary theoretical models can explain both types. Whereas RA is rooted in the frustration-anger theory of aggression (Dollard et al., 1939, cited by Vitaro et al., 2006; Scarpa et al., 2010), PA can be explained by the social learning theory (Bandura, 1973). The first theory states that a person immediately reacts to the anger-frustration stimulus and injures the perpetrator of the real or perceived threat or provocation. In contrast, according to the social learning theory, PA is often planned or pre-mediated, and rather than an act of frustration, this type of aggressive behavior is thought to be regulated by learned reinforcement (Scarpa et al., 2010).
Aggression and Heart Rate
A mechanism that has been shown to be strongly related to aggression is stress regulation (Summers & Winberg, 2006). Two dominant theories of stress regulation in relation to aggression have been proposed. They both explain the relationship between
reduced ANS arousal and the development of aggression; the fearlessness theory and the sensation-seeking theory. The fearlessness theory (Raine, 1993) posits that the low level of arousal noted in testing situations can be taken as evidence of a lack of normal fear. Low levels of arousal do not indicate rest, but instead indicate the lack of anxiety and fear. In this theory, learning by punishment is less successful due to the fact that low arousal causes the experience of negative emotions to be weaker than normal and thus a person will be less likely to avert from an action causing negative emotions. The autonomic underarousal reflects lack of anxiety and fear, and this is assumed to predispose one to aggressive behavior because such behavior requires a degree of fearlessness to execute. Fearlessness, especially in
childhood, might account for poor socialization because low fear of punishment would likely reduce the effectiveness of behavioral conditioning (Schiffer, Rao, & Fogel, 2003). An alternative to the fearlessness theory is the sensation-seeking theory (Zuckerman, 1979). This theory presumes that low arousal represents an aversive physiological state. This observed hypoarousal is experienced by affected individuals as unpleasant and is compensated for with risk-taking/thrill-seeking behaviors. Therefore, individuals with antisocial behavior seek out stimulation in order to restore their arousal levels back to an optimal or normal level (Raine, 1996). According to this theory, the state of low reactivity is aversive; therefore, increased stimulation is sought and this may lead to risk seeking, criminality, and aggression. In line with both aforementioned theories (fearlessness and sensation-seeking) it is thus expected that basal ANS activity and ANS responsivity to psychosocial stress would be attenuated in
multiproblem young adults.
A system that is very important in stress regulation is the ANS (re)activity (Kreibig, 2010). ANS (re)activity can be assessed by measuring heart rate. Cross-sectional studies have shown low resting heart rate to be a consistent biological correlate of antisocial behavior as shown in a meta-analysis conducted by Ortiz and Raine (2004). Their review demonstrated
that resting heart rate appears to be the best-replicated biological correlate to date of antisocial behavior in children and adolescents. Antisocial individuals have a lower heart rate, either during rest or during a stressor, as compared to healthy controls. A strong effect size was revealed for low heart rate during a stressor in relation to antisocial behavior. These results indicate that antisocial individuals lack an optimal level of arousal in order to appropriately deal with (stressful) situations. In line with these results, Raine, Venables, and Williams (1990) conducted a prospective study using heart rate responses of participants at age 15 to predict criminality at age 24. Results claimed that criminals had lower resting heart rate than noncriminals. Another study found no difference in resting heart rate between antisocial children and normal controls (De Wied, Van Boxtel, Posthumus, Goudena, & Matthys, 2009). In addition, several longitudinal studies found a relation between lower heart rate and future antisocial behavior or delinquency (De Vries-Bouw, et al., 2011; Raine, Venables, &
Mednick, 1997; Raine, Venables, & Williams, 1995). In sum, heart rate can be linked to antisocial behavior, including aggression.
ANS (re)activity can also be assessed by measuring skin conductance (SC). Gao, Raine, Venables, Dawson and Mednick (2010) found that low SC conditioning at 3-year-olds was associated with aggression at the age of eight, and criminal behavior 20 years later. Baker, Shelton, Baibazarova, Hay, & Van Goozen (2013) discovered that low SC in infancy predict aggressive behavior at age 3. The study of Raine, Venables, and Williams (1990) claimed that criminals had lower skin conductance activity than noncriminals
In conclusion, low autonomic arousal appears to be linked with aggression. These findings suggest that differences in individuals in the level of resting autonomic arousal and autonomic reactivity may be predictive of recidivism and other treatment outcome measures. It is therefore important to examine whether ANS measures add information to the risk profile of recidivism in a group of multiproblem young adults.
The Current Study
Young adults (18-27) have rarely been studied as a separate group (D’Oosterlinck, Broekaert, & Van der Heaghen, 2006; Osgood, Foster, & Courtney, 2010). So far, they were mainly included in studies that investigated adults from ages 18 to approximately 65 (Osgood et al., 2010). Although it has been shown that (re)activity of the ANS is correlated with aggression, there is limited knowledge on the association between young adults’ reactive and proactive aggression and responses of the ANS. Given that the majority of multiproblem male young adolescents demonstrate antisocial behavior (AW-DNK, 2015), it is substantially relevant to study the predictive value of ANS (re)activity for persistence of aggressiveness in this specific group. To the best of our knowledge ANS data on young male multiproblem adolescents is currently missing. Hence, this study will focus on this specific age group (18-27) with a multiproblem background.
The aim of the current study is to examine the predictive value of heart rate and skin conductance in resting and in response to emotion-eliciting (sadness) film clips. Accordingly, this study will examine the association between multiproblem young male adults’ reactive and proactive aggression and heart rate as well as skin conduction. This leads to the following research question: To what extent are measures of aggression (i.e., proactive and reactive aggression) associated with autonomic responses (measured by heart rate and skin conductance) to sadness in multiproblem young adults (aged 18-27)?
The present study will focus on ANS responses associated with the emotion sadness. There is some evidence that aggressive individuals are impaired in reacting to emotional stimuli (Blair, 2007; Schechtman, 2003). Sadness is an emotion that plays a crucial role in adaption (Darwin, 1872; Ekman, 1984, 1999). It is commonly described as a passive emotion and occurs when a goal can no longer be achieved, as when one is separated from a valued person or object, or when one loses a sense of control. Sadness can be experienced as feelings
of loneliness, discouragement, rejection and dissatisfaction with oneself (Ekman 1999; Kreibig, Wilhelm, Roth & Gross, 2007). Although emotions in general have been connected to urgent action tendencies (Frijda, 1986), sadness appears to be associated with a decrease in responsivity to environment because it is characterized by an enhanced parasympatic tone causing HR deceleration (Kreibig, Wilhelm, Roth & Gross, 2007). In general, we expect HR to decelerate in response to sadness. In addition, on the basis of the results of Scarpa, Haden, and Tanaka (2010) we expect that in this sample of multiproblem male young adults, reactive aggression will be related to higher HR, whereas proactive aggression will be related to lower HR in association with sadness.
Little is known about aggressive individuals and their SC response to sadness. Khalfa, Isabelle, Jean-Pierre, and Manon (2002) studied changes in skin conductance resulting from listening to fearful, happy, sad, and peaceful musical excerpts. Their results showed that higher SC was associated with fearful and happy musical excerpts but they did not find any differences in SC response to sad excerpts. Other studies revealed a decrease in SC during sad music (Krumhansl, 1997; Salimpoor, Benovoy, Longo, Cooperstock, & Zatorre, 2009, Van der Zwaag, Westerink & Van den Broek, 2011). Emotionally powerful music tends to
increase skin conductance more than less emotionally powerful music (Rickard, 2004). Thus, in general SC rises when responding to powerful emotions, but SC response to sadness is still ambiguous.
Based on the aforementioned fearlessness theory of Raine (1993), the current hypothesis is that reactive aggressive individuals are hypersensitive to emotional stimuli, whereas proactive aggressive individuals tend to be hyposensitive to such emotional stimuli (Blair, 2010). In other words, it is expected that lower HR is associated with higher self-reported proactive aggressive behavior. In addition, reactive aggression is expected to be related to HR in the opposite way, indicating that higher HR is associated with higher
self-reported reactive aggressive behavior. SC is expected to be low in aggressive individuals (Baker et al., 2013).
Method Participants
Participants were 44 male young adults (between ages 18 and 27, M = 21.82 years, SD = 2.50). The sample was derived from the original sample of 352 multiproblem young adults who participated in a large ongoing cohort study of AW-DNK, which will eventually include a group of 500 participants. For the neurobiological study, within a subgroup (n = 150), EEG measures, fMRI measures and ANS measures will be collected. The subsample for the current study consisted of the first 44 participants from the neurobiological study and can be seen as a pilot study, used as a pre-test or try out for the complete neurobiological study group.
The participants had severe psychosocial problems, occurring in several important life areas. For example, they were not employed, and not enrolled in formal education.
Approximately 80% had an ethnic minority background (AW-DNK, 2015). The following are criteria for inclusion and exclusion. The men had to be between the age of 18 and 27. They needed to have a sufficient knowledge of the Dutch language to understand the study procedure and to fill out the questionnaires. Moreover, participants needed to have a Self-Sufficiency Matrix (SSM; Hammink & Schrijvers, 2013) score of ‘muliproblematic’, which means a score 1 or 2 on the domains Income and Daily activities, a score of 1, 2 or 3 on the domains Justice or Addiction or Mental Health or Social Network, a score 3, 4, or 5 on the domain Physical Health (AW-DNK, 2014). Participants were excluded from participation when having an IQ-score below 70. IQ was assessed using the Information, Block Design, Arithmetics and Digit Symbol-Coding subtests of the Wechsler Adult Intelligence Scale for the Netherlands (WAIS-III-NL; Wechsler, 1997). Characteristics of the sample are included in Table 1.
Procedure
The participants were recruited via a municipal facility for adolescents and young adults, called Het Jongerenloket (JOLO) in Rotterdam, the Netherlands. JOLO helps adolescents and young adults with problems on several primary life domains, such as receiving income, housing, and finding a daytime activity (work, school). All multiproblem young adults referred to a treatment program by JOLO or already in treatment, were asked to participate.
The Self-Sufficiency Matrix (SSM) was administered at JOLO, as part of the standard procedure, to all young adults at intake. If the candidate had an SSM score based on the ‘multiproblematic’ definition of AW-DNK (2014), he was eligible to be included in the study. He was asked whether he agreed to receive information about the research project. If he did, the youth care professional at JOLO referred the candidate to the researcher and the
researcher personally explained the study to the candidate. The candidate also received an information letter. If the candidate was willing to participate, informed consent was obtained and the study procedure started. All data were handed anonymously and marked only by an identification code consisting of three numbers for each individual. The key to this code was stored separately from the collected data on a computer in the VUmc department for child and adolescent psychiatry and was accessible only to the research team by use of a password. The Medical Ethics Committee of the VU University Medical Centre Amsterdam approved the study.
Materials
Relaxing video. In order to measure the autonomic nervous system in rest,
participants watched a five-minute excerpt from an aquatic relaxing video (Coral Sea Dreaming, Small World Music Inc.; De Wied, Van Boxtel, Matthys, & Meeus, 2013). This has been shown to be a more effective manner of lowering cardiovascular activity levels than
simply resting quietly (Piferi, Kline, Younger, & Lawler, 2000). To ensure recovery from emotional arousal induced by film clips, participants also viewed 1-minute excerpts from the same aquatic video prior to each emotional film clip.
Emotional film clips. In addition to the resting state measurements, an
emotion-eliciting task was administered. The emotion-emotion-eliciting task consisted of watching two
adaptations of film clips presenting the emotion sadness. Film clips are powerful in capturing attention because of their dynamic display that includes both visual and auditory modalities (Fancsali, 2011). The two film clips that were used to evoke emotional reactions were one sadness clip involving a boy whose father is dying (The Champ; Zeffirelli, 1979) and one involving a boy who fails at selection training for a soccer tournament (Mohammed). The clips were assembled from documentary films (broadcast by the Dutch public broadcasting companies) and pretested in a pilot study, where more than 95% of the adolescents agreed upon which prominent emotion was represented (De Wied, Van Boxtel, Matthys, & Meeus, 2012; Gross & Levenson, 1995). The film clips varied in length (Champ = 99s, Mohammed = 147s). While the complete clip of Champ was taken as a target, in the clip of Mohammed a target of 34s (1:51 - 2:25) was taken. In this target scene the story character portrayed intense facial and vocal expressions of emotion. Two different film clips were chosen to compare how similar the ratings for the two film clips were. Heart rate and skin conductance level were analyzed during the target episodes and during the five-minute aquatic video, which served as baseline. Average heart rate and skin conductance level during the different conditions was compared. In total, the measurements took approximately ten minutes.
Apparatus and stimulus presentation. Participants were individually tested in a
laboratory room at the Erasmus University of Rotterdam. The room contained a testing area and observation unit partitioned by a one-way screen. Both units were equipped with a personal computer, one for stimulus presentation (in the test room) and one for online control
and storage of data collection (in the observation unit). Both computers were connected to a small portable digital recorder for the preprocessing and storage of physiological data that was attached to the participant; VU-Ambulatory Monitoring System (VU-AMS; De Geus, Willemsen, Klaver, & van Doornen, 1995; Willemsen, De Geus, Klaver, van Doornen, & Carroll, 1996). Using the VU-AMS, heart rate in rest and as reaction to the video clips
portraying sadness was measured. Heart rate was continuously measured during the entire test session as an index of autonomic activity. The two film clips were digitized and presented in randomized order: all participants with an odd identification code first watched the film clip Champ and all participants with an even identification code first watched the film clip
Mohammed. The VU-AMS also generated signals marking onset and offset of each film clip, which were stored on the VU-AMS recorder.
Psychophysiological measures. Heart rate (HR) and skin conductance (SC) were
measured using the VU-AMS during the acclimation period (relaxing video) and the emotional film clips. Indices of HR (in beats per minute; bpm) were derived from the
electocardiogram (ECG), using the VU-AMS. Mean HR and SC were determined during the baseline and target periods of each film clip. Resting HR and SC were calculated during the 300s presentation of the aquatic film clip. SC level (measured in microsiemens; µS) was collected through leads on the thenar and hypothenar eminences of the non-dominant hand and average over 30s intervals. The sampling rate of the VU-AMS for collecting SC data is 2 Hz and a resolution of 0.0125 µS. SC reflects sympathetic functioning.
Reactive-Proactive Aggression Questionnaire (RPQ). Aggression was examined by
the Dutch version of the Reactive-Proactive Aggression Questionnaire (RPQ), a 23-item self-report questionnaire (Raine et al., 2006). All questions were read out loud to the participants by the researcher and answers were marked at the questionnaire by the researcher. It took approximately ten minutes to administer. The RPQ consisted of two subscales (reactive and
proactive aggression; Raine et al., 2006). These subscales were used to confirm the presence of the different types of aggression. For both the original and the Dutch version, a two-factor structure has been found to be the best fitting model, internal consistency for the subscales and the total scale has been shown to be good to excellent (Cronbach’s alpha ≥ .83; Cima, Raine, Meesters, & Popma, 2013; Raine et al., 2006). Cronbach’s alpha for this study was .902, which is considered to be very high.
Data analysis
Prior to conducting the main analyses of the study, initial analyses were performed using the IBM SPSS program to inspect the data. First, descriptives were checked to verify plausible values of the participants on each of the variables. Second, the assumptions for the variables were checked. The outcome variables in this study are heart rate (HR) and skin conductance (SC) and the predictor variable is aggression, specified in PA and RA. To get an impression of normality, histograms of the residuals were inspected. The shape of the
histogram of the variable SC indicated skewness; SC data were skewed to the right. Therefore, data transformation was needed (Tabachnick & Fidell, 2013). To reduce the extreme skewness, the variable SC was logarithmically transformed. The other residual distributions reflected normality. Scatterplots revealed linear relationships between every predictor and the outcome variable. Verification of homogeneity was done using the residuals of the model, by plotting them. The residual variation was similar. Tests to see if the data met the assumption of collinearity indicated that multicollinearity was not a concern (PA,
Tolerance = .53, VIF = 1.89; RA, Tolerance = .53, VIF = 1.89). In addition, homogeneity of variance in the different relationships was confirmed using Levene’s test. Levene’s test as well as Box’s test was non-significant for all dependent variables, meaning that the assumption of homogeneity of variance was met. The histogram of standardized residuals indicated that the data contained approximately normally distributed errors, as did the normal
P-P plot of standardized residuals, which showed points that were not completely on the line, but close. The scatterplot of standardized residuals showed that the data met the assumptions of homogeneity of variance and linearity. The data also met the assumption of non-zero variances. In summary, the assumptions were satisfied. Descriptive characteristics of the participants are included in Table 1.
In order to answer the research question, a number of analyses were performed. First, to examine the difference in HR and SC between baseline and both the film clips, paired samples t-tests were calculated. In this way the autonomic nerve system responses to the film clips portraying sadness for all young adults (aggressive and non-aggressive) were
determined. Second, linear regression analysis was used to examine the association between the variables PA and RA on the one hand and HR and SC during the baseline relaxation video on the other hand. Third, a linear regression analysis was used to examine to what extent the variables PA and RA were associated with autonomic responses (HR and SC) to sadness. For each clip the difference score of HR and SC between film clip and baseline was calculated. The HR response was expressed as the difference score between mean HR during the target and baseline period. Probabilities of all tests were two-tailed and a significance level of .05 was adopted in all tests.
Results Autonomic Responses to Target Film Clips
As displayed in Table 2, there are statistically significant differences between baseline and target scores for HR in Champ and Mohammed, and for SC in Champ but not for those in Mohammed. First, there was a significant difference between HR at baseline (M = 64.4, SD = 8.57) and HR at the target movie Champ (M = 63.0, SD = 8.15) conditions, t(43) = 3.29, p = .002. In line with this result, there was also a significant difference between SC at the baseline (M = 5.2, SD = 3.44) and SC at the target Champ (M = 5.8, SD = 4.30) conditions, t(43) =
-2.54, p = .015. Therefore, the null hypothesis of equal means between baseline and Champ was rejected. One unit higher on HR Champ indicated 3.29 units higher HR at the baseline. Thus, the baseline HR was significantly higher and the baseline SC was significantly lower than the HR and SC during the target movie Champ. In comparison, results of the second paired samples t-test showed that SC differed between baseline and target movie Mohammed (M = 5.2, SD = 3.45); t(43) = -2.24, p = .030. In contrast with the target movie Champ, HR at baseline did not significantly differ from HR at the target movie Mohammed. Table 2 shows the results from the paired samples t-tests performed in this study.
Overall, results show that HR decreased for the target movie Champ, and that SC increased after exposure to both the target movies. In conclusion, these results suggest that the target movie Champ did have an effect on the autonomic responses. Specifically, when the target movie Champ was showed, HR decreased and SC increased. The target movie Mohammed had an effect on SC but not on HR. When the target movie Mohammed was shown, SC increased.
Predicting Value of Aggression on Autonomic Responses
Linear regressions were conducted to examine if the level of PA and RA predicted autonomic nerve system responses on the baseline movie. Simple linear regressions were conducted to determine if HR and SC were associated with the levels of PA and RA. The analysis showed that RA level did not significantly predict the value of HR, F(1,42) = 1,492, p = .229, neither did RA predict the value of SC, F(1,42) = 1,034, p = .315. In line with these results, the level of PA also did not significantly predict SC value, F(1,42) = 1,006, p = .322. However, PA significantly predicted HR, F(1,42) = 5.041, p = .030. Participants’ average HR decreased -.583 bpm for each unit increase on the level of PA. Approximately 11% (R2 = .107) of the variation in HR was predicted by the level of PA. According to Cohen (1988), this suggests a small to medium effect. In conclusion, PA did significantly predict the level of
HR but not SC during relaxation, while RA did not significantly predict the autonomic nerve system responses HR and SC.
Association Aggression and Autonomic Responses to Sadness
First, when HR response to sadness was predicted with the target movie Champ it was found that PA, F(1,42) = .449, p = .507, and RA F(1,42) = .003, p = .959, were both
nonsignificant. The same conclusion was drawn with the target movie Champ when SC response to sadness was predicted by PA, F(1,42) = .002, p = .976 and RA, F(1,42) = .222, p = .640. Second, identical analyses were conducted to interpret the effect of the target movie Mohammed where similar results were found. When HR response to sadness was predicted it was found that PA, F(1,42) = 3.982, p = .052, and RA, F(1,42) = .217, p = .644, were both nonsignificant predictors. Also for this target movie, the same conclusion could be drawn when SC response to sadness was predicted by PA, F(1,42) = .275, p = .603, and RA, F(1,42) = .281, p = .599. To conclude, none of the aforementioned analyses resulted in a significant prediction of HR or SC to sadness by PA and RA.
Discussion
The main goal of this study was to examine HR and SC in response to emotion eliciting film clips portraying sadness in multiproblem young male adolescents and the associations with proactive and reactive aggressive behavior. Therefore, first the effects of sadness on the ANS were measured by comparing HR and SC at baseline with HR and SC during the two filmclips portraying sadness. When the target movie Champ was showed, HR decreased and SC increased. When the target movie Mohammed was shown, HR did not change but SC increased as well. Second, the association between autonomic arousal at rest (baseline) and the two subtypes of aggressive behavior, RA and PA, was inspected. Results showed that a higher score on PA predicted a lower heart rate. Third, results indicated that aggression did not interact with sadness in predicting HR and SC.
Consistent with the hypotheses, both sadness clips significantly increased SC. The film clip Champ also significantly decreased HR, but no significant difference emerged between the baseline and the film clip Mohammed. These findings support the hypothesis that film clips portraying sadness are associated with lower HR and higher SC. The film clip from Champ did have an effect on ANS responsivity, but the Mohammed clip did not affect HR. This finding suggests that the film clip of Mohammed does not represent the prominent emotion sadness in the same way as the film clip Champ does. In other words, losing a loved one seems to elicits more sadness than failing at selection training for a soccer tournament.
A higher score on PA predicted a lower HR. This indicates that multiproblem male young adults with relatively high levels of proactive aggressive behavior are more likely to show lower heart rate at rest. Based on the sensation seeking theory and the fearlessness theory (Raine, 2002), it was hypothesized that lower resting autonomic arousal would be associated with higher levels of aggressive behavior. The present study’s finding is consistent with this prediction. This finding supports the suggestion that low resting HR may reflect an aversive state of underarousal and that this group of multiproblem male young adults engages in proactive aggressive behaviors in an effort to increase their physiological arousal to more comfortable levels.
In contrast, the results did not support the hypothesis that heightened aggression would predict lower levels of resting SC. The association of SC and aggression was nonsignificant. A possible explanation for this finding is that aggressive individuals have a lower SC in general (Baker et al., 2013; Gao et al., 2010), regardless the type of aggressiveness (RA or PA). It is possible that there is a negative relation between aggressive behavior and SC (as well as for HR) compared to a control group with almost no aggressive behavior. This sample already scored quite high on aggression and so there was little variation in order to find a relationship between aggression and SC. Because a control group was not assessed in the
present study, we cannot examine this possible explanation. However, this would be interesting for future research.
Multiproblem young male adults who had an increased level of proactive and reactive aggression were not found to have a decreased level of the ANS in reaction to sadness. Contrary to the expectations, aggression did not interact with sadness in predicting HR and SC. The results of these analyses supported the findings of Khalfa, Isabelle, Jean-Pierre, and Manon (2002), who also found a lack of significant difference in SC in response to sadness. They explain these results by the fact that sadness is not a strongly arousing emotion, and therefore it may not heighten the level of SC in the same way as stronger arousing emotions such as happiness and fear might do (Khalfa, Isabelle, Jean-Piere, & Manon, 2002). Further, there was no significant HR difference in response to sadness within this group, neither with PA nor with RA. This finding was not consistent with predictions that HR would decelerate in response to sadness (Kreibig, Wilhelm, Roth, & Gross, 2007). These findings suggest that although a film clip portraying sadness is showed, there was no reaction in HR and SC. In other words, PA and RA are not predictors of ANS responses to sadness in this sample. An explanation might be that in delinquent samples such as in the current study, there may be less variance in antisocial and aggressive behavior compared to general population samples, making differences smaller and therefore harder tot detect.
Despite the strengths of the present study, including a well-defined sample of
multiproblem male adolescent with a higher score on PA showing significantly lower HR, a number of important limitations must be acknowledged. One of the most important
limitations is the lack of scientific evidence for the construct ‘multiproblematic’. The
researchers of the Academic Workplace (AW-DNK, 2014) defined this term themselves and this raises questions about the validity. Questions concern for example the weight given to the dimensions of the Self-Sufficiency Matrix (SSM) (AW-DNK, 2014): Is a participant with a
score of 1 on Justice (indicating frequent contact with police and/or due process in court) but 4-5 on the other life domains more or less multiproblematic comparing to a participant scoring 3 on all domains of life? It is unknown whether the inter-rater reliability and validity of the SSM is conducted by the research team of AW-DNK and what the possible results would be (2014). There is a need to determine inter-rater reliability and validity of the SSM in order to support the use of the construct ‘muliproblematic’.
Second, the sample size was relatively small. Although in neurobiological research n = 44 is quite large, the power to detect significant relations between variables was limited. In addition, we studied a specific population of multiproblem young male adolescents. Although studying such a specific group has evident relevance, results cannot be generalized to other samples. For instance, because the current study included only male adolescents, the findings require replication with female samples before they can be generalized to female adolescents with a multiproblem background. Future research including both males and females would help to understand the influence of gender in multiproblem young adults’ HR and SC in response to sadness. Besides, greater confidence in these results would be warranted if they were replicated with larger, more diverse, and more representative samples. In addition, our sample was relatively complicated with probably a big diversity in multiproblem background. Psychophysiological factors show stronger relationships to aggression among people from benign social backgrounds than among those from difficult ones (Baker et al., 2013; Raine, 2002). It is therefore possible that the association between HR and SC and aggression in this sample was not consistently found, because the present study focused on a multiproblematic group. Future research should examine whether the observed effects generalize to different populations, that is, with different backgrounds. As known from the literature, psychosocial factors interact with psychophysiological risk factors, and antisocial behavior increases exponentially when social and biological risk factors combine (Raine, 2002). Further studies
on aggressive behavior should focus on multiple factors, neurobiological as well as
psychological and social factors. Research should therefore also incorporate features of the participant’s environment, including for example parenting behavior, peers, education and SES into the study of ANS responses and aggressive behavior. Nevertheless, the present study has explored a relevant field; young adults had thus far rarely been studied as a separate group.
Third, the fact that the experimenter instead of the participants themselves selected the film clips is a limitation of this study’s design. This meant that appreciation of, and familiarity with the film clips was not controlled. To be sure that the participants would feel actual (self reported) sadness, they could bring their own selected film clip (Van der Zwaag, Westerink, & Van den Broek, 2011). Future research should further investigate the impact of sadness on ANS responses using personally selected film clips. However, in such a small sample, using two selected clips helped to keep the results constant, and because of the differences in reactions to the two clips, some insight is gained in the effect of the separate film clips.
Fourth, the data on aggressive behavior were collected using a self-reporting questionnaire (RPQ; Raine et al., 2006). Responses to questions about behavior may be biased by individuals’ willingness to self-disclose their feelings, and their desire to present themselves in a socially desirable way. Future research could include questionnaires filled in by their social workers and/or therapist (if present) and, for example, recorded and registered information from the Judicial Documentation System (JDS), with approval of the Ministry of Security and Justice. The JDS provides data on convictions of juvenile offenders (AW-DNK, 2014). Notwithstanding, results of a study done by Burt and colleagues (2015) revealed that youths rated themselves higher on questionnaires measuring aggression than did their parents, indicating that although information from only one informant was used, the most favorable informant was chosen.
Despite the limitations of the study, the results are an important contribution toward advancing knowledge of the role of aggressive behavior on ANS measures in response to sadness. In conclusion, the current study produced two important findings. First, we confirmed the relation between proactive aggression and a significant lower heart rate in a sample of multiproblem young adults. Second, we demonstrated that multiproblem young male adults with higher levels of reactive or proactive aggression showed relatively similar responses to film clips portraying sadness as those with lower levels of aggression.
References
Academische Werkplaats bij De Nieuwe Kans [AW-DNK] (2014). Multiproblem Young Adults: a study on service use trajectories, neurobiological mechanisms and treatment effect in young adults with severe criminal, psychiatric and social problems. Research Protocol for Institutional Review Board VUmc. Amsterdam, The Netherlands.
Academische Werkplaats bij De Nieuwe Kans [AW-DNK] (2015). Academische Werkplaats. Factsheet resultaten pilot. Retrieved from:
http://www.aw-dnk.nl/wp- content/uploads/2015/03/Factsheet-Academische-Werkplaats-bij-De-Nieuwe-Kans.pdf
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (DSM-IV-TR, 4th ed., text revision). Washington: American Psychiatric. Arnett, J. J. (2000). Emerging adulthood: A theory of development from the late teens
through the twenties. American Psychologist, 55, 469–480. doi:10.1037//0003- 066X.55.5.469
Baker, E., Shelton, K. H., Baibazarova, E., Hay, D. F., & Van Goozen, S. H. (2013). Low skin conductance activity in infancy predicts aggression in toddlers 2 years later. Psychological science, 24, 1051-1056. doi:10.1177/0956797612465198
Bandura, A., 1973. Aggression: A Social Learning Analysis. Prentice-Hall: Oxford England. Blair, R. J. R. (2010). Neuroimaging of psychopathy and antisocial behavior: a targeted
review. Current Psychiatry Reports, 12(1), 76-82. doi:10.1007/s11920-009-0086-x Blokland, A., Palmen, H., & Van San, M. (2012). Crimineel gedrag in de jongvolwassenheid.
Tijdschrift voor Criminologie, 54(2), 85–98.
Brendgen, M., Vitaro, F., Tremblay, R. E., & Lavoie, F. (2001). Reactive and proactive aggression: Predictions to physical violence in different contexts and moderating
effects of parental monitoring and caregiving behavior. Journal of Abnormal Child Psychology, 29(4), 293-304.
Burt, S. A., Rescorla, L. A., Achenbach, T. M., Ivanova, M. Y., Almqvist, F., Begovac, I., ... & Verhulst, F. C. (2015). The association between aggressive and non-aggressive antisocial problems as measured with the Achenbach System of Empirically Based Assessment: A study of 27,861 parent–adolescent dyads from 25 societies.
Personality and Individual Differences, 85, 86-92.
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(5), 466-480.
Cima, M., Raine, A., Meesters, C., & Popma, A. (2013). Validation of the Dutch Reactive Proactive Questionnaire (RPQ): Differential correlates of reactive and proactive aggression from childhood to adulthood. Aggressive Behavior, 39(2), 99–113. doi:10.1002/ab.21458
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale. NJ: Erlbaum.
Crone, E. A. (2009). Executive functions in adolescence: inferences from brain and behavior. Developmental Science, 12(6), 825–830. doi:10.1111/j.1467-7687.2009.00918.x D’Oosterlinck, F., Broekaert, E., & Vander Haeghen, C. (2006). Probleemjongeren te vroeg
het tehuis uit? Tijdschrift voor Orthopedagogiek, Kinderpsychiatrie, en Klinische Kinderpsychologie, 31, 58–68.
Darwin, C. (1872). The expression of the emotions in man and animals. London: Albemarle. De Geus, E. J. C., Willemsen, G. H. M., Klaver, C. H. A. M., & van Doornen, L. J. P. (1995).
Biological psychology, 41(3), 205–27. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/8608201
De Wied, M., Van Boxtel, A., Matthys, W., & Meeus, W. (2012). Verbal, facial and autonomic responses to empathy-eliciting film clips by disruptive male adolescents with high versus low callous-unemotional traits. Journal of Abnormal Child
Psychology, 40(2), 211–23. doi:10.1007/s10802-011-9557-8
De Wied, M., Van Boxter, A., Posthumus, J. A., Goudena, P. P., & Matthys, W. (2009). Facial EMG and heart rate responses to emotion-inducing film clips in boys with disruptive behavior disorders. Psychophysiology, 46, 996-1004.
Dodge, K. A., Lochman, J. E., Harnish, J. D., Bates, J. E., & Pettit, G. S. (1997). Reactive and proactive aggression in school children and psychiatrically impaired chronically assaultive youth. Journal of Abnormal Psychology, 106, 37–51.
Dollard, J., Doob, L., Miller, N., Mowrer, O., Sears, R., 1939. Frustration and Aggression. Yale University Press, New Haven, CT.
Doreleijers, T. A. H., & Fokkens, J. W. (2010). Minderjarigen en jongvolwassenen: Pleidooi voor een evidence based strafrecht. Rechtstreeks, 2, 9–47.
Fancsali, S.E. (2011). Variable construction for predictive and causal modeling of online
education data. In: Proceedings of the 1st International Conference on Learning
Analytics and Knowledge (54-63). NY, USA: ACM New York.
Gao, Y., Raine, A., Venables, P. H., Dawson, M. E., & Mednick, S. A. (2010). Reduced electrodermal fear conditioning from ages 3 to 8 years is associated with aggressive behavior at age 8 years. Journal of Child Psychology and Psychiatry, 51, 550–558. Gross, J., & Levenson, R. (1995). Emotion elicitation using films. Cognition & Emotion, 9(1),
87–108. Retrieved from
Hammink, A., & Schrijvers, C. (2013). Klantprofielen van kwetsbare jongeren die zich melden bij het Jongerenloket. Rotterdam.
Hoeve, M., Colins, O. F., Mulder, E. A., Loeber, R., Stams, G. J. J. M., & Vermeiren, R. R. J. M. (2015). The association between childhood maltreatment, mental health problems, and aggression in justice-involved boys. Aggressive Behavior. Advance online
publication. doi: 10.1002/AB.21586
Khalfa, S., Isabelle, P., Jean-Pierre, B., & Manon, R. (2002). Event-related skin conductance responses to musical emotions in humans. Neuroscience Letters, 328, 145–149.
Kreibig, S. D. (2010). Autonomic nervous system activity in emotion: A review. Biological psychology, 84, 394-421.
Krumhansl, C. (1997). An exploratory study of musical emotions and psychophysiology. Canadian Journal of Experimental Psychology, 51, 336–353.
Mun, E. Y., Windle, M., & Schainker, L. M. (2008). A model-based cluster analysis approach to adolescent problem behaviors and young adult outcomes. Development and
psychopathology, 20(1), 291-318.
Murray-Close, D. (2006). Gender Differences in the Association Between Autonomic Arousal and Aggressive Conduct. Ann Arbor: ProQuest.
Ortiz, J., & Raine, A. (2004). Heart rate level and antisocial behavior in children and adolescents: A meta-analysis. Journal of the American Academy of Child & Adolescent Psychiatry, 43(2), 154-162.
Piferi, R. L., Kline, K. A., Younger, J., & Lawler, K. A. (2000). An alternative approach for achieving cardiovascular baseline: viewing an aquatic video. International Journal of Psychophysiology, 37(2), 207–217. doi:10.1016/S0167-8760(00)00102-1
Raine, A. (1993). The Psychopathology of Crime: Criminal behavior as a clinical disorder. San Diego, CA: Academic Press
Raine, A. (1996). Autonomic Nervous System Factors Underlying Disinhibited, Antisocial, and Violent Behavior Biosocial Perspectives and Treatment Implicationsa. Annals of the New York Academy of Sciences, 794(1), 46-59.
Raine, A., Dodge, K., Loeber, R., Gatzke-kopp, L., Lynam, D., Reynolds, C., … Liu, J. (2006). The Reactive–Proactive Aggression Questionnaire: Differential Correlates of Reactive and Proactive Aggression in Adolescent Boys. Aggressive behavior, 32(2), 159–171. doi:10.1002/ab.20115
Raine, A., Venables, P. H., & Mednick, S. A. (1997). Low resting heart rate at age 3 years predisposes to aggression at age 11 years: evidence from the Mauritius Child Health Project. Journal of the American Academy of Child and Adolescent Psychiatry, 36(10), 1457–1464. doi:10.1097/00004583-199710000-00029
Raine, A., Venables, P. H., & Williams, M. (1990). Relationships between central and autonomic measures of arousal at age 15 years and criminality at age 24 years. Archives of General Psychiatry, 47(11), 1003-1007.
Raine, A., Venables, P. H., & Williams, M. (1995). High Autonomic Arousal and Electrodermal Orienting at Age 15 Years as Protective Factors Against Criminal Behavior at Age 29 Years. American Journal of Psychiatry, 152(11), 1595–1600. Rickard, N. (2004). Intense emotional responses to music: A test of the physiological arousal
hypotheses. Psychology of Music, 32, 371–388.
Salimpoor, V. N., Benovoy, M., Longo, G., Cooperstock, J. R., & Zatorre, R. J. (2009). The rewarding aspects of music listening are related to degree of emotional arousal. PLoS One, 4, e7487.
Scarpa, A., Haden, S. C., & Tanaka, A. (2010). Being hot-tempered: Autonomic, emotional, and behavioral distinctions between childhood reactive and proactive aggression. Biological Psychology, 84(3), 488–496. doi:10.1016/j.biopsycho.2009.11.006
Schechtman, Z. (2003). Cognitive and affective empathy in aggressive boys: Implications for counselling. International Journal for the Advancement of Counseling, 24, 211-222. Schiffer, R.B., Rao, S.M., & Fogel, B.S. (Eds.). (2003). Neuropsychiatry. Philadelphia:
Lippincott Willimas & Wilkins.
Summers, C. H., & Winberg, S. (2006). Interactions between the neural regulation of stress and aggression. Journal of Experimental Biology, 209, 4581-4589.
Tabachnick, B. G., & Fidell, L. S. (2013). Using Multivariate Statistcis, 6th ed. Boston: Pearson
Van der Zwaag, M. D., Westerink, J. H., & Van den Broek, E. L. (2011). Emotional and psychophysiological responses to tempo, mode, and percussiveness. Musicae Scientiae, 15(2), 250-269.
Van Domburgh, L., Vermeiren, R., Blokland, A. A. J., & Doreleijers, T. A. H. (2009).
Delinquent development in Dutch childhood arrestees: developmental trajectories, risk factors and co-morbidity with adverse outcomes during adolescence. Journal of
Abnormal Child Psychology, 37(1), 93–105. doi:10.1007/s10802-008-9260-6 Vitaro, F., Brendgen, M., Barker, E.D., 2006. Subtypes of aggressive behaviors: a
developmental perspective. International Journal of Behavioral Development 30, 12– 19.
Vries-Bouw, D., Popma, A., Vermeieren, R., Doreleijers, T.A., Van de Ven, P.M., & Jansen, L. (2011). The predictive value of low heart rate variability during stress for
reoffending in delinquent male adolescents. Psychophysiology, 48(11), 1597-1604. Wechsler, D. (1997). WAIS-III: Administration and scoring manual: Wechsler adult
Willemsen, G. H. M., De Geus, E. J. C., Klaver, C. H. A. M., Van Doornen, L. J. P., & Carroll, D. (1996). Ambulatory monitoring of the impedance cardiogram. Psychophysiology, 33(2), 184–193. doi:10.1111/j.1469-8986.1996.tb02122.x
Zeffirelli, F. (producer). (1979). The Champ [Dvd]. United States of America: Warner Bros. Zuckerman, M. (1979). Sensation seeking: Beyond the optimal level of arousal. Hillsdale, NJ:
Table 1
Characteristics of multiproblem young adolescents
M SD Minimum Maximum Questionnaires RPQ total 18.77 8.46 4.00 41.00 Proactive Aggression 6.59 4.81 0.00 19.00 Reactive Aggression 12.18 4.39 3.00 22.00 ANS variables HR rest (baseline) 64.42 8.57 48.30 87.12 HR movie 1 (Champ) 63.03 8.15 48.70 86.89 HR movie 2 (Mohammed) 64.17 7.76 49.80 84.34
HR total movie 2 (Mohammed) 64.68 8.01 51.71 89.15
SCL rest (baseline) 5.22 3.45 1.48 15.79 SCL movie 1 5.79 4.30 1.06 20.35 SCL movie 2 5.63 4.37 1.09 20.27 SCL total movie 2 5.76 4.37 1.14 20.99 Demographic variables Educational level 3.07 2.43 0.00 10.00 Age 21.82 2.50 18.00 26.00
Note. N = 44. RPQ = Reactive-Proactive Aggression Questionnaire; movie 1 = Champ, movie 2 = Mohammed (target), total movie 2 = total
Mohammed; HR = heart rate (beats per minute); SCL = skin conductance level (µS); Educational level: 0 = no education, 1 = primary school, 2 = lower vocational education basic (VMBO basis/beroeps), 3 = lower vocational education theoretical (VMBO kader/theoretisch), 4 = high school/pre-university education (havo/vwo), 5 = intermediate vocational education level 1 (MBO-1), 6 = intermediate vocational education level 2 (MBO-2), 7 = intermediate vocational education level 3 (MBO-3), 8 = intermediate vocational education level 4 (MBO-4), 9 = university (HBO/universiteit),10 = different.
Table 2
Descriptive Statistics and t-test results for HR and SC
Baseline Champ
Outcome M SD M SD 95% CI for Mean
Difference R t df HR 64.4 8.57 63.0 8.15 0.54, 2.24 .95* 3.29* 43 SC 5.2 3.45 5.8 4.39 -1.02, -0.12 .95* -2.54* 43 Baseline Mohammed
Outcome M SD M SD 95% CI for Mean
Difference R t df HR 64.4 8.57 64.7 8.57 -1.28, 0.75 .92* -0.52 43 SC 5.2 3.45 5.8 4.37 -1.02, -0.05 .94* -2.24* 43 * p < .05
