U
NIVERSITY OFA
MSTERDAMM
ASTERP
SYCHOLOGY,
B
RAIN ANDC
OGNITIONMasterthese (18 EC)
E
FFECT OF THE
‘C
OGNITIEVE
V
AARDIGHEDEN
T
RAINING
’
ON
THE
P
SYCHOPHYSIOLOGICAL
F
UNCTIONING OF
I
NMATES
.
J. van der Velden
Studentnummer:
0529915
Email adres:
sl_jemai@hotmail.com
Begeleiders vanuit het NSCR: Liza Cornet, Peter van der Laan en Cathy de Kogel (WODC)
Eerste beoordelaar:
Winni Hofman
Tweede beoordelaar:
Thelma Schilt
Onderzoeksinstelling:
NSCR
List of abbreviations:
ANS Autonome Nervous System
CBT Cognitive Behavioral therapy
CoVa Cognitieve Vaardigheden training (Cognitive Skill training)
HR Heart Rate
HRV Heart Rate Variability
LAT Low Arousal theory
NSCR Nederlands Studiecentrum Criminaliteit en Rechtshandhaving (Dutch Institute for
Criminoligy and Law)
PA Proactive
PEP Pre Ejection Period
PFC Prefrontal Cortex
PNS Parasympathetic Nervous System
RA Reactive
Risc Risico inschattings schalen
RPQ Reactive proactive Questionnaire
SCL Skin Conductance Level
SDAS Social Dysfunction and Aggression scale
TriPM Triarchic Psychopathic Measure
ABSTRACT
Background: There is a substantial subgroup of prisoners that does not respond to treatment
favorably, exhibiting high levels of antisocial and disruptive behavior. It is crucial that underlying mechanisms for these differential effects are identified to maybe improve treatment efficacy or selection procedures. Objectives: The present study investigated whether individual differences in psychophysiological factors contribute to explaining differences in responsiveness to cognitive behavioral therapy in this case the “cognitieve vaardigheden training”. Methods: The participants were divided between two clusters. Cluster one existed of inmates with a relatively low heart rate and cluster two of inmates with a relatively high heart rate. Several psychophysiological measures: heart rate, heart rate variability and pre ejection period were measured at rest and during the D2 task. These measures were assessed before and after completion of treatment. We also investigated whether cluster 1 and 2 differed on psychopathy, proactive aggression, reactive aggression and behavior according to their trainers and mentor. Results: Heart rate changed for both clusters, but this change was not significant. Heart rate variability increased for prisoners in cluster 1 and decreased for prisoners in cluster. No differences between the clusters were found on: psychopathy, proactive aggression and reactive aggression and behavior according to their trainers and mentor.
Conclusion: The CoVa influences the psychophysiological functioning of inmates. These
psychophysiological changes were not reflected in the behavior they exhibited, according to their trainers and mentor.
Key Words: heart rate, heart rate variability, pre ejection period, cognitive behavioral therapy,
INTRODUCTION ...4
Physiological reactions ... 5
Proactive aggression, reactive aggression and psychopathy ... 7
Prefrontal cortex activation ... 9
Cognitive behavioral therapy ... 10
Present study ... 10
METHOD ... 11
Participants... 12 Procedure ... 12 Measures ... 13 Data Analyses ... 16RESULTS ... Error! Bookmark not defined.
Physiological measures ... 18Treatment outcome measures ... 21
DISCUSSION ... 21
REFERENCES
...25
INTRODUCTION
Antisocial and aggressive behavior is a common problem in prison. Impulsive antisocial behavior or misconduct may be generated by impairments in the ability to accurately perceive emotional/social cues and to regulate responses to them. These regulatory abilities are modulated by the prefrontal cortex (PFC) and its communication with lower structures in the limbic system. The limbic system is responsible for emotional reactions to social information (Fishbein & Sheppard, 2006). Offending is linked to structural and functional deficits in the same areas (Vaske, Galyean, & Cullen, 2011).
Low autonomic arousal has shown to be a consistent biological correlate of antisocial behavior (Thayer & Brosschot, 2005; Ortiz & Raine, 2004). These findings are consistent with the ‘low arousal theory’ (LAT) and the ‘fearlessness theory’. The LAT states that individuals seek out stimulation, for example, by displaying antisocial behavior, to optimize their low arousal, which represents an unpleasant physiological state. The fearlessness theory holds that antisocial individuals do not fear the negative consequences of their actions and therefore are more likely to engage in antisocial behaviors. (de Vries-Bouw, Popma, Vermeiren, Doreleijers, van de Ven, & Jansen, 2011). From both perspectives it is expected that basal autonomic nervous system (ANS) activity and ANS responsivity to psychosocial stress would be attenuated in antisocial delinquents and may put them at risk for further antisocial behavior.
If detainees with behavioral problems want to successfully re-integrate into society, they have to change their behavior and improve their executive functions. Executive functions are cognitive functions that are necessary to perform complex behaviors (Mullin & Simpson, 2007). Executive function deficits that are implicated in antisocial behavior are thought to be responsible for poor social skills and decision-making ability, insensitivity to punishment, impulsivity and inability to regulate emotional responses.
In penal institutions behavioral interventions can be an important tool to improve (re-) integration of offenders into society and to reduce criminal recidivism (Lipsey & Cullen, 2007). Many different training programs are aimed at changing behavior, but the effects of these programs vary widely. Treatment effects are for a large part determined by the type of treatment, the way the intervention is executed and the population of offenders that is being treated (Lipsey & Cullen, 2007).
In the present study the “Cognitieve vaardigheids training” (CoVa) will be examined .The CoVa is a correctional cognitive behavioral therapy (CBT). Cognitive behavioral therapy is a successful treatment choice when working with offenders according to a large number of meta-analyses (Andrews, Zinger, Hoge, Bonta, Gendreau, & Cullen, 1990; Pearson, Lipton, Cleland, & Yee, 2002; Dowden & Andrews, 2000; Landenberger & Lipsey, 2005). Paquette, et al., (2003) suggest that a
psychotherapeutic approach, such as cognitive behavioral therapy, has the potential to modify the dysfunctional neural circuitry associated with anxiety disorders. Their findings indicate that changes made at the mind level, within a psychotherapeutic context, are able to functionally “rewire” the brain. Although the participants in the former research suffered from an anxiety disorder instead of an behavioral disorder. It may be possible that changes in brain functioning and psychophysiology underlie the relationship between changes in cognitive functioning and changes in criminal behavior. According to Vaske, Galyean, & Cullen, (2011) Cognitive behavioral therapy may be effective in reducing problem behavior, including crime, because the intervention affects specific areas of the brain. CBT changes in cognitionchanges in brain functioningchanges in behavior.
Although it’s demonstrated that CBT works, the recidivism rates are high. Aos and colleagues (2006) estimated only a 6.3% reduction in crime outcomes for cognitive behavioral therapy. There is a substantial subgroup that does not respond to treatment favorably, exhibiting high levels of antisocial and disruptive behavior within and outside of the correctional environment. It is crucial that underlying mechanisms for these differential effects are identified to maybe improve treatment efficacy.
Within populations who exhibit antisocial behavior different patterns of emotion regulation are found that seem to be associated with various forms of antisocial behavior. Low activity of the sympathetic nervous system at rest seems to be associated with proactive (see paragraph: Proactive aggression, reactive aggression and psychopathy) aggressive behavior and high sympathetic activity seems to be associated with reactive aggressive behavior (Stadler, Grasmann, Fegert, Holtmann, Poustka, & Schmeck, 2008; Hawes, Brennan, & Dadds, 2009). These individual differences in psychophysiological characteristics may affect treatment response. The objective of this research is to gain more insight into the psychophysiological functioning of aggressive detainees and the effect of the “CoVa” on the psychophysiological functioning of aggressive inmates. Understanding of the psychophysiological characteristics of inmates possibly contributes to a more targeted and effective treatment or better selection methods for treatment.
Physiological reactions
The autonomic nervous system, regulates a large number of functions that are unconsciously taking place. Examples of such unconsciously functions are breathing, digestion and widening and narrowing of the blood vessels. The autonomic nervous system is also the most prominent factor that determines cardiac functions such as heart rate (HR) and heart rate variability (HRV) (Levy, 1990).
The autonomic nervous system is generally conceived to have two major branches, the sympathetic nervous system, associated with energy mobilization, and the parasympathetic nervous system, associated with vegetative and restorative functions. Both branches of the autonomic nervous system are tonically active with sympathic activity associated with heart rate acceleration and parasympathetic activity associated with deceleration. Heart rate variability has been studied as an important marker of autonomic nervous system modulation and used as an indicator of the sympathovagal balance (the activity of both the sympathic nervous system and the parasympathetic nervous system, which have generally opposing actions on organ systems) (Berntson, Cacioppo, & Quigley, 1991). The parasympathetic (primarily vagal) influences are associated with a high heart rate variability, particularly the high frequency component (or respiratory sinus arrhythmia) of heart rate variability (De Bernardi Luft, Takase, & Darby, 2009). Respiratory sinus arrhythmia (RSA) is a cardiorespiratory phenomenon characterized by the rhythmic waxing and waning of cardiac vagal efferent effects upon the sinoatrial node that are in phase with inhalation and exhalation (Grossman & Taylor, 2005). Heart rate increases during inspiration due to a decrease of vagal outflow. During expiration vagal outflow restores and HR decreases.
“The sympathetic effects are on the time scale of seconds whereas the parasympathetic effects are on the time scale of milliseconds. Therefore, the parasympathetic influences are the only ones capable of producing rapid changes in the beat-to-beat timing of the heart” (Lane R. D., McRae, Reiman, Chen, Ahern, & Thayer, 2008).
The hart is under tonic inhibitory control by parasympathetic influences. Thus, during rest cardiac autonomic balance is regulated by way of parasympathetic dominance over sympathetic influences (Jose & Collison, 1970). Under conditions of uncertainty and threat, the prefrontal cortex becomes hypoactive. This hypoactive state is associated with disinhibition of sympathoexcitatory circuits. As a consequence heart rate will accelerate and heart rate variability will decrease.
The autonomic nervous system is regulated by the prefrontal cortex. This neural network permits the prefrontal cortex to inhibit subcortical structures associated with defensive behaviors and promotes flexible responsiveness to environmental changes (Thayer & Brosschot, 2005). Sympathetic and parasympathetic tone in a resting state may reflect individual differences in the capacity to respond adaptively to internal en external demands placed upon a person (Thayer, Hansen, Saus-Rose, & Johnse, 2009; Scarpa, Haden, & Tanaka, 2009). According to Lane R. D., McRae, Reiman, Chen, Ahern, & Thayer (2008) vagal tone has been hypothesized to play an important role in emotion regulation. To react to emotional stimuli a person has to select an optimal response and to inhibit less functional responses from a brad behavioral repertoire. Relatedly, high-frequency HRV may be considered a resource that can be drawn upon in support of emotion regulation. Individuals with
greater cardiac vagal control (higher HRV) are better prepared to respond efficiently (Leon, Hernandez, Roderiguez & Vila, 2009). Thayer and Bronsschot (2005) for instance, showed that individuals with low heart rate variability reacted to neutral, harmless stimuli as if they were aversive and threatening. In contrast, individuals with high heart rate variability were better able to match their response to situational demands and therefore respond more appropriately to the energy requirements of the situation. Efficient vagal regulation appears to facilitate coping with stress and appears to be related to positive engagement strategies (Leon, Hernandez, Roderiguez & Vila, 2009). A study by Johnsen et al. (2003) brings support to the idea that high levels of vagally mediated heart rate variability are associated with efficient attentional regulation and greater ability to inhibit prepotent but inappropriate responses.
Offenders show abnormalities in physiological arousal at rest (van Goozen & Fairchild, 2008; Raine, 1996). Low autonomic arousal has been most repeatedly related to antisocial and violent behavior in both child and adult samples. In general the resting heart rate and skin conductance levels are lower in antisocial individuals (Kempes, Matthys, de Vries, & van Engeland, 2005; Raine, 2002). Offenders and no offenders can also be distinguished on the basis of their cortisol levels (stress reactivity). Offenders show lower amounts of salvatory cortisol, which also indicates low autonomic arousal (Cima, Smeets, & Jelicic, 2008).
It is clear that the psychophysiology of antisocial individuals is very different compared to the psychophysiology of ‘normal’ individuals. Knowledge and understanding of the psychophysiological functioning of antisocial individuals is important and can be used for several purposes. It may even be possible to distinguish different subgroups of antisocial prisoners on the basis of their individual physiologic characteristics.
Proactive aggression, reactive aggression and psychopathy
According to information processing models, behavior is a function of several steps of processing, including encoding of social cues, interpretation of social cues, clarification of goals, response access or construction, response decision, and behavioral enactment (Lemerise & Arsenio, 2000; Dodge & Crick, 1990). An adequate response to a social cue represents a selection of an optimal response and the inhibition of less functional ones from a broad behavioral repertoire. Biased or deficient processing is hypothesized to lead to aggressive and antisocial behavior. At least two general social information processing patterns have found to be characteristic of aggressive behavior. The first pattern involves processing at the ‘interpretation step’ of the previous described model. Antisocial persons react to neutral, harmless stimuli as if they were aversive and threatening, and also have the tendency to react similarly to positive stimuli (Thayer & Brosschot, 2005). This type of aggression is
reactive aggression. The second pattern involves the ‘response decision step’ of processing. During this step, individuals evaluate possible behavioral responses to a particular social situation. Thus, for individuals who exhibit this second social information-processing pattern, aggression may function as a viable means for obtaining desired goals (Crick & Dodge, 1996). This type of aggression is proactive aggression.
Reactive aggression (RA), also referred to as hostile or impulsive aggression, is generally defined as aggression that occurs as an angry response to provocation or threat. Aggressive behavior of this type is often characterized by high emotional and physiological arousal, such as during an argument or fight. As stated above reactive aggression is generated by the impairment in the ability to accurately interpret social and emotional cues and then to regulate the response to them. Reactive aggressive behavior is primarily committed to intimidate and dominate others and it’s not planned behavior. In contrast, proactive aggression (PA), also referred to as instrumental aggression, is generally defined as aggression that is unprovoked and typically involves planning and forethought. This type of aggression is used for some sort of instrumental gain, such as to obtain goods or services, to obtain dominance over others, or to enhance one’s social status. Proactive aggressive offenders are provocative and relatively unemotional (Kempes, Matthys, de Vries, & van Engeland, 2005; Fishbein & Sheppard, 2006). Both in adults and children, different patterns of emotion regulation have been found that seem to be associated with proactive/reactive aggressive behavior. Autonomic underarousal may underlie proactive aggressive behavior, which reflects low levels of heart rate and skin conductance. In contrast to heart rate variability, which is high. Physiologically arousal is associated with impulsive/reactive aggressive behavior. Such physiologically hyper-arousal is thought to reflect an automatic stress response, in which the sympathoexcitatory circuits are disinhibited. Heart rate variability is low and heat rate and skin conductance levels are high in rest (Stadler, Grasmann, Fegert, Holtmann, Poustka, & Schmeck, 2008; Hawes, Brennan, & Dadds, 2009).
Most research concerning proactive and reactive aggression has been done in children samples. There appears to be a group of highly aggressive children who show both types of aggressive behavior and another group of children who are less aggressive overall and who show only reactive aggression (Frick, Cornell, Barry, Bodin, & Dane, 2003). Children who are high on proactive aggression, show often reduced levels of emotional reactivity in contrast to children who are high on reactive aggression showing high rates of angry reactivity and low frustration tolerance (Munoz & Frick, 2012). It appears that children who are high on both reactive and proactive aggression show a number of characteristics that are similar to adults with psychopathy such as their insensitivity to potential punishment for their aggressive behavior and deficits in their emotional reactivity. Several
studies have shown that criminals identified as perpetrating predominantly proactive violent offenses have higher scores on the Psychopathy Checklist than those with a history of reactice violent offenses (Raine, et al., 2006). In A sample of juvenile offenders incarcerated in adult prison, offenders who showed more severe, repeated, proactive, and sadistic violence against their victims scored higher on a self-report measure of psychopathic traits (Munoz & Frick, 2012). These findings suggest that proactive aggression is an adolescent indicator of psychopathic personality.
According to Hart and Hare (1996) psychopathy is defined by antisocial behavior in addition to emotional impairment such as lack of empathy. Recently Bolt, Hare, Vitale, & newman (2004) described four facets of psychopathy. The first two facets give a measure of the clinical aspect of psychopathy and include features of narcissism which is characterized by egocentricity, grandiosity, arrogance, envy and lack of empathy. The last two facets are related to behavioral aspects such as impulsive and violent behavior.
Numerous studies have shown that crime and violence are closely linked to antisocial behavior and psychopathy. Differentiation between antisocial behavior and psychopathy might have implications with regard to development of preventive and treatment programs as well as risk assessment. And research on the physiological mechanisms may help us to better understand their complexity.
Prefrontal cortex activation
Physiological, cognitive and affective regulation are all linked with prefrontal cortex (PFC) activity. As stated above, PFC activity is associated with vagal mediated cardiac functions and indexed by heart rate variability and heart rate (Thayer & Brosschot, 2005). Thus, heart rate variability may serve to index the functional capacity of a set of brain structures (the PFC) that also supports the effective and efficient performance on cognitive executive-function tasks. Individual differences in heart rate variability are related to performance on tasks associated with executive function and prefrontal cortical activity. Persons with greater vagal mediated HRV perform better on executive-function tasks in a wide range of situations (Thayer, Hansen, Saus-Rose, & Johnse, 2009).
Neuroimaging studies in humans have reported increased activity in the prefrontal cortex during tasks involving executive functioning. Emotional arousal is associated with a decrease in HRV and concomitant decreases in brain activation in the prefrontal cortex. During emotional stress the PFC is taken offline to let automatic, proponent processes regulate behavior (Lane, McRae, Reitman, Chen, Ahern, & Thayer, 2009; ).
As stated above, activity in the PFC is associated with vagal function (Giannakos, Van Der Vein, & Jennings, 2004). When the activation of the PFC decreases, for example during emotional arousal,
heart rate variability will also decrease. When brain activation of the prefrontal cortex increases, heart rate variability will also increase.
Cognitive behavioral therapy
Cognitive behavioral interventions focus on the relationship between cognition and behavior. According to cognitive behavioral therapy (CBT) programs, high-risk situations lead to antisocial thoughts and feelings. These thoughts and feelings increase the likelihood of antisocial behavior. CBT attempts to reduce antisocial behavior by reducing the antisocial cognitive and -emotional response to (high-risk) stimuli. Thus, changes in antisocial/criminal behavior can be affected by altering the thinking patterns of offender’s.
In most cognitive behavioral therapy programs, offenders improve their social skills and executive functions, such as means-ends problem solving, critical reasoning, moral reasoning, cognitive style, self-control, impulse management and self-efficacy. There are three key skill sets that CBT attempts to improve, namely social skills, coping skills, and problem-solving skills. These three skill sets are related to a specific area of the brain, namely the prefrontal cortex. Offending is linked to structural and functional deficits in these same brain areas and in the limbic structures (emotion-regulation). For more detailed information of the specific areas see Vaske, Galyean, & Cullen, (2011).
All detainees in this study participate in the “Cognitieve Vaardigheden training” (CoVa). The CoVa is a cognitive behavioral intervention for prisoners with cognitive deficits. The CoVa is the Dutch translation and adaptation of the widely used and studied for effectiveness “Enhanced Thinking Skills” training (Sadler, 2010). The aim of this intervention is to reduce the risk of recidivism. The training focuses on the development of (thinking) skills that enables participants both inside and outside the prison to think and act on a more pro-social way. The four major domains of the CoVa are: impulsivity, problem solving, critical/moral reasoning and perspective taking.
Present study
The percentage of prisoners who are released from a penal institution but come back into contact with justice is high, even after they have completed a correctional intervention. Although the CoVa is a successful treatment choice when working with offenders, why are the recidivism rates high? There appears to be a large group of prisoners who have completed the CoVa, but who are not taking advantage of the training. A possible reason for the low responsivity to the CoVa is that the selection criteria for the CoVa do not make a distinction between psychopaths and non-psychopaths. Psychopaths generally exhibit more cold-blooded aggressive behavior (proactive aggression)
compared to non-psychopaths who exhibit generally more impulsive aggressive behavior. The CoVa is mainly aimed at people who exhibit impulsive aggressive behavior (reactive aggression).
According to Vaske, 2011 we need to move towards a biosocial theory of offender rehabilitation. There have been many studies concerning the executive functioning of prisoners in relation to treatment, but there is little attention for psychophysiological factors in relation to treatment. Up to now little is known at biological level. Most studies concerning physiological factors and treatment, investigate the predictive value of psychophysiological factors on treatment response (Mullin & Simpson, 2007; Fishbein & Sheppard, 2006). According to van der Wiel, Jeglic, & Donovick, (2004) children with a relatively small physiological reaction to stress benefit less from treatment. Also a low heart rate in rest predicts poorer treatment response (Stadler, Grasmann, Fegert, Holtmann, Poustka, & Schmeck, 2008).
Major part of all research done is with children who exhibit antisocial and aggressive behavior, but it is not clear to what extent the results can be generalized for adults (Stadler, Grasmann, Fegert, Holtmann, Poustka, & Schmeck, 2008; van der Wiel, Jeglic, & Donovick, 2004; van Goozen & Fairchild, 2008).
The aim of this present paper is to gain more insights into the underlying mechanisms of anti-social behavior of adult male prisoners and the effectiveness of interventions through a biological approach. This study examined the effect of the CoVa on HR, HRV and PEP. We expected changes in cardiac vagal tone because the intervention affects brain regions which are linked to ANS regulation. In this study a distinction was made between two subgroups of inmates. The “low ANS activity” types (i.e., low HR and high HRV) and the “high ANS activity” types (i.e., high HR and low HRV). Based on the above review, it was hypothesized that psychophysiological functioning of inmates who were in the high ANS activity group changed in contrast to the psychophysiological functioning of inmates in the low ANS activity group, because of the underlying process that initiates proactive behavior. Finally, it was hypothesized that changes in psychophysiological functioning go along with changes in behavior.
METHOD
This study is part of a lager project that examines the predictive value of neurobiological and neurocognitive factors in relation to treatment response to the CoVa. The procedure that was used for the collection of data was already determined (for an overview see appendix 1). For the present study only a selection of test was used.
Participants
The full sample included twenty-nine male inmates (Mage=27.62, SD=8.45; range 19-58). All
participants were selected for the CoVa by their probation officer, based on the chance of recidivism (medium to high) as measured by the “Recidive Inschattingsschalen”. Inmates were included if: the time that the inmate still has to be in prison after finishing the CoVa did not exceeded twelve months, they were over 18 and their IQ score was >70. Subjects were excluded if their Dutch language understanding was not sufficient to understand and answer the questionnaires and when at the time of invitation it was undesirable that the test protocol was taken due to a personal situation. This research is conducted in accordance with the principles of the Declaration of Helsinki (October 2008; www.wma.net) and in accordance with the Law Medical- Scientific Research with humans. The ethics committee METc of the VUmc has approved this study. Prior to the assessment all participants were briefed about the procedures, and provided with informed consent.
Procedure
The pre-measurement took place two weeks before the inmate had started with the CoVa. The inmate had already answered the treatment motivation questionnaire, RPQ and the TriPM (see psychopathy data collection) before he started with the pre-measurement. At the beginning of the pre-measurement the questionnaires were checked for the absence of answers. If a response was missing this was yet answered, along with the prisoner. After that a few general questions were asked regarding personal health and amount of drinks with caffeine the inmate had consumed within the past 24 hours. Next the electrodes were placed on the participant, the VU-AMS was connected to a laptop (Dell Precision M4600) and the measurement of the heart was checked. After that baseline 1 and the ECG/ICG measure started. The participant sat quietly in a chair, while he listened to relaxing music and watched serene pictures on the computer (this took 5 minutes). The participant first started with the paper tasks, seven in total. During the tasks markers were sent to the VU-AMS via E-Prime 2.0 to pinpoint the start and finish on the ECG-measurement of every task. After finishing the paper tasks, baseline 2 started (also 5 minutes). After that the participant started with the computer tasks, three in total. In the next step the Mini International Neuropsychiatric Interview (MINI) was conducted and after that baseline 3 started. The participant was now finished with the pre-measurement and all the electrodes were removed. (See figure 1: timeline pre-pre-measurement)
The post-measurement took place right after the inmate had finished the CoVa. First the inmate completed the self-evaluation questionnaire. The rest of the assessment was identical to the pre-measurement except for the M.I.N.I. and the NLV (first of the paper tasks).
Figure 1. Timeline pre-measurement
The M.I.N.I. and the NLV were only conducted during the pre-measurement. The pre-measurement took approximately 3 h and the post-measurement 2 h.
Trainers administered the questionnaire “gedrag tijdens de CoVa” over the telephone, two weeks after the start and two weeks before the end of the CoVa. Mentors administered the questionnaire “behandelgedrag” over the telephone also two weeks before the start and two weeks before the end of the CoVa. For completing the pre- and post-measurement the inmate received twenty-five euro.
Measures
Psychophysiological measures heart rate, heart rate variability and pre-ejection period were measured using The Vrije Universiteit Ambulatory Monitoring System 36 (VU-AMS). Six disposable Ag/AgCl electrodes were attached four on the chest and two on the back of the participant. Electrode resistance is kept below 10Kohm by cleaning with alcohol and rubbing (See appendix 1.b). Heart rate was measured in beats per minute (bpm), using the R-peak of the QRS-complex as a marker (See figure 2). The time between two R-peaks is measured over a given period of time, resulting in a RR-interval (or NN-RR-interval, referring to Normal beats). Variability of this inter beat RR-interval (IBI) is known as the Heart rate variability. Heart rate variability is assessed in two ways namely, as the root mean squared of the standard deviation (RMSSD) and as the respiratory sinus arrhythmia (RSA). The pre-ejection period (PEP) is measured as the time between the ECG Q wave (onset of ventricular depolarization) and the impedance cardio graphic B wave (onset of left ventricular ejection) (See figure 2). Mean scores for HR, RMSSD, RSA and PEP were calculated for baseline 2 and for the D2 task (pre- and post measurement). The software that was used is AmsComfigure.
Figure 2. QRS-complex and PEP
HR reflects both sympathetic and parasympathetic control of the heart. RMSSD provides a sample of the short-term variance of the inter beat intervals and reflects vagal control of the heart as an index of parasympathetic tone. Larger mean standard deviations indicate increased HRV. This measure of HRV has been shown to strongly relate to cardiac vagal tone as assessed by pharmacological blockade (Hayano et al., 1991), thus providing evidence for its use as an appropriate index of vagal tone. RSA was also used to index parasympathetic functioning and sympathetic-linked cardiac activity was indexed by PEP (two measures are used to index vagal tone (RSA and RMSSD) because, the initiator of the bigger project (Liza Cornet) had indicated that both measures should be included in this research).
Neurocognitive measures The D2-task (Brickenkamp, 1998) is a concentration- and attention task. In this research it is used to induce a stress reaction. This task was the best option because it focuses on speed, accuracy, concentration and tempo variation. This task is also chosen because the individual scores are easy to compare. The D2-task has a set time, thus no extra calculations had to be performed in order to compare the physiologic values recorded during the task. The task consists of 14 lines of ‘p’s and ‘d’s with a maximum of 4 stripes surrounding it, either above, underneath or both. The subject was given a maximum of 20 seconds per line to tick off the ‘d’s with two stripes surrounding it, while ignoring the other symbols. In total this task takes 4 minutes and 40 second. Two different kinds of errors could be made, the F1 error and the F2 error. F1 indicates that a correct symbol (a ‘d’ with two stripes surrounding it) is not ticked off, F2 indicates that an incorrect symbol is ticked off. The internal consistency of the total number of processed items ranged in different studies from α=0.84 till α=0.98 (Brickenkamp 1998).
Psychopathy data collection The Reactive proactive aggression questionnaire (RPQ) is a self-reported scale developed by Raine, et al. (2006) and is used to assess reactive and proactive aggression. As described by Lai-Chu Fung, Raine, & Gao (2009). “The questionnaire consists of 23 behavioral items rated on a 3-point scale, with 0= never, 1 = sometimes, and 2 = often. A total of 11 items assess reactive aggression (e.g., “Reacted angrily when provoked by others”), and 12 items assess proactive aggression (e.g., “Hurt others to win a game”). Scores are summated to form measures of reactive of proactive aggression together with an overall score of total aggression (Raine, et al., 2006). Internal consistency in adolescents has previously been reported as α=0.86 for proactive aggression, α=0.84 for reactive aggression, and α=0.90 for total aggression (Raine, et al., 2006)”.
The Triarchic Psychopathy Measure (TriPM) (Patrick, 2010) is a 58-item self-report measure used to discriminate between three facets of psychopathy, namely Boldness (19 items), Meanness (19 items) and Disinhibition (20 items). It has three scales that capture distinct constructs of the three facets. Using a 4-point, Likert-type scale (“true”, “somewhat true”, “somewhat untrue”, “untrue”), participant’s rate the degree to which the items, consisting of personal characteristics (e.g., “I’m more often than not optimistic”, “I sympathize with people’s problems”), apply to them.
Treatment outcome measures The Social and Dysfunction and Aggression Scale (SDAS) is used to assess whether a detainee has benefited from the CoVa and is telephonically conducted by the mentors of the detainees, before and after the training. The mentor has rated the detainees on 11 items such as “irritability” by giving a score between 0 to 4 (0 is “not present” and 4 for “very serious”). A high score indicates socially dysfunctional and aggressive behavior. The measure for treatment response is operationalized by the score attested in the first interview minus the scores attested in the second interview. A positive number indicates an improvement in socially dysfunctional and aggressive behavior (Wisted, 1990). This is a frequently used reliable and valid questionnaire.
The questionnaire Gedrag tijdens de CoVa (behavior during the CoVa) is telephonically conducted by the CoVa trainers. The measure for treatment response is operationalized by the score attested in the second interview minus the scores attested in the first interview. For answer a, b, c and d is respectively, the score 1, 2, 3 and 4 assigned. A positive score indicates an improvement. The higher the subject’s score on the questionnaire, the better was his behavior. The items on the questionnaire represent a combination of knowledge, participation and competence. This questionnaire is edited and translated from the English version Treatment Gain: Short Scale, which has been used in a large-scale survey by (Fishbein & Sheppard, 2006). Treatment motivation is measured before the start of the CoVa by the “treatment motivation questionnaire”, which was completed by the prisoners
themselves. The questionnaire consists of 12 items, which can be answered with “false”, “don’t know” or “true”. An example of a question is: ”I need help”.
Treatment response is measured by the “self-evaluation questionnaire”. This was completed by the prisoners themselves. An example of a question is: “Do you think the CoVa was useful”. This questionnaire has been specially developed for this study (four items, α=.90).
Data Analyses
All data is processed with SPSS (Statistical Package for the social Sciences), version 19.0. In all statistical analyses an alpha of 0.05 is used as level of significance.
First a number of tests were conducted to describe the sample population and distribution of the variables. The Cronbach’s alpha coefficient for internal consistency is calculated for the self-evaluation questionnaire.
Missing variables for HR, RSA, RMSSD and PEP were analyzed with analyze patterns. This was done by SPSS to see if the missing data was random or if there was a pattern in the missing data. This was necessary for choosing a suitable way of imputing the missing data. Next the parameters for the imputation of the missing data were set in the random number generator (under the heading transform). After that the missing data were imputed with multiple imputation. Clusters were made with K-Means Cluster Analysis based on HR, RSA, RMSSD and PEP during baseline 2 and during the D2 task of the pre-measurement. Group differences were examined by using multivariate analysis of variance (ANOVA) with clusters as between and BMI, amount of exercise per week, amount of cigarettes per day and age, scores on the TriPM, scores on the RPQ, scores on the treatment motivation questionnaire and scores on the self-evaluation questionnaire as dependent variables. A repeated measures ANOVA with measurement as within, clusters as between-subjects and HR during baseline as dependent variable was used to determine if a greater physiologic change had occurred after completion of the CoVa for cluster two compared to cluster one. The repeated measures ANOVA was performed for a total of six times, only the dependent variables were different. For the second repeated measures ANOVA: measurement as within, cluster as between-subjects and HR during D2 as dependent variable. For the third: measurement as within, cluster as between-subjects and RMSSD and RSA during baseline as dependent variable. For the fourth: measurement as within, cluster as between-subjects and RMSSD and RSA during D2 as dependent variable. For the fifth: measurement as within, cluster as between-subjects and PEP during baseline as dependent variable. For the last repeated measures ANOVA: measurement as within, cluster as between-subjects and PEP during D2 as dependent variable.
To compare increases in HR, RSA, RMSSD and PEP during the D2 task, the formula adopted form Davies and Maliphant, 1971b, is used as follows:
Mχ-Bχ Bχ =baseline χ Δ χ = ____________ X 100. Mχ=maximum χ Bχ
This formula takes into account baseline differences (Wilder, 1957), thus enabling comparisons to be made between individuals. A paired sample t-test is performed to compare HR during baseline with HR during the D2 task based on the pre-measurement to determine if the D2 task deservedly is used to induce stress. To examine if the physiological stress response of the prisoners had changed as a result of the CoVa and whether this change is different for the clusters, a repeated measures ANOVA with clusters as between-subjects, measurement as within and Δ HR as dependent variable is used. The repeated measures ANOVA is repeated twice with the same between and within variables, only the dependent variables were different. For the second repeated measures ANOVA: measurement as within, cluster as between-subjects and Δ RSA and Δ RMSSD as dependent variables. For the third: measurement as within, cluster as between-subjects and Δ PEP during baseline as dependent variable. A Pearson correlation test was performed to examine possible associations between the physiological variables and the treatment outcome measures.
RESULTS
Due to equipment malfunction psychophysiological data of several inmates was lost. The missing data for the baseline (pre-measurement): HR, RSA and RMSSD of one inmate and PEP of three inmates; the baseline (post-measurement): HR, RSA and RMSSD of four inmates and PEP of six; D2 (pre-measurement): PEP of four inmates; D2 (post-measurement): PEP of one inmate. Subjects with missing data were not included in the statistical analyses, unless a statistical analysis was conducted in which the missing data was not important. In addition statistical analyses were performed on a dataset where the missing’s were imputed. This dataset is only used to determine whether a trend towards significance could be observed with a higher number of subjects.
Clusters were made with K-Means Cluster Analysis. HR, HR during D2task, RMSSD, RMSSD during D2task, RSA and RSA during the D2task of the pre-measurement are used as variables for clustering. Table 1 shows means and standard deviations of HR, RMSSD and RSA based on the original data. Based on the imputed data Cluster 1 and 2 also differ significantly for HR. Multivariate analysis of variance performed on the data revealed no significant between group differences for the clusters, for amount of cigarettes smoked per day, hours of exercise per week, age and BMI. Also no significant between group differences were found for “behavior during training” and “behavior on
Table 1
Average HR, RMSSD and RSA of the Clusters Based on the Pre-measurement.
Physiological Mean Standard Deviation Sig.
variables Cluster 1 a Cluster 2b Cluster 1a Cluster 2b
Average HR (bpm) 64.37 72.65 8.63 8.85 .177
Average HR during D2 task 70.52 78.49 6.59 10.16 .145
RMSSD (msec) 128.11 40.27 49.19 18.75 .000
RMSSD during D2 task 125.95 33.31 45.14 19.32 .000
RSA (msec) 182.08 63.59 86.43 28.95 .000
RSA during D2 task 169.16 53.05 51.83 22.64 .000
a N=4 b N=24
the ward (SDAS)”. Multivariate ANOVA did reveal a trend towards significance between the clusters for “treatment motivation”, F(1, 22)=4.00, p=.058, Mcluster1=18, Mcluster2=22.95. Cluster two (high arousal group) was more motivated prior to the training than cluster one (low arousal group). For scores on the RPQ and the TriPM also no significant between group differences for the clusters were found. The clusters scored similarly on the TriPM and the RPQ, which means there was no difference between the clusters in psychopathy and aggressiveness.
Physiological measures
A paired-sample t-test was conducted to compare baseline HR with HR during the D2 task. A significant difference was found between baseline HR (M=70.04) and HR during D2 (M=77.35), t(27)=-9.15, p<.001. These results show that the D2 task induces stress (elevated heart rate). A repeated measures ANOVA revealed an marginally significant (p<0.10) interaction effect between training and clusters for HR during baseline (F(1,22)= 3.18, p=.089) and HR during D2 (F(1,26)=2.98, p=.096). Although marginally significant, the CoVa had a different effect on HR and HR during D2 for cluster one compared to cluster two. HR of cluster one decreased (Mpre=64.37; Mpost=57.49) and HR of cluster two increased (Mpre=72.65; Mpost=74.30) after training. HR during D2 also decreased for cluster one (Mpre=70.52; Mpost=62.04) and increased for cluster two (Mpre=78.49; Mpost=79.73). Figure 3A and B illustrate the direction to which HR and HR during the D2 task changes for both clusters. A repeated measures ANOVA showed a significant main effect of training for RSA (F(1,22)=9.33, p=.006) and RMSSD (F(1,22)=10.99, p=.003). As a result of the CoVa RSAduring baseline increased for cluster one (Mpre=182.08; Mpost=240.10) and decreased for cluster 2 (Mpre =61.86; Mpost=54.96). RMSSDbaseline also increased for cluster one (Mpre=128.11; Mpost=169.39) and decreased for cluster 2
A B
Figure 3
Direction of Change between pre- and post-measurement for both clusters A) Estimated Marginal Means of HR during baseline. SD cluster 1 measurement 1= 8.6. SD cluster 2 measurement 1= 8.5. SD cluster 1 measurement 2=6.6. SD cluster 2 measurement 2= 11.7. B) Estimated Marginal Means of HR during the D2 task. SD cluster 1 measurement 1= 6.6. SD cluster 2 measurement 1= 10.2. SD cluster 1 measurement 2=5.8. SD cluster 2 measurement 2= 12.5.
A B
Figure 4
Direction of Change between pre- and post-measurement for both clusters A) Estimated Marginal Means of RSA during baseline. SD RSA cluster 1 measurement 1= 86.4. SD RSA cluster 2 measurement 1= 26.5. SD RSA cluster 1 measurement 2=124.2. SD RSA cluster 2 measurement 2= 23.6. B) Estimated Marginal Means of RMSSD. SD RMSSD cluster 1 measurement 1= 49.2. SD RMSSD cluster 2 measurement 1= 18.2. SD RMSSD cluster 1 measurement 2=49.6. SD RMSSD cluster 2 measurement 2= 13.5.
70,53 62,04 78,49 79,73 50 55 60 65 70 75 80 85 90 95 100 1 2
HR
D2(b
pm
)
Cluster 1 Cluster 2 64,37 57,49 72,65 74,3 50 55 60 65 70 75 80 85 90 95 100 1 2HR
ba sel in e(b
pm
)
Measurement Cluster 1 Cluster 2 70,53 62,04 78,49 79,73 50 55 60 65 70 75 80 85 90 95 100 1 2HR
D2(b
pm
)
Measurement Cluster 1 Cluster 2 182,08 240,1 61,86 54,96 0 50 100 150 200 250 1 2RSA
ba se lin e(m
se
c)
Measurement Cluster 1 Cluster 2 128,11 169,39 39,38 36,27 0 50 100 150 200 250 1 2RM
SSD
ba se lin e(m
se
c)
Measurement Cluster 1 Cluster 2A B
Figure 5
Direction of Change between pre- and post-measurement for both clusters A) Estimated Marginal Means of RSA during the D2 task. SD RSAD2 cluster 1 measurement 1= 51.8. SD RSAD2 cluster 2 measurement 1= 22.6. SD RSAD2 cluster 1 measurement 2=28.2. SD RSAD2 cluster 2
measurement 2= 19.1. B) Estimated Marginal Means of RMSSD during the D2 task. SD RMSSDD2 cluster 1 measurement 1= 45.1. SD
RMSSDD2 cluster 2 measurement 1= 19.3. SD RMSSDD2 cluster 1 measurement 2=74.3. SD RMSSDD2 cluster 2 measurement 2= 11.9.
More importantly, the interaction effect between training and cluster was significant for both RSA (F(1,22)=15.04, p=.001) and RMSSD (F(1,22)=14.87, p=.001). Figure 4A and C illustrate the direction to which RSAbaseline and RMSSDbaseline changes for both clusters.
A significant main effect for training was found for RSA and RMSSD during the D2 task, respectively F(1,27)=12.0, p=.002 and F(1,27)=14.3, p<.001. The interaction effect between training and cluster was also significant for RSA during the D2 task (F(1,26)=18.6, p<.001) and for RMSSD during the D2 task (F(1,26)=19.1, p<.001). These results show that the CoVa had the same effect on RSA and RMSSD during the D2 task as it had on RSA and RMSSD during baseline. RSAD2 increased for cluster one
(Mpre=169.16; Mpost =239.54) and decreased for cluster 2 (Mpre=53.05; Mpost=45.48). RMSSDD2 also
increased for cluster one (Mpre=125.95; Mpost=195.15) and decreased for cluster 2 (Mpre=33.31; Mpost=28.33). Figure 5A and B illustrate the direction to which RSA and RMSSD changes during the D2 task for both clusters. For PEP neither the difference between clusters nor the difference between pre- and post-measurement even approached significance.
An independent sample t-test was conducted to examine whether cluster one and two differ in ΔHR, ΔRMSSD, ΔRSA and ΔPEP based on the pre-measurement. No significant differences were found between the clusters. These results show that the physiological reaction on the D2 task was similar
for both clusters. Three repeated measures ANOVA’s were performed to determine if the CoVa had any influence on the physiologic stress reaction of the inmates on the D2 task and if so, if this effect
169,16 239,54 53,05 45,48 0 50 100 150 200 250 1 2
RSA
D2(m
se
c)
Measurement Cluster 1 Cluster 2 125,95 195,15 33,31 28,33 0 50 100 150 200 250 1 2RM
SSD
D2(m
se
c)
Measurement Cluster 1 Cluster 2these result the CoVa had no effect on ΔHR, ΔRMSSD, ΔRSA and ΔPEP. Thus the CoVa had no influence on the physiologic stress reaction.
Treatment outcome measures
To examine a possible relationship between the treatment outcome measures (behavior on the ward and during the training, treatment motivation and self-evaluation). Pearson correlation tests were performed. Four significant positive correlations were found. Only the association between treatment motivation and treatment self-evaluation (r=.709, p<.001) was strong, thus increases in treatment motivation scores correlated with increases in self-evaluation scores. In addition, moderate to weak correlations were found between “behavior during the training (pre-measurement)” and “behavior during the training (post-(pre-measurement)” (r=.666, p<.001) and between “SDAS pre-measurement” and “SDAS2 post-measurement” (SDAS= Social and Dysfunctional and Aggression Scale) (r=.0470, p=.012). These results suggest that behavior during the pre-measurement (assessed by trainer and mentor) positively correlated with behavior during the post-measurement. Also a weak correlation was found between “treatment self-evaluation” and “behavior during training (pre-measurement) (r=.431, p=.028). This means that higher self-evaluation rates correlated with increased good behavior during training. The results also show a trend towards significance between “behavior during treatment” (pre-measurement)” and “treatment motivation” (r=.394, p=.051) and between “behavior during treatment (post-measurement)” and “self-evaluation”. A repeated measures ANOVA performed on the data revealed a significant main effect of the CoVa on ratings of the trainer concerning the inmates behavior during the training, F(1,25)=26.8, p<.001. No significant interaction effect was found. According to the trainers behavior of the inmates had improved equally for both clusters (cluster one: Mpre=15.25; Mpost=21.50, cluster two: Mpre=18.21; Mpost=21.83). This effect was not found for scores on the “SDAS”. According to the mentors behavior of the inmates did not improve, thus the CoVa had no effect on behavior that was being exhibited on the ward.
DISCUSSION
The current study examined the effect of the CoVa on heart rate, RMSSD, RSA and pre-ejection period of adult male convicts during rest and during a stress task. The aim of this present paper was to gain more insights into the underlying mechanisms of anti-social behavior of adult male convicts and the effectiveness of the CoVa through a biological approach. We expected changes in cardiac vagal tone because according to prior research cognitive behavioral therapy, in this case the CoVa, affects brain regions that are linked to Autonomic Nervous System regulation. The inmates were
assigned to cluster 1 (i.e., low HR, high RMSSD and RSA) or cluster 2 (i.e., high HR, low RMSSD and RSA), based on their RMSSD, RSA and HR in rest. Autonomic under arousal may underlie proactive aggressive behavior, which is reflected by cluster 1. Hyper-arousal is associated with reactive aggressive behavior, which is reflected by cluster 2 (Hawes, Brennan & Dadds, 2009). It was hypothesized that the psychophysiological functioning of inmates who were assigned to cluster 2 changed in contrast to the psychophysiological functioning of inmates assigned to cluster 1, because of the underlying process that initiates reactive aggressive behavior. Therefore we also expected a difference between the clusters in psychopathy and aggressiveness. Finally, it was hypothesized that changes in psychophysiological functioning go along with changes in behavior.
In summary, it appears that as a result of the CoVa the psychophysiological functioning of inmates had changed, although the results are opposite to the expectations. The HR of the subjects in cluster 1 decreased and the HR of the subjects in cluster 2 increased. RMSSD and RSA of the subjects in cluster 1 increased and decreased for subjects in cluster 2. If the clusters were to be translated into behavioral profiles this would mean that the psychopaths (cluster 1) were becoming more psychopathic after intervention and the reactive aggressive persons (cluster 2) were becoming more reactive aggressive. The CoVa is mainly aimed at people who exhibit reactive aggressive behavior. Therefore it was expected that for cluster 1 HR, RMSSD and RSA would remain unchanged and that HR of persons in cluster 2 would decrease and their RMSSD and RSA increase. We also questioned whether the physiological stress response for the clusters differed from each other. This was not the case. The results show that the CoVa had no influence on the physiologic stress reaction of inmates. As stated above the groups are created based on physiological characteristics. With reference to that physiological profile, one would also expect a behavioral profile. The expectation was, considering the Low Arousal Theory, that reactive aggression correlates with high HR and low HRV (cluster 2), and proactive aggression correlates with low HR and high HRV (cluster 1) (Stadler, Grasmann, Fegert, Holtmann, Poustka, & Schmeck, 2008; Hawes, Brennan, & Dadds, 2009; de Vries-Bouw, Popma, Vermeiren, Doreleijers, van de Ven, & Jansen, 2011). It was also expected that cluster 1 could be characterized by high scores on the Tri-PM, particularly on the subscale “meanness”. For cluster two we expected high scores on the Tri-PM subscale “disinhibition”. Results showed no difference between the clusters in psychopathy and aggressiveness. According to the mentors behavior did not improve or deteriorate, thus the CoVa had no effect on behavior that was being exhibited on the ward. Former research of James, Ogloff, Wong, & Greenwood (1990) shows that psychopaths are less motivated compared to non psychopaths prior to treatment. These findings accord with the observations of the trainers in the present study. Although the difference in motivation between
cluster 1 and 2 was not significant. If we take into account the cut off scores, cluster 1 is not motivated and cluster 2 is reasanably motivated.
Initially the groups were to be created on the basis of a HR of 1SD under the mean (group 1) and a HR of 1SD above the mean (group 2). However this was not possible because the groups became too small to perform any statistical analyses on. Therefore the groups had to be created in another way. This had as a consequence that the two groups did not consisted of merely extremes. This maybe explains why against expectation psychophysiological changes had occurred in both clusters and why the behavioral profiles were not reflected.
The findings of the current study are consistent with those of Bruce, McDermott, Fisher, & Fox, (2009) who found no changes in behavioral measures but did found changes in physiological measures. These findings corroborates the ideas of (Karayanidis, Robaey, Bourassa, De Koning, Geoffroy, & Pelletier, (2000), who suggested that this pattern of results represents a subtle change in cognitive processing that physiological measures are sensitive enough to detect but behavioral measures are not. The CoVa is an intervention designed for detainees with cognitive deficits. The training focuses on 4 domains: problem solving, critical/moral reasoning and perspective taking. The CoVa may have caused a subtle change in cognition, because the treatment changed the way they think and react. “Think first, than act” according to the slogan of the treatment. To examine this hypothesis the scores on the cognitive tasks (conducted in the lager project) before treatment should be compared with the scores on the cognitive tasks after treatment.
Our study contributes new data to the discussion on aggression, physiologic measures and cognitive behavioral therapy. Up to now no comparative research has done using behavioral and physiological measures to asses the effects of cognitive behavioral therapy on the psychophysiological functioning of adult male convicts.
This study had several methodological limitations and the outcomes have to be seen as preliminary. First, the sample size was too small. Second, there was no “no treatment control group”. Therefore it is not clear to what extend the CoVa is responsible for the psychophysiological changes. It is possible that the psychophysiological changes are due to the general effects of imprisonment. For
completeness a control group consisting of persons who do not exhibit antisocial behavior could also be added to the design. Third, the sample consisted only of male convicts. Whether these results could also be replicated with female convicts is unknown. Fourth, although the physiologic measures show that de D2 task induces stress, the D2 task is not particularly designed to induce stress. Fifth, the “mentor evaluation” (SDAS) scored the absence or presence of negative behavior, the “trainer evaluation” scored the absence or presence of positive behavior. This limited the possibility to compare the outcome measures.
According to Vaske (2011) Cognitive behavioral therapy may be effective in reducing problem
behavior, including crime, because the intervention changes cognition. It is very likely that changes in brain functioning and psychophysiology underlie the relationship between changes in cognitive function and changes in aggressive behavior. From the results it appears that behavioral measures may not be sensitive enough to distinguish certain groups and to detect small changes in cognition, while physiological measures maybe can. Although the physiologic results are not consistent with the expectations, they do show that the CoVa influences the physiology of male prisoners to some extend. This combination of findings provides some support for the premise that it is important to perform various types of assessments to observe small changes in cognition and to discriminate certain groups (for example PA aggressive and RA aggressive persons). Generally this does not happen. Most screening instruments are behavioral in nature.
This was a preliminary study and should be repeated with a larger research group, for better results. Several questions remain unanswered, future studies on the current topic are therefore
recommended. Research should focus on whether biological parameters have a predictive relation with different types of aggression and whether HR and HRV and HR and HRV reactivity is related to behavioral outcome measures after cognitive behavioral therapy.
In conclusion, the results of this study suggest that the CoVa might impact the psychophysiological functioning of aggressive inmates. Additionally, the results emphasize the importance of assessing behavioral and physiological measures together to reveal subtle differences in cognition. To this day, there have been few studies concerning the effects of treatment on psychophysiological factors. This is understandable considering the controversial nature of the subject, although we do want to emphasize the need for further research concerning this subject, because this may have great clinical implications. In the future it may be possible to determine the most appropriate treatment for each individual by using physiological screening in addition to regular screening.
REFERENCES
Andrews, D., Zinger, I., Hoge, R. D., Bonta, J., Gendreau, P., & Cullen, F. T. (1990). Does correctional treatment work? A clinically relevant and psychologically informed meta-analysis. Criminology , 28 (3), 369-404.
Aos, S., Miller, M., & Drake, E. (2007). Evidence-Based Public Policy Options to Reduce Future Prison Construction, Criminal Justice Coasts, and Crime Rates. Federal Sentencing Reporter , 19 (4), 275-290. Berntson, G. G., Cacioppo, J. T., & Quigley, K. S. (1991). Autonomic determinism: The modes of autonomic control, the doctrine of autonomic space, and the laws of autonomic constraint. Psychological Review , 98, 459-487.
Bolt, D. M., Hare, R. d., Vitale, J. E., & newman, J. P. (2004). A Multigroup Item Response Theory Analysis of the Psychopathy Checklist-Revised. Psychological Assessment , 16 (2), 155-168. Bruce, J., McDermott, J. M., Fisher, P. A., & Fox, N. A. (2009). Using Behavioral and
Electrophysiological Measures to Assess the Effects of a Preventive Intervention: A Preliminary Study with Preschool-Aged Foster Childeren. Prevention Science , 10 (2), 129-140.
Cima, M., Smeets, T., & Jelicic, M. (2008). Self-reported trauma, cortisol levels, and aggression in psychopathic and non-psychophatic prison inmates. Biological Psychology (78), 75-86.
Cornell, D. G., Warren, J., Hawk, G., Stafford, E., Oram, G., & Pine, D. (1996). Psychopathy in instrumental and reactive violent offenders. Journal of Consulting and Clinical Psychology (64), 783-790.
Crick, N. R., & Dodge, K. A. (1966). Social Information-Processing Mechanisms in Reactive and Proactive Aggression. Child Development (67), 993-1002.
De Geus, E. J., & van Doornen, L. J. (1996). Ambulatory assessment of parasympathetic/sympathetic balance by impedance cardiography. Computer-assisted psychological and sychophysiological methods in monitoring and field studies , 141-163.
De Geus, E. J., Willemsen, G. H., Klaver, C. H., & van Doornen, L. J. (1995). Ambulatory measurement of respiratory sinus arrhythmia and respiration rate. Biological Psychology , 3 (41), 205-227.
de Vries-Bouw, M., Popma, A., Vermeiren, R., Doreleijers, T. A., van de Ven, P. M., & Jansen, L. M. (2011). The predictive value of low heart rate and heart rate variability during stress for reoffending in delinquent male adolescents. Psychopgysiology , 48, 1596-1603.
Dowden, C., & Andrews, D. A. (2000). Effective correctional treatment and violent reoffending: a meta-analysis. Canadian Journal of Criminology , 42, 449-476.
Fishbein, D., & Sheppard, M. (2006). Assessing the role of neuropsychological functioning in inmates treatment response. .U.S. Department of Justice.
Gianaros, P. J., Van Der Veen, F. M., & Jennings, J. R. (2004). Regional cerebral blood flow correlates with heart period and high-frequency heart periot variability during working memory tasks:
Implications for the cortical and subcortical regulation of cardiac autonomic activity. Psychophysiology (41), 521-530.
Goedhart, A. D., van der Sluis, S., Houtveen, J. H., Willemsen, G., & De Geus, J. C. (2007). Comparison of time and frequency domain measures of RSA in ambulatory recordings. Psychophysiology (44), 203-215.
Hart, S. D., & Hare, A. E. (1996). Psychopathy and antisocial personality disorder. Current Opinion in Psychiatry (9), 129-132.
Hawes, D. J., Brennan, J., & Dadds, M. R. (2009). Cortisol, callous-unemotional, and pathways to antisocial behavior. Current opinion in psychiatry , 22, 357-362.
Hayano, J., Sakakibara, Y., Yamada, A., Yamada, M., Mukai, S., Fujinami, T., et al. (1991). Accuracy of assessment of cardiac vagal tone by heart rate variability in normal subjects. Am J Cardiol , 199-204. James, R. P., Ogloff, J. D., Wong, S., & Greenwood, A. (1990). Treating Criminal Psychopaths in a Therapeutic Community Program. Behavioral Science and Law , 8, 181-190.
Jose, A. D., & Collison, D. (1970). The normal range and determinants of the intrinsic heart rate in man. Cardiovasc Res. (4), 160-167.
Karayanidis, F., Robaey, P., Bourassa, M., De Koning, D., Geoffroy, G., & Pelletier, G. (2000). ERP differences in visual attention processing between attention-deficit hyperactivity disorder and control boys in the absence of performance differences. Psychophysiology , 37, 319-333.
Kempes, M., Matthys, W., de Vries, H., & van Engeland, H. (2005). Reactive and proactive aggression in children A review of theoryy, findings and relevance for child and adolescent psychiatry. Eur Child Adolesc Psychiatry (14), 11-19.
Lai-Chu Fung, A., Raine, A., & Gao, Y. (2009). Cross-Cultural Generalizability of the Reactive Proactive Aggression Questionnaire (RPQ). Journal of Personlity Assessment , 91 (5), 473-479.
Landenberger, N. A., & Lipsey, M. W. (2005). The Positive Effects of Cognitive-Behavioral Programs for Offenders: A Meta-Analysis of Factors Associated with Effective Treatment. Journal of
Experimental Criminology .
Lane, R. D., McRae, K., Reiman, E. M., Chen, K., Ahern, G. L., & Thayer, J. F. (2009). Neural correlates of heart rate variability during emotion. Neuroimage (44), 213-222.
Lemerise, E. A., & Arsenio, W. F. (2000). An Integrated Model of Emotion Processes and Cognition in Social Information Processing. Child Development , 71 (1), 107-118.
León, I., Hernándes, J. A., Rodríguez, S., & Vila, J. (2009). When head is tempered by heart: Heart rate variability modulates perception of other-blame reducing anger. Motiv Emot (33), 1-9.
Levy, M. N. (1990). Autonomic interactions in cardiac control. Ann. NY Acad. Sci. , 209-221. Lipsey, M. W., & Cullen, F. T. (2007). The effectiveness of correctional rehabilitation: A review of
Logan, G. D., & Cowan, W. B. (1984). On the ability to inhibit thought and action: A theory of an act of control. Psychological Review (91), 295-327.
Mullin, S., & Simpson, J. (2007). Does executive functioning predict improvement in offenders' behaviour following enhanced thinking skills training? An exploratory study with implications for rehabilitation. Legal and Criminological Psychology (12), 117-131.
Munoz, L., & Frick, P. J. (2012). Callous-Unemotional Traitl and Their Implication for Understanding and Treating Aggressive and Violent Youths. Criminal Justice and Behavior , 39 (6), 794-813. 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 and Adolescent Psychiatry (43), 154-162. Paquette, V., Lévesque, J., Mensour, B., Leroux, J., Beaudoin, G., Bourgouin, P., et al. (2003). "Change the mind and you change the brain": effects of cognitive-behavioral therapy on the neural coorelates of spider phobia. NeuroImage , 401-409.
Pearson, F. S., Lipton, D. S., Cleland, C. M., & Yee, D. S. (2002). The effects of behavioral/cognitive-behavioral programs on recidivism. Crime and Deliquency , 48, 467-496.
Raine, A. (2002). Annotation: The role of prefrontal deficits, low autonomic arousal, and early health factors in the development of antisocial and aggressive behavior in children. Journal of Child
Psychology and Psychiatry , 4 (43), 417-434.
Raine, A. (1996). Autonomic Nervous System Factors Underlying Disinhibited, Antisocial, and Violoent Behavior Biosocial Perspectives and Treatment Implications. (794), 46-59.
Raine, A., Dodge, K., Rolf, L., Gatzke-Kopp, L., Lynam, D., Reynolds, C., et al. (2006). The Reactive-Proactive Aggression Questionnaire Differential Correlates of Reactive and Reactive-Proactive Aggression in Adolescent Boys. Aggressive Behavior , 32, 159-171.
Riese, H., Groot, P. F., van der Berg, M., Kupper, N. H., Magnee, E. H., Rohaan, E. J., et al. (2003). Large-scale ensemble averaging of ambulatory independance cardiograms. Behavioural Res. Methods Instrum. Comput. , 3 (35), 467-477.
Sadlier, G. (2010). Evaluation of the impact of the HM prison service Enhanced Thinking Skills programme on reoffending outcome of the surveying proisoners crime reduction (SPCR) sample. Ministry of Justice Reasearch Series 9/10.
Scarpa, A., Haden, S. C., & Tanaka, A. (2009). Being hot-tempered: Autonomic, emotional, and behavioral distinctions between childhood reactive and proactive aggression. (84), 488-496.
Stadler, C., Grasmann, D., Fegert, J., Holtmann, M., Poustka, F., & Schmeck, K. (2008). Heart Rate and Treatment Effect in Children with Dirsuptive Behavior Disorders. Children Psychiatry and Human Decelopment , 39, 299-309.
Thayer, J. F., & Brosschot, J. F. (2005). Psychosomatics and psychopathalolgy: looking up and down from the brain. Psychoneuroendocrinology , 30, 1050-1058.
Thayer, J. F., Hansen, A. L., Saus-Rose, E., & Johnse, B. H. (2009). Heart Rate Variability, Prefrontal Neural Function and Cognitive Performance: The Neurovisceral Integration Perspective on Self-regulation, Adaptation, and Health. The Society of Behavioral Medicine , 37, 141-153.
van der Elst, W., van Boxtel, M. P., van Breukelen, G. J., & Jolles, J. (2006). The Stroop Color-Word Test Influence of Age, Sex an Education; and Normative Data for a Large Sample Across the Adult Age Range. Assessment , 1 (13), 62-79.
van der Wiel, H., Jeglic, E. L., & Donovick, P. J. (2004). Cortisol and treatment effect in children with disruptive behavior disorders: a preliminary study. Journal of Correctional Health Care (1), 51-60. van Goozen, S. H., & Fairchild, G. (2008). How can the study of biological processes help design new interventions for children with severe antisocial behavior? Psychopathology (20), 941-947.
Vaske, J., Galyean, K., & Cullen, F. T. (2011). Toward a biosocial theory of offender rehabilitation: Why does cognitive-behavioral thrapy work? Journal of Criminal Justice .
Verbruggen, F., Logan, G. D., & Stevens, M. A. (2008). STOP-IT: Windows Executable software for the stop-signal paradigm. Behavior Research Methods , 2 (40), 479-483.
Walsh, A., & Beaver, K. M. (2009). Biosocial criminology: New directions in theory and research. New York: Routledge (Eds.).
Wright, J. P., & Tibbetts, S. G. (2008). Criminals in the making: Criminality across the life course. Thousand Oaks, CA: Sage .
