Facts, Frameworks, and Forecasts

Joan McCord

• Publisher: Routledge
• Year: 2017
• Print ISBN: 9781138523326
• eBook ISBN: 9781351520775
• Topic: Social Sciences
• Subtopic: Criminology
• Index: Social Sciences

Facts, Frameworks, and Forecasts calls for rethinking the development of criminological theory. In her introduction, Joan McCord argues that the field is ready for new approaches and that its progress depends on a sound factual base. Examining the discipline's research design, methodology, and quantitative analysis efforts, the contributors identify significant advances in criminological theory. This empirical orientation results in a balanced blend of information and speculation.This book contains a comprehensive review. The first chapter describes biological conditions that have theoretical links with criminal behavior - ending with a discussion of how biological and social conditions may interact to influence criminal behavior. Early chapters discuss general issues related to crime. These are followed by expositions of theoretical orientations not typically found in criminological literature. The second half of the book describes seven longitudinal studies in four countries. The authors interpret their data to expose biological, social, and psychological factors they believe may influence criminal behavior.These contributors include: Guenther Knoblich and Roy King, Daniel Glaser, Robert A. Rosellini and Robin L. Lashley, Robert J. Sampson, Ellen S. Cohn and Susan O. White, Joan McCord, L. Rowell Huesmann and Leonard D. Eron, Robert Cairns and Beverly Cairns, Richard E. Tremblay, Patricia Cohen and Judith S. Brook, David P. Farrington and David Magnussen, Britt af Klinteberg, and Hakan Stattin.Facts, Frameworks, and Forecasts addresses the observation of noted criminologist Marvin Wolfgang that criminological theory had stagnated. This groundbreaking work, available in paperback for the first time, is as relevant now as when first published. It should be read by all concerned with data-related approaches to criminology.

1

Biological Correlates of Criminal Behavior

Guenther Knoblich and Roy King

The origins of antisocial behavior have been studied from multiple vantage points including sociological, psychological, and biological perspectives, all of which have competed somewhat unnecessarily to be recognized as possessing the definitive model of criminal behavior. An accurate review would transcend overly reductive boundaries of academics to draw a more complete picture of the criminal. Therefore, although this review will focus on the biological concepts of violent behavior, it will be informed by current psychosocial models of criminality. With the recent development of strict criteria in the psychiatric literature for the diagnosis of antisocial personality disorder, it has been possible to start to delineate its etiology. Unfortunately, psychiatric literature has not fully considered the broader concept of psychopathy. That is, antisocial personality disorder emphasizes behavioral misconduct continuing from childhood to adulthood rather than interpersonal defects such as lack of empathy and egocentricity that are encompassed by the construct of psychopathy. This neglect of the idea of psychopathy is unfortunate because in attempting to uncover the etiology of enduring behavior, it is useful to look at more specific characteristics of individuals and how those contribute to the development of long-lasting conduct as seen in sociopathy. It is not a priori clear that biological differences would predict misconduct as well as it does interpersonal-affective problems; however, such differences are helpful in uncovering how affective disorders affect behavior, violent or otherwise. Hence, the biological paradigm of human behavior is most useful when looking at how such environmental influences as isolation, neglect, abuse, and other conditioning variables are imprinted into individuals to cause predictable, incorrigible behavioral disorders years later. So biological psychiatry does not contradict any of the conclusions of other fields as to the importance of contextual variables on personality and behavior; it simply addresses the possibility that these conditioning experiences are mediated into long-ranging behavior through known and newly discovered biological and neurobiological systems. Indeed, many of the animal models of behaviors hypothesized to be related to psychopathy, such as aggressiveness, fit a social interaction paradigm. Therefore, a useful scheme for integrating these viewpoints is to segment the complex construct of antisocial behavior to consider it a composite of more easily investigated factors such as arousal, dominance, and aggression. Each of these component factors has been studied in humans using a dimensional model of personality traits leading to interesting hypotheses about the relationship between psychosocial variables, neurobiological function, and personality. The structure of this paper will be to report on studies of impulsivity and related traits, such as sensation seeking and extroversion, as well as presenting important studies on human aggressiveness. Finally these two sets of findings will be integrated to organize a conceptual heuristic for exploration of models of sociopathy.

1 Arousal models of psychopathic behavior

The classic neurobiological explanation of criminal behavior suggests that individuals prone to engage in nonconforming, impulsive, or rule breaking behavior have low levels of baseline “arousal” in some central arousal system. Several measures of arousal have been investigated including peripheral autonomic activity, electro-encephalographic (EEG) patterns, plasma/urinary catecholamine levels, and plasma/urinary cortisol levels. Thus certain studies show that psychopaths have lower skin conductance levels (Raine, Venables, and Williams 1990), others indicate that these subjects often have increased slowing of the EEG similar to a drowsy pattern (Howard 1984). Likewise, a study of conduct-disordered children showed them to have lower indices of cardiovascular arousal; namely, lower heart rate and blood pressure (Rogeness et al. 1990). On the neurochemical side, urinary adrenaline appears to be reduced in certain impulsive and psychopathic individuals (Lindberg et al. 1978). A problem with generalizing from these findings is that most evidence suggests that there are multiple arousal systems in the brain. For example, autonomic arousal and electrocortical arousal may not be tightly linked (Raine, Venables, and Williams 1990) and peripheral catecholamine levels often fluctuate independently of corticosteroid levels in a variety of experimental paradigms (Coover 1983). Thus, some of the failure to replicate this low arousal theory of psychopathy may be related to the individual response specificity of a particular person’s arousal mechanisms as well as to the influence of situational factors on modifying arousal levels. Indeed, the conflicting reports on psychopathy and arousal may be due to protocols that have only looked at one arousal parameter. One way of circumventing this is assessing arousal across the variety of response systems and weighting these measures to optimally distinguish an overall arousal factor for the individual. Raine et al. attempted such an unique design in which psychophysiological measures of male adolescents were taken at age fifteen in order to relate low arousal in response modes to criminal behavior at age twenty-four. In particular, those with criminal records at twenty-four had lower skin conductance levels, low resting heart rate and slower EEG activity at age fifteen than their cohorts with no criminal status. In fact, this discriminate analysis showed promise in distinguishing criminal from noncriminal groups (Raine, Venables, and Williams 1990). This evidence within a single study implicates the interrelationship between both cortical and autonomic arousal and antisocial acts. However, such mathematical manipulations involving weighted scores are often difficult to replicate across studies and ignore the qualitative distinctions between the arousal systems themselves. Also, although we have thus far focused on conventional arousal system correlates and predictors of criminality, there exist less well studied systems of equal importance, such as neuroendocrine and biochemical arousal states. Several studies, using populations of criminals and individuals with criminal and criminal related behaviors, have looked at neuroendocrine levels of cortisol, a hormone implicated in many different forms of stress response. Virkunnen (1985) has shown that habitually violent sociopaths had lower levels of urinary cortisol that other less violent offenders. In addition, King et al. (1990) found that plasma cortisol was lower in a sample of patients with a diagnosis of substance abuse than controls. Furthermore, within the group of normals, impulsivity as measured by the Eysenck Personality Inventory correlated negatively with plasma cortisol. This result is interesting in light of Ballenger’s findings that, within a group of normals, cortisol was negatively correlated with the mania scale of the MMPI (Ballenger, Post, and Goodwin 1983). Thus, the measures of activity of the hypothalamic-pituitary-adrenal axis seem to reflect a lower baseline arousal in impulsive, disinhibited, or violent individuals. This is consistent with measurements of other metabolic parameters showing a less excitable state. For instance, low glucose nadirs upon a glucose tolerance test (GTT) have been found in habitually violent sociopaths (Virkunnen 1982). This probably reflects a parasympathetic bias in these individuals causing a release of pancreatic enzymes from sympathetic inhibition. This is supported by findings of low GTT levels and increased insulin secretion in arsonists and individuals with intermittent explosive disorder (Virkunnen 1986).

Most frequently the arousal models of psychopathy predict that in psychopaths, anticipatory anxiety is reduced to an aversive stimuli. Thus, psychopaths may show deficient socialization, at least compared to individuals with normal or high levels of arousal. This assumes, of course, that much of socialization may involve the inhibition of impulsive or reward driven behavior. A second interpretation of the low arousal models is the idea that for every individual there is an optimal level of arousal, and that the relationship between arousal level and performance follow an inverted U-shaped curve. Psychopaths situated at the lower end of the arousal performance curve can increase performance and/or pleasure through manipulating the environment to increase their level of arousal. Such behavior is often termed sensation seeking behavior, and although it is classically associated with criminal acts such as violence or illicit drug use, such arousal-driven behavior is also purported to fall with a dimension of normal variations, embodied by activities such as rock climbing or skydiving. However, such models simplify the complex dynamics of interpersonal processes. All of social behavior cannot be reduced to punishment; for instance, prosocial interactions can clearly, of themselves, be rewarding. Thus, low arousal per se may not lead to antisocial acts but could lead to heightened sociability under certain circumstances. It is possible that interpersonal defects such as those defined under the characteristic of psychopathy (or using a more recent personality disorder scheme, those who show narcissistic qualities) would be prone to manifest enduring antisocial acts. Thus, low arousal may be a necessary but not complete explanatory factor in the construct of criminality.

2 Testosterone and Aggression

Few neurobiological models have received such widespread societal attention as that of testosterone and its effects on aggression. Associations between this steroid hormone and violent and criminal behavior appear in popular media on a regular basis. However, the ideas surrounding this biological model within the scientific community are unclear at best. Numerous groups have studied the effects of testosterone on behavior dating back over forty years (Beeman 1947), but the mass of data produced in those four decades has failed to provide a cohesive, encompassing picture of its role in human behavior. It has, however, provided a base of findings that may prove helpful in the ensuing years in which new methodology and innovative techniques in the field of neurobiology may produce a clearer model of aggression and its biological source.

The many animal and human studies on the effects of testosterone on behavior break up into three main groupings. First, studies focusing on developmental effects have shown associations between higher levels of the androgen and later aggression, as is seen in the behavior of offspring when natural prenatal steroid exposure was extrapolated from intrauterine contiguity of male fetuses (Gandelman, vom Saal, and Reinisch 1977). The second related type of animal studies are those in which testosterone has been shown to have significant effects on later aggression when animals are exposed in the perinatal period (Bronson and Desjardins 1968; Edwards 1969), during which important neural connections are made and neural development occurs rapidly, and environmental stimuli or their absence can be predicted to have long lasting effects. This effect of androgen on the development of aggression has also been replicated in humans, linking umbilical levels of testosterone to aggression some eighteen months down the line. The findings of increased testosterone in more aggressive boys strongly points at a steroid-dependent neural development of antagonistic behavior, even at the human level (Jacklin, Maccoby, and Doering 1983). Both of these first two effects are clearly mediated by regulation of the genome at a molecular level, determining later factors such as neurotransmitter receptor number and other “hardwiring” of the neural system, and surely further research in this direction will produce a clearer understanding of violent behavior, the cause of its genetic linkage, and the role of testosterone at a molecular level.

There is a third area of study, however, that is much less clear in terms of biological modeling. This area is that of the effects of testosterone levels on the adult, in which environmental factors make the association between biological findings, such as testosterone, and overt behavior, such as aggression, much more complex. In order to organize the mass of observations it helps to define precisely which types of behavior we mean by aggression. Moyer (1976) has broken these up into six categories, two of which, intermale and irritable aggression, seem to be related to testosterone and which break up the different observations well.

Intermale aggression is, as the name indicates, aggression between two males, which in animal studies is characterized by attacks, grabbing, biting, and other dominance-related actions. Such aggression can be seen to be testosterone dependent through use of animal experiments. For instance, castrated mice will not display intermale aggression while their castrated cohorts with testosterone implants do (Albert, Dyson, and Walsh 1987; Albert, Dyson, Walsh, and Wong 1988). Such behavior is not only dependent on the steroid, however, as the environmental stimulus of provocation is also necessary to produce aggressive attacks, using such experimental paradigms as competition for food (Albert, Petrovic, and Walsh 1989). This fits the human data in which in a population of normal males, levels of testosterone were best correlated with aggression indices that included provocation (Olweus et al. 1980).

Such studies are interesting in that they indicate that there exists an aggression that is state dependent. That is, given similar development, a change in aggression can be seen in the adult by changes in testosterone levels. This observation is supported by clinical anecdotal reports that describe previously psychiatrically normal young men who begin to use anabolic steroids and suddenly commit suicide (Brower et al. 1989), homicide, and other violent, antisocial acts (Pope and Katz 1990). This change in behavioral state accompanying biological state changes also fits the present prevailing “biosocial theory of status” (Mazur 1985). That is, a change in a biological parameter effects a state change in an individual’s temperament or behavior, and then this behavior feeds back on the level of the biological progenitor itself. Similar feedback systems have been implicated in other neurobiological systems associated with behaviors such as drug addiction (King et al. 1990). In the case of testosterone, this theory would indicate that the increase in intermale aggression seen with increases of testosterone would effect a rise in the individual’s dominant status by augmentation of winning in social conflicts. This winning behavior would then regulate the testosterone levels themselves. Such a model is supported not only by the previously mentioned correlations of testosterone and aggressive attack, but by data showing that winning causes increases in testosterone in animal studies (Rose, Bernstein, and Gordon 1975) as well as in human studies of college tennis players (Booth et al. 1989) and wrestlers (Elias 1981). Additionally, the increase in testosterone is seen not only in individual wins but is sequentially elevated by each victory, increasing the chance of social dominance, presumably by increasing aggression, and setting up the type of cybernetic feedback that Mazur’s theory implicates in enduring behavior.

The other type of aggression detailed by Moyer is irritable aggression. This subset of behaviors has less easily identifiable releasers and therefore corresponds to such behaviors as spontaneous, unprovoked violence. This paradox of an aggression system independent of stimuli seems to contradict theories that all biological response is environmentally bound, and seems to form a more deterministic picture of behavior. Indeed, an unsophisticated view of any of the correlational studies would imply that testosterone is a causal factor in violence. This is obviously not the case, as it is doubtful that any neurobiological system follows such simplistic cause-effect relationships. There exist several possible alternatives that may explain this concept of spontaneous irritable aggression as well as the contradicting evidence against testosterone affecting adult aggression. First, retrospective correlation studies of human aggression rely on baseline levels of androgen correlated to either self-reported feelings of aggression or history of antisocial acts. However, in primates, the relationship between testosterone and aggression is related to the social and developmental context in which one observes the behavior (Coe and Levine 1983; Coe, Smith, Mendoza, and Levine 1983). Furthermore, Sapolsky has found that although baseline levels of testosterone were unrelated to dominance behavior, the response of circulating testosterone to stress in individuals differs (Sapolsky 1982, 1986). Taking these facts together, it is plausible that irritable aggression may actually not be a trait-like behavior, but simply dependent on the social contex...

Table of contents

• Cover
• Title Page
• Copyright
• Table of Contents
• Introduction
• 1. Biological Correlates of Criminal Behavior
• 2. Reconceiving Some Confounding Domains of Criminology
• 3. Opponent-Process Theory
• 4. Family Management and Child Development
• 5. Taking Reasoning Seriously
• 6. Understanding Motivations
• 7. Childhood Aggression and Adult Criminality
• 8. The Sociogenesis of Aggressive and Antisocial Behaviors
• 9. The Prediction of Delinquent Behavior from Childhood Behavior
• 10. A Developmental Perspective on Drug Use and Delinquency
• 11. Explaining the Beginning, Progress, and Ending of Antisocial Behavior from Birth to Adulthood
• 12. Autonomic Activity/Reactivity, Behavior, and Crime in a Longitudinal Perspective
• Contributors
• Subject Index
• Name Index