Determining the etiology of antisocial behavior/aggression is critical. Antisocial behavior and aggression generally are a major cause of child and adolescent referrals to mental health clinicians. Antisocial/aggressive individuals cost society up to ten times more than their healthy counterparts in aggregate health care and social service expenditures. Moreover, a variety of psychiatric disorders are associated with an increased risk for aggression (e.g., anxiety disorders, particularly posttraumatic stress disorder; mood disorders, particularly bipolar disorder, and; personality disorders, particularly psychopathy). At least two forms of aggression can be distinguished from a behavioral and a neurobiological perspective: reactive and instrumental aggression (Blair, 2008). Reactive aggression is unintentional and impulsive and typically accompanied by negative emotional states (e.g., anger, frustration, and hostility). Instrumental aggression is directed; the antisocial behavior is used to achieve the individuals goal (e.g., gaining money). Typically, there is a relative absence of intense emotion. Mood and anxiety disorders selectively increase the risk for reactive aggression. Psychopathy increases the risk for both reactive and instrumental aggression.
There have long been suggestions that reduced levels of empathy are associated with an increased risk for antisocial behavior (e.g., Miller & Eisenberg, 1988). The article in this issue by Rhee and colleagues on typically developing children (Rhee et al., 2012) is important because it is one of the few studies to longitudinally examine the relationship between early, behavioral indicators of empathic responding (assessed at age 14–36 months) and later antisocial behavior (assessed at age 4–17 years). Particularly, usefully empathic responding in this study is not only indexed by maternal reports on the child but also observations of the child’s reactions to the pain of others (indeed it was these observational measures that were the best predictors of future antisocial behavior).
Empathy research has been hampered by definitional issues with very different computational processes being referred to as empathy. However, a consensus is emerging allowing a distinction between cognitive empathy (the representation of the mental states of others) and emotional empathy (the emotional response to another individual’s emotional display) based on differential neurobiology (e.g. Blair, 2008). The study by Rhee and colleagues (Rhee et al., 2012) interestingly examines the predictive power of two different forms of response to the distress of others: (a) ‘concern for others’ marked by proximity to the victim and prosocial helping and comforting behavior toward the victim and; (b) ‘disregard for others’ indexed by anger, hostility, and aggression toward (or laughter at) the victim or the object causing the distress. Although these two types of response probably recruit both cognitive and emotional empathy, they appear most to reflect two behavioral manifestations of an emotional response to a victim’s distress. Unsurprisingly, Rhee and colleagues’ data here show that the likelihood of presenting with ‘concern for others’ was inversely related to the likelihood of showing ‘disregard for others’. Very interestingly though, Rhee et al. report that it is only behaviors associated with ‘disregard for others’ that longitudinally predict antisocial behavior.
The psychiatric condition where reduced emotional empathy is causally related to the development of antisocial behavior and aggression is psychopathy (Blair, 2008). Indeed, reduced emotional empathy is part of the definition of psychopathy (the affective, callous–unemotional component of psychopathy). Other psychiatric conditions (e.g., mood and anxiety disorders) associated with increased aggression are not so obviously linked to reduced empathy – and in fact are associated with heightened emotional responding to provocative stimuli. However, the data here do not neatly fit into the psychopathy literature. Psychopathy is associated with a reduction in responses to the distress of others whether these are indexed by autonomic activity or amygdala responding. Reduced responsiveness would suggest that the behavioral reaction to the distress of others would also be muted. But Rhee et al. did not observe a relationship between a muted response to another’s distress and future antisocial behavior. Instead, they found that a form of increased response (involving hitting behavior, anger, and hostility) was associated with future antisocial behavior.
Several possibilities can be considered for these data. One possibility is that the response to the distress cues of others typically inhibits the behaviors associated with ‘disregard for others’ and thus the children with greater psychopathic tendencies showed the greatest amounts of these behaviors. Such a hypothesis might be compatible with the increased laughter at the victim’s plight. However, it is not clear how this hypothesis would predict that increased levels of anger, hostility, and attacks on the offending object would be associated with future antisocial behavior (although it might relate to the nonclinical nature of the sample).
A second possibility is that behaviors associated with ‘disregard for others’ reflect heightened emotional volatility; i.e., an increased, albeit nonoptimal emotional empathic response. Reactive aggression (also associated with anger and hostility) reflects a heightened response to emotional provocation. This second possibility is particularly interesting given Rhee and colleagues’ behavioral genetic results. They found that all of the covariance between observed disregard and antisocial behavior was due to shared environmental influences. Moreover, the heritability of observed ‘disregard for others’ was near zero. In contrast, considerable research now indicates that the empathy dysfunction associated with psychopathy, the observed callous–unemotional traits, is highly heritable (e.g., Viding, Blair, Moffitt, & Plomin, 2005). However, environmental variables, particularly those relating to stress and neglect, have been shown to increase the sensitivity of the neural systems mediating reactive aggression (particularly in specific genetic contexts); i.e., these environmental variables increase emotional volatility.
It is worth considering this last point a little more. As noted, Rhee and colleagues found that all of the covariance between observed disregard and antisocial behavior was due to shared environmental influences. These data suggest that interventions addressing these environmental influences might reduce the development of antisocial behavior – at least in nonclinical samples. There is now a considerable literature showing that Social Emotional Learning programs do reduce the development of antisocial behavior (Durlak, Weissberg, Dymnicki, Taylor, & Schellinger, 2011). Such programs provide instruction in social and emotional skills and establish safe, caring learning environments within the home and classroom. In other words, these programs address the environmental influences that increase emotional volatility.
Of course, the above possibilities remain speculative. However, they raise clear predictions. Given this, it will be interesting to learn what the relationship between the affective (callous–unemotional) component of psychopathy and the behavioral response to another’s distress is. It will be interesting to determine what the impact of environmental variables such as stress, neglect, and socialization practices have on the behavioral response to another’s distress. Rhee and colleagues are well positioned to address these issues. Certainly, the current study, showing that the behaviors associated with ‘disregard for others’ at age 14–36 months predict antisocial behavior (assessed at age 4–17 years), stresses the importance of understanding empathy for understanding the development of antisocial behavior. Moreover, the behavioral genetic findings again stress the importance of psychosocial intervention to prevent the development of these problematic behaviors.