Impulsivity and aggression: A meta-analysis using the UPPS model of impulsivity☆
Introduction
Aggression and violence are prevalent and costly public health problems that have negative consequences on victims (e.g., injury or death) and perpetrators (e.g., incarceration). Decades of research in anthropology (e.g., Fry, 1998), neuroscience (Gollan, Lee, & Coccaro, 2005; Rosell & Siever, 2015; Verona & Bresin, 2015), psychiatry (Coccaro, 2012; Vitiello & Stoff, 1997), and psychology (Anderson & Bushman, 2002; Berkowitz, 1989; Dodge, 1991; Finkel & Hall, 2018; Wilkowski & Robinson, 2010), among other disciplines has provided some insights into the etiology of aggression; however, more work is needed to identify individual differences that exacerbate or attenuate the likelihood of responding aggressively in order to develop effective prevention and intervention strategies to reduce the public health burden of aggression. Trait impulsivity has long been proposed to play a role in aggression, and there is empirical support for this prediction (e.g., Miller, Zeichner, & Wilson, 2012; Weiss, Connolly, Gratz, & Tull, 2017). Because impulsivity is a multifaceted construct, what remains to be elucidated is whether some, but not all, facets of impulsivity are related to aggression and whether aggression is more strongly related to some aspects of impulsivity than others. The goal of this paper was to conduct a meta-analysis of the relation between impulsivity and aggression using a unifying multifaceted model of trait impulsivity (i.e., the UPPS model; Cyders et al., 2007; Whiteside & Lynam, 2001) to determine the association between unique aspects of impulsivity and aggression.
Research on trait impulsivity has a long history in psychology and has been a focus of personality, clinical, and developmental psychologist (Barratt, 1993; Eysenck & Eysenck, 1985; Rothbart, 2007; Whiteside & Lynam, 2001). Over the decades of research, several scales have been created to assess impulsivity (Dickman, 1990; Eysenck, Pearson, Easting, & Allsopp, 1985; Patton, Stanford, & Barratt, 1995; Zuckerman, 1994). These scales, which came from different theoretical backgrounds and areas in psychology, had some conceptual and empirical overlap, but also areas of divergence, which led some to believe that impulsivity was not a unitary construct. In order to clarify the structure of trait impulsivity, Whiteside and Lynam (2001) conducted a factor analysis across several commonly used impulsivity scales. Their results showed a four-factor structure: (negative) urgency, lack of premeditation, lack of perseverance, and sensation seeking. A fifth factor, positive urgency, was later added by Cyders et al. (2007; Cyders & Smith, 2008).
Negative urgency refers to an individual's tendency to engage in rash action when feeling negative emotions, whereas positive urgency refers to an individual's tendency to engage in impulsive behavior when feeling positive emotions. Both urgency traits are related to neuroticism in the Big 5 model of personality (Cyders & Smith, 2008). Lack of premeditation refers to an individual's propensity to act without planning. Lack of perseverance indicates an individual's tendency to leave tasks unfinished or to give up. These two factors are related to conscientiousness in the Big 5 (Cyders & Smith, 2008). The final factor, sensation seeking, indicates an individual's engagement in thrill-seeking behaviors and is related to extraversion (Cyders & Smith, 2008).
These studies by Whiteside and Lynam (2001) and Cyders et al. (2007), among others, are important in many ways. First, they provided a unifying framework to organize the multiple facets of impulsivity, which can be used to make predictions about how different facets of impulsivity might be related to different behaviors. For example, a meta-analysis found that negative urgency is more strongly related to nonsuicidal self-injury than lack of perseverance (Hamza, Willoughby, & Heffer, 2015). Similar differential correlations have been found for alcohol and marijuana use (Coskunpinar, Dir, & Cyders, 2013; VanderVeen, Hershberger, & Cyders, 2016). Another advantage of the UPPS model, particularly for meta-analysis is that because it was designed from other impulsivity measures, it is possible to place studies that did not use the UPPS scales into the factor structure. This is particularly advantageous for meta-analysts who wish to include studies prior to 2001.
Throughout time researchers have used several definitions of aggression (e.g., Buss, 1961; Dollard, Doob, Miller, Mowrer, & Sears, 1939; Parrott & Giancola, 2007; Zillmann, 1979). This paper uses the definition put forth by Parrott & Giancola, 2007 among others (e.g., Anderson & Bushman, 2002; Baron & Richardson, 1994) that aggression is a behavior direct toward another with the goal of causing physical or psychological harm that the target wants to avoid. Aggression then is distinct from anger, which is an emotion often associated with aggression, and hostility, which are thoughts associated with aggression (Parrott & Giancola, 2007). Similar to impulsivity, aggression is a heterogeneous construct and theorist has suggested several ways to divide aggression into more homogeneous categories (e.g., direct versus indirect; active versus passive; Gollan et al., 2005; Lesch & Merschdorf, 2000; Parrott & Giancola, 2007).
Based on the available studies, I focused on two dimensions: form and function. The form of aggression refers to the method that is used to deliver harm to another (Anderson & Bushman, 2002; Verona, Sadeh, Case, Reed II, & Bhattacharjee, 2008). Common methods include physical (e.g., punching), verbal (e.g., insulting), relational (spreading rumors), and sexual (insisting that a partner engage in sexual acts; Buss & Perry, 1992; Crick, Ostrov, & Werner, 2006; Straus, Hamby, Boney-McCoy, & Sugarman, 1996; Verona et al., 2008). Different forms of aggression tend to have medium-to-large size correlations with each other (e.g., Evans et al. 2018; Verona et al., 2008), suggesting that people who engage in one form of aggression are likely to engage in multiple forms of aggression. Some scales have unique subscales for each form of aggression (Buss & Perry, 1992; Raine et al., 2006; Straus et al., 1996), whereas other scales consider multiple forms together to create a general aggression score (e.g., Coccaro et al., 1997).
The functions, or purposes, of aggression, are usually divided into proactive (sometimes premeditated) and reactive (also known as impulsive or hostile; Crick & Dodge, 1996; Raine et al., 2006). Proactive aggression is where the physical or psychological pain inflicted is a means to an end. That is, aggression (e.g., hitting someone) is used to obtain some other outcome (e.g., money). Reactive aggression is aggression, where harming the target is the ultimate goal, and this is in response to some form of real or perceived provocation. Some have called into question the utility of the reactive/proactive distinction given that acts of aggression fit neatly into one category (Barratt & Slaughter, 1998; Bushman & Anderson, 2001). In line with this, meta-analytic research has shown a large correlation between reactive and proactive aggression (Polman, de Castro, Koops, van Boxtel, & Merk, 2007). Moreover, many aggression measures do not make the distinction between reactive and proactive aggression. Despite this overlap, there is evidence of unique associations when adjusting for overlap between reactive and proactive aggression (e.g., Miller et al., 2012). Furthermore, there is theoretical overlap between aspects of impulsivity and reactive and proactive aggression; therefore, it seemed useful to examine in this meta-analysis.
Across theories of aggression, there are at least two common aspects, reactivity, and failure to think ahead that are proposed to play a critical role in aggressive behavior(Anderson & Bushman, 2002; Berkowitz, 1989; Verona & Bresin, 2015; Wilkowski & Robinson, 2010). For instance, I3, which is a meta-theory of aggression, suggests that aggression is a function of three processes or forces: instigation, impellance, and inhibition (Finkel & Hall, 2018). Instigators are situational factors that create the possibility of an aggressive response (e.g., being cut-off in traffic). Impellors are dispositional (or situational) factors that increase the urge to respond aggressively (i.e., reactivity) to the instigator (e.g., generally being an aggressive driver). Inhibitors are factors that help the individual withhold the aggressive response (i.e., inhibit a prepotent response) in the face of urges (e.g., the presence of a police officer) and disinhibitors are factors that increase the likelihood of an aggressive response (e.g., alcohol intoxication). This framework may help make predictions for which aspects of impulsivity might be related to aggression.
Negative urgency (and to a lesser extent positive urgency) is a trait that could increase the risk of aggression in two ways. First, urgency may function as an impellor and increase reactivity to aggression-instigating situations. That is, for the same aggression-instigating event (e.g., being cut off in traffic) individuals high in negative urgency, may have a stronger anger reaction than individuals low in negative urgency. This fits with the idea that negative urgency is closely related to neuroticism (Cyders & Smith, 2008). Second, urgency may also be a disinhibitor, making it easier for aggressive responses to occur when provoked. By definition, urgency is acting rash in the face of affect. This could be easily translated into acting aggressively in the face of anger. Along these lines, studies have found significant positive associations between negative and positive urgency and aggression (Derefinko, DeWall, Metze, Walsh, & Lynam, 2011; Leone, Crane, Parrott, & Eckhardt, 2016; Miller et al., 2012).
Lack of premeditation could increase aggression primarily as a disinhibitor. On the one hand, people who have a difficult time planning for the future will likely have difficulty thinking about the future consequences of their aggressive actions (e.g., causing problems in a relationship). This may allow aggressive behavior to occur in the presence of an instigator. On the other hand, people who are high in premeditation may be better able to inhibit aggression when the urge is high because they can think about the long-term repercussions of behaving aggressively. Understanding these long-term consequences may help these individuals quell their urges to aggress. Consistent with this, research has found positive correlations between lack of premeditation and aggression (Derefinko et al., 2011; Leone et al., 2016).
The final two aspects of impulsivity, sensation seeking and lack of perseverance, are presumed to have weak positive associations with aggression. Due to their quest for novelty, individuals high in sensation seeking may be more likely to be in aggression-instigating situations. For instance, someone high (versus low) in sensation seeking might be more likely to provoke someone to “get a rise” out of them, which may result in an aggressive interaction. Lack of perseverance may be related to aggression to the extent that inhibiting an aggressive urge requires persistence; however, lack of perseverance, at least as measures by the UPPS scales, is focused more on task persistence than emotion regulation, as evidenced by its association with conscientiousness (Cyders & Smith, 2008). In line with this, research using the UPPS has found weak and inconsistent correlations between sensation seeking and lack of perseverance and aggression (Derefinko et al., 2011; Latzman & Vaidya, 2013; Leone et al., 2016).
It is possible that different aspects of impulsivity might have stronger (or weaker) relations with different forms and/or functions of aggression. Individual studies have generally found similar correlations between facets of impulsivity and aggression across different forms of aggression (e.g., Carlson, Pritchard, & Dominelli, 2013; Weiss et al., 2017). Thus, there is little a priori reason to predict different relations among facets of impulsivity and different forms of aggression. Still, an empirical test would be helpful for future theory development. Prior studies have shown similar sized zero-order correlations between different facets of impulsivity and reactive and proactive aggression (Hecht & Latzman, 2015; Miller et al., 2012). When considering the unique variance (i.e., adjusting for overlap between functions), different results have appeared for the distinct facets of impulsivity. Specifically, negative urgency appears to have a significant positive relation to the unique variance of reactive aggression (Hecht & Latzman, 2015; Miller et al., 2012). This is consistent with the intuition that people who tend to engage in impulsive behavior when upset may be more prone to engage in aggression that is reactive. There is also some evidence that lack of premeditation is positively related to the unique variance of proactive aggression (Hecht & Latzman, 2015).
The goal of the current meta-analysis was to determine the relation between the different facets of impulsivity, as defined by the UPPS, and aggression. This was done using two-stage meta-analytic structural equation modeling, which allowed for the examination of zero-order relations and relations adjusting for the other facets of impulsivity.1 Based on prior research and theory (Miller et al., 2012; Weiss et al., 2017; Cyders & Smith, 2008), it was predicted that negative urgency, positive urgency, and lack of premeditation would have significant and small-to-medium zero-order and partial correlations with aggression. No strong predictions were made for sensation seeking or lack of perseverance as results have been fairly inconsistent in prior studies.
In addition to the primary goal, I also conducted moderation analyses to see whether the correlations were similar across different forms and functions of aggression and across different samples. It was assumed that the correlation between different facets of impulsivity and aggression would be similar across different forms of aggression (e.g., physical, verbal). For function, it was predicted that negative and positive urgency would be related to the unique variance of reactive aggression. In addition to form and function, I explored whether sample demographic characteristics (e.g., sample age, sample gender, sample type) were related to the correlations between impulsivity and aggression in the Supplemental Materials.
Section snippets
Literature search
Search terms for impulsivity were derived from prior meta-analyses using the UPPS model (Coskunpinar et al., 2013; Kale, Stautz, & Cooper, 2018). Specifically, I searched the combination of each scale (see Supplemental Table 1) that has been proposed to map onto UPPS with the word aggression. The search yielded 627 unique papers. Fig. 1 shows the number of papers involved in each stage of determining eligibility. During the search, each paper's method section was reviewed to determine whether
Study characteristics
Table 1 displays the aggregate characteristics of the studies included in the meta-analysis and Table 2 displays this information by study. The average sample size across the samples was 345, which according to G*Power (Faul, Erdfelder, Lang, & Buchner, 2007) would give the power of 0.45 to detect small correlations as defined by Cohen (r = 0.10; 1988) and 0.99 power for medium and large correlations (r > 0.30). The samples were fairly young and contained more men than women. The studies that
Discussion
The goal of this meta-analysis was to determine the relation between the unique facets of impulsivity and aggression. The results showed significant and small-to-medium correlations between each facet of impulsivity and aggression across several different forms of aggression. Consisting with my prediction negative urgency and lack of premeditation had significantly stronger associations with aggression than did sensation seeking and lack of perseverance. These results occurred both for
Declaration of competing interest
None.
References2 (156)
- et al.
Suicidal and violent behavior: The role of anger, emotion dysregulation, and impulsivity
Personality and Individual Differences
(2015) - et al.
Impulsive and premeditated aggression: A factor analysis of self-reported acts
Psychiatry Research
(1999) - et al.
Driving anger, emotional and instrumental aggressiveness, and impulsiveness in the prediction of aggressive and transgressive driving
Accident Analysis and Prevention
(2013) - et al.
Personality correlates of aggressive and non-aggressive antisocial behavior
Personality and Individual Differences
(2008) - et al.
Externalizing behavior, the UPPS-P impulsive behavior scale and reward and punishment sensitivity
Personality and Individual Differences
(2013) - et al.
Eysenck’s personality dimensions as predictors of male adolescent trajectories of physical aggression, theft and vandalism
Personality and Individual Differences
(2006) - et al.
Personality traits associated with problematic and non-problematic massively multiplayer online role playing game use
Personality and Individual Differences
(2012) - et al.
Tryptophan via serotonin/kynurenine pathways abnormalities in a large cohort of aggressive inmates: Markers for aggression
Progress in Neuro-Psychopharmacology & Biological Psychiatry
(2016) - et al.
Risky and aggressive driving in young adults: Personality matters
Accident Analysis and Prevention
(2011) - et al.
Impulsivity and internalizing disorders in childhood
Psychiatry Research
(2011)
Trait aggression and trait impulsivity are not related to frontal cortex 5-HT2A receptor binding in healthy individuals
Psychiatry Research: Neuroimaging
The influence of dysfunctional impulsivity and alexithymia on aggressive behavior of psychiatric patients
Psychiatry Research
Trait aggression is related to anger-modulated deficits in response inhibition
Journal of Research in Personality
How impulsivity and intelligence are related to different forms of aggression
Personality and Individual Differences
Dispositional impulsivity in normal and abnormal samples
Journal of Psychiatric Research
Impulsivity, aggressiveness, and DSM-IV personality disorders
Psychiatry Research
Predicting borderline and antisocial personality disorder features in nonclinical subjects using measures of impulsivity and aggressiveness
Psychiatry Research
Prefrontal regional correlates of self-control in male psychiatric patients: Impulsivity facets and aggression
Psychiatry Research: Neuroimaging
A multivariate approach to aggression and the orbital frontal cortex in psychiatric patients
Psychiatry Research: Neuroimaging
Revealing the nuanced associations between facets of trait impulsivity and reactive and proactive aggression
Personality and Individual Differences
Aggressive behavior among military veterans in substance use disorder treatment: The roles of posttraumatic stress and impulsivity
Journal of Substance Abuse Treatment
Does choice of measure matter? Assessing the similarities and differences among self-control scales
Journal of Criminal Justice
Aggression on the road: Relationships between dysfunctional impulsivity, forgiveness, negative emotions, and aggressive driving
Transportation Research Part F: Traffic Psychology and Behaviour
Distinctive profiles of traits predisposing to violence in schizophrenia and in the general population
Schizophrenia Research
Impulsivity traits as correlates of antisocial behaviour in adolescents
Personality and Individual Differences
Psychometric characteristic of the Italian version of the temperament and character inventory—Revised, personality, psychopathology, and attachment styles
Comprehensive Psychiatry
A test of the four-factor model of impulsivity-related traits
Personality and Individual Differences
Personality correlates of pathological gambling derived from Big Three and Big Five personality models
Psychiatry Research
Inhibitory control and trait aggression: Neural and behavioral insights using the emotional stop signal task
NeuroImage
Negative urgency and reward/punishment sensitivity in intermittent explosive disorder
Journal of Affective Disorders
The impact of craving and impulsivity on aggression in detoxified cocaine-dependent patients
Journal of Substance Abuse Treatment
Personality and laboratory-based aggression: Comparing the predictive power of the Five-Factor Model, BIS/BAS, and impulsivity across context
Journal of Research in Personality
Etiology of the impulsivity/aggression relationship: Genes or environment?
Psychiatry Research
Pathways to crime and risk factors among Arab female adolescents in Israel
Children and Youth Services Review
Salivary cortisol, personality, and aggressive behavior in adolescent boys: A 5-year longitudinal study
Journal of the American Academy of Child & Adolescent Psychiatry
Impulsivity, gender, and response to fenfluramine challenge in borderline personality disorder
Psychiatry Research
Irritability and impulsiveness: Relationship to self-reported impulsive aggression
Personality and Individual Differences
Association of low-activity MAOA allelic variants with violent crime in incarcerated offenders
Journal of Psychiatric Research
Alexithymia, emotion dysregulation, impulsivity and aggression: A multiple mediation model
Psychiatry Research
Aggression and inhibition deficits, the role of functional and dysfunctional impulsivity
Personality and Individual Differences
Prenatal testosterone and personality: Increasing the specificity of trait assessment to detect consistent associations with digit ratio (2D:4D)
Journal of Research in Personality
Anxiety, depression, impulsivity and substance misuse in violent and non-violent adolescent boys in detention in China
Psychiatry Research
Predicting delinquency by self-reported impulsivity in adolescents in Ghana
Criminal Behaviour and Mental Health
Assessing indirect aggression in aggressors and targets: Spanish adaptation of the Indirect Aggression Scales
Psicothema
Relations between traumatic stress, dimensions of impulsivity, and reactive and appetitive aggression in individuals with refugee status
Psychological Trauma: Theory, Research, Practice, and Policy
Differentiating forms and functions of aggression in emerging adults: Associations with hostile attribution biases and normative beliefs
Journal of Youth and Adolescence
Predicting inpatient aggression by self-reported impulsivity in forensic psychiatric patients
Criminal Behaviour and Mental Health
Pain, affect, and rumination: An experimental test of the emotional cascade theory in two undergraduate samples
Journal of Experimental Psychopathology
Adulthood personality correlates of childhood adversity
Frontiers in Psychology
Chinese version of impulsive/premeditated aggression scale: Validation and its psychometric properties
Journal of Aggression, Maltreatment & Trauma
Cited by (49)
Electrodermal lability and overt aggression: Testing the trait-like nature of skin conductance responses
2024, Personality and Individual DifferencesA multi-sample investigation of psychological flexibility processes and reactive and proactive functions of aggression
2024, Journal of Contextual Behavioral ScienceCognitive control moderates the maturation of relational aggression in adults
2023, Personality and Individual DifferencesFacet-level impulsivity and proactive and reactive relational aggression
2023, Personality and Individual Differences
- ☆
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
- 2
⁎ = Study was included in the meta-analysis