Measures of aggression
Psychlopedia -- Measures and manipulations -- Implicit measures -- Measures of aggression
Jump to the comments Section
Researchers often need to assess the level of aggression that individuals exhibit-defined as behavior with intent to harm a person (e.g., Baron & Richardson, 1994; Berkowitz, 1993; Geen, 1990). For example, Denzler, Forster, and Liberman (2009) attempted to ascertain whether stabbing a voodoo doll, representing a person who had acted offensively, curbs aggressive behavior. A variety of procedures have been developed to gauge aggressive behavior.
Aggression is difficult to measure, however. First, individuals often deny or conceal their aggressive behavior (e.g., James, McIntyre, Glisson, Green, Patton, LeBreton, et al., 2005). Second, aggressive acts are often unethical or consequential. Hence, researchers need to apply techniques that circumvent these complications.
Classifications of aggressive behavior
Although many forms of aggression have been differentiated, researchers often divide these acts into two main categories (e.g., Anderson & Carnagey, 2004): sometimes known as affective and predatory. Affective aggression, also called reactive, defensive, and hostile, refers to aggressive behavior that is elicited by a sense of threat, even fear. This form of aggression is impulsive, devoid of careful planning. In contrast, predatory aggression, also called instrumental, premeditated, and proactive, is more deliberate and calculated.
Indeed, as evidence indicates, these two forms of aggression tend to inhibit one another. Specifically, the medial hypothalamus, when coupled with the periaqueductal gray, in the midbrain, tends to mediate affective aggression. In contrast, the lateral hypothalamus tends to mediate predatory aggression. These two regions are connected by GABBA neurons, which inhibit one another (for a review, see Siegel & Victoroff, 2009).
Specifically, aggression is partly ascribed to the limbic system, assumed to include the hippocampus, amygdala, nucleus accumbens, ventral striatum, septal area, and segments of both the anterior cingulate and prefrontal cortex. The limbic system projects neurons to the hypothalamus and periaqueductal gray, in the midbrain. The limbic system moderates activity in the hypothalamus and periaqueductal gray; if impaired, these regions are not moderated effectively and aggression often ensues.
Dispensation of aversive stimuli
Perhaps the most prevalent approach to gauge aggressive behavior or inclinations is to offer participants an opportunity to dispense aversive stimuli or punishments to someone else. Traditionally, participants are granted an opportunity to dispense electric shocks (e.g., Berkowitz, 1964; Hammock, G., & Richardson, 1992), to hit someone with a foam sword (e.g., Diener, 1976; Diener, Dineen, Endresen, Beaman, & Fraser, 1975), or to ask someone to consume hot sauce (e.g., Lieberman, Solomon, Greenberg, & McGregor, 1999; McGregor, Lieberman, Greenberg, Solomon, Arndt, Simon, & Pyszczynski, 1998; Warburton, Williams, & Cairns, 2006).
To illustrate the applicant of hot sauce, in the study conducted by McGregor, Lieberman, Greenberg, Solomon, Arndt, Simon, and Pyszczynski (1998), individuals read an essay, written by another participant. Later, they were granted to opportunity to allocate a particular quantity of very spicy, hot sauce to the author-and informed this person does not like spicy food.
In particular, they were informed they were participating in a study, designed to assess the association between personality and food preferences. They were informed they will be tasting specific foods. They were also informed they need to allocate various foods to other participants; this process was putatively included to ensure experimenters are oblivious to which foods each participant receives.
Next, they were informed they would need to rate the dry foods and allocate the spicy foods to someone else. Specifically, they rated a dry cracker on a nine point scale. Then, participants were told to prepare a sample of hot sauce for someone else to consume. They could choose the quantity themselves, but were also informed the other person must consume the entire quantity. Furthermore, they were told they will be allocating this sauce to the person who wrote the essay they read earlier.
Participants allocated more hot sauce if this author wrote an essay that diverges from their own political perspective. This tendency was especially pronounced after participants had contemplated their own mortality-a manipulation that amplifies the rejection of contradictory opinions.
Many other studies have also validated this technique. Individuals tend to allocate more hot cause in similar contexts after they are exposed to guns (see Klinesmith, Kasser, & McAndrew, 2006)
The Taylor paradigm is sometimes used to assess aggression. A typical example was reported by Denson, Capper, Oaten, Friese, and Schofield (2011). First, participants were provoked by another person. That is, participants prepared and presented a speech to someone during a contrived videoconference and received unfavourable feedback. Later, in the context of a game with this person, participants were granted an opportunity to emit a loud blast to that individual. In particular, as soon as a stimulus appeared on the screen, both individuals were supposedly granted the opportunity to press a button as rapidly as possible. The person who did not win would receive a loud noise. Participants were permitted to set the intensity and duration of this noise. Loud, long noises represent an aggressive reaction. Self control was shown to curb aggression, as gauged by this measure.
Some studies have merely asked to individuals to choose which pictures another person should be asked to observe. To illustrate, in one of their studies, Denzler, Forster, and Liberman (2009), participants were asked to assist the experimenter, by choosing which 10 of 30 photographs should be utilized in a forthcoming study. Some of the pictures were negative in tone, and other pictures were positive in tone. The selection of negative photographs was assumed to reflect aggressive tendencies. That is, as pilot tests confirmed, participants recognize that such pictures will evoke discomfort in the viewer
Consistent with this proposition, participants who had read about someone who had cheated were more inclined to choose negative pictures, as rated by another set of individuals. This tendency to select negative pictures diminished, however, if these individuals could firstly stab a voodoo doll--which purportedly diminishes their aggressive tendencies.
Accessibility of aggression
Denzler, Forster, and Liberman (2009), in one of their studies, determined the accessibility of words that relate to aggression. If individuals are more inclined to act aggressively, words that coincide with aggressive actions, such as fist, are likely to be accessible.
In particular, participants completed a lexical decision task. That is, a series of items were presented, and participants had to decide whether or not the item was a word. Overall, 14 of the items were non-words; 7 of the items were words that relate to aggression, such as FIST; and 7 of the items were words that were unrelated to aggression, such as STOVE (for words, see Mussweiler & Forster, 2000). Rapid recognition of words related to aggression, compared to recognition of words unrelated to aggression, was regarded as a measure of aggressive inclinations.
As hypothesized, procedures that thwarted or fulfilled goals relating to aggression did reduce accessibility of aggressive words. For example, stabbing a voodoo doll, representing a person who had acted inappropriately, did indeed prolong reactions times to recognize aggressive words. These findings partly validate this approach.
Word completion tasks
A word completion task has been used to assess the accessibilit of aggression or hostility. In a study conducted by DeWall and Bushman (2009), participants received 98 word fragments. Their task was to ascertain the words these fragments represent. Furthermore, 49 of these fragments represent either an aggressive word or another word; ki - - , for example, might correspond to kill or kiss. The other 49 fragments did not represent aggressive words.
If participants had previously been exposed to the concept of hot temperatures, which DeWall and Bushman (2009) maintain should evoke aggression, they were more likely to specify aggressive words when they completed the fragments. This finding is consistent with the proposition that such answers represent aggressive inclinations.
Biases that epitomize aggression
James, McIntyre, Glisson, Green, Patton, LeBreton, et al. (2005) developed a procedure, called the conditional reasoning test, that can implicitly gauge cognitions and beliefs that predict aggressive behavior (see also James, 1998; James & Mazerolle, 2002; James, McIntyre, Glisson, Bowler, & Mitchell, 2004). In particular, participants read scenarios, such as an extract about the car manufacturing industry. The scenarios are followed by four statements. Participants must select the statement that is demonstrated or implied in the scenario.
Embedded within these choices are assumptions that are not implied by the scenario but often embraced by individuals who act aggressively. One of the alternatives might imply that car manufacturers deliberately ensure their vehicles are not durable, ultimately to increase sales. Aggressive individuals, putatively, embrace this cynical assumption to justify their aggression. Hence, aggressive participants will often choose this alternative.
Indeed, according to James, McIntyre, Glisson, Green, Patton, LeBreton, et al. (2005), aggressive individuals tend to espouse a variety of distorted assumptions, all intended to justify their behavior. For example, aggressive participants more likely to believe that other individuals are motivated to harm, not help. For instance, they assume surveillance systems are merely an excuse to bother employees.
Similarly, they perceive social interactions as usually a process of attempting to establish who is dominant rather than submissive. They feel that managers often fear strong employees, for instance. Third, they feel that retribution is more important than striving to maintain harmony. To illustrate, they feel that peacetime generals are not appropriate during wars. Fourth, they feel that powerful individuals strive to impose harm, and conceptualize themselves as victims of injustice. They assume, for example, that hiring a lawyer during a divorce implies the other party will act with hostility. Fifth, they feel that social customs tend to restrict autonomy and independence; for example, bonuses are perceived as a means to control employees. Finally, they like to perceive their victims as immoral and evil.
The test that is designed to predict aggression comprises 22 of these items. Three traditional items, which assess inductive reasoning, are included to enhance face validity. Several studies have shown that indeed performance on this test does predict counterproductive or aggressive behavior (e.g., James & Mazerolle, 2002; James & McIntyre, 2000; James, McIntyre, Glisson, Green, Patton, LeBreton, et al., 2005), and effect sizes are reasonably large, with correlations exceeding .40. Estimates of internal consistency also exceed .74.
Interesting, performance on this test is unrelated to self report measures of aggression. Furthermore, performance on this test predicts counterproductive or aggressive behavior more effectively than self report measures (James & Mazerolle, 2002; James & McIntyre, 2000; James, McIntyre, Glisson, Green, Patton, LeBreton, et al., 2005). Finally, performance on this test is unrelated to cognitive ability.
Self report measures
Researchers have also developed some self report measures of aggression or related traits and behaviors. Some of these measures assess the extent to which respondents feel they exhibit aggressive behavior. Other measures, instead, explore the degree to which respondents perceive other individuals as hostile-a perception that often provokes aggressive behavior. Finally, some measures assess the belief of respondents, such as whether retribution is appropriate.
Buss and Perry (1992) developed a scale that assesses the extent to which individuals exhibit aggressive behavior. This measure of trait aggressiveness comprises three subscales. The first subscale represents the degree to which respondents exhibit anger. A typical item is "When frustrated, I let my irritation show". The second subscale concerns the extent to which respondents express aggressive remarks, called verbal aggression. A sample item is "When people annoy me, I may tell them what I think of them". The third subscale refers to the degree to which participants perceive other individuals as hostile. An example is "When people are especially nice, I wonder what they want".
Some researchers collapse the three subscales to derive an overall measure of trait aggression. Keller,, Hurst, and Uskul (2009), for example, showed this overall measure generates an alpha reliability of .85. A state variant of this questionnaire is also administered occasionally (see Farrar & Krcmar, 2006; Keller, Hurst, & Uskul, 2009).
Cynical hostility represent the extent to which participants assume the motivations of other individuals are competitive and uncaring. Cook and Medley (1954) developed a measure of cynical hostility, with items such as "Most people will use somewhat unfair means to fain profit or an advantage rather than to lose it". Keller, Hurst, and Uskul (2009) showed that alpha reliability can range from .69 to .73.
Several researchers have developed scales that assess the extent to which individuals feel they should reciprocate the behavior of someone else (e.g., Eisenberger, Lynch, Aselage, & Rohdieck, 2004; Perugini, Gallucci, Presaghi, & Ercolani, 2003). These scales usually comprise a mixture of reciprocity in response to favorable behaviors, such as "If someone does me a favor, I feel obligated to repay them in some way" (Eisenberger, Lynch, Aselage, & Rohdieck, 2004), as well as reciprocity in response to unfavorable behaviors, such as "If someone dislikes you, you should dislike them" (Eisenberger, Lynch, Aselage, & Rohdieck, 2004).
Stankov, Saucier, and Knezevic (2010) developed a measure that represents the extent to which individuals espouse an extremist mindset. In short, this measure comprises three subscales, each consisting of eight items. The first subscale, called proviolence, represents the extent to which individuals are wiling to engage in violence to solve social problems. A typical item is "Was is the beginning of salvation" or "Killing is justified when it is an act of revenge". The second subscale, called vile world, represents negative beliefs about the world and society, typified by items like "The world is headed for destruction" or "The present day world is vile and miserable". The final subscale, called divine power, relates to the extent to which individuals refer to a sacred entity to justify violence. A sample item is "Martyrdom is the act of a true believer in the cause, not an act of terrorism" or "At a critical moment, a divine power will step in to help our people".
Perhaps the most impressive feature of this scale is the procedures that were undertaken to compilate the initial set of items. First, information about militant attitudes was derived from academic sources on military extremism as well as relevant psychological, historical, and literary articles. Twenty main themes emerged; 122 items were developed to represent these themes. Second, the texts and documents of militant groups were subjected to content analyses; 16 themes emerged and 106 items were developed. Third, the frequency of words and themes in texts written by military extremists were compared to the frequency of words and themes in texts written by other political parties in that nation. Words that differed between these two sets of sources were then converted to items; 133 items were derived from this process, called a linguistic approach.
An initial pilot study was conducted in Australia, Serbia, and USA; the number of items was reduced from 361 to 56. These items were then completed by 2424 participants, in Anglo, Latin American, South East Asian, and Confusion nations. In addition, these individuals completed questionnaires that assess personality, social attitudes, and values. The measure of social attitudes determined the extent to which participants value tradition, self interest, civic ideals, and spirituality.
Exploratory factor analysis identified three factors; after redundancies were eliminated, only 24 items remained. Vile world and divine power were positively correlated, with r = .35. Proviolence was not correlated with the other factors. Item response theory confirmed this factor structure. Specifically, each of the three factors were conceptualized as distinct scales, conforming to a 2 parameter logistic model. Fit indices were sound; correlations between each item and the slope parameters almost always exceeded.4. This structure was confirmed in all the nations that were tested, except in Belarus, in which the proviolence factor did not emerge.
The proviolence factor was negatively associated with extraversion, resilience, conscientiousness, agreeableness, originality, and honesty but positively related to psychoticism. Furthermore, individuals who reported proviolence espoused attitudes that justify self interest, such as materialism and hedonism. The vile world factor was related to psychoticism and social cynicism, which represents a negative perception of human nature. The divine power subfactor was related to positive attitudes towards tradition and convention.
Similarly, Stankov, Higgins, Saucier, and Knezevic (2010) developed a similar measure that assesses the degree to which individuals espouse militant, extremist attitudes. These items were derived from the linguistic approach only. The scale comprises three distinct factors. The first factor, comprising eight items, represents the extent to which individuals justify and champion violence. A typical item of this subscale, called the War factor, is "We should respond to terror with terror".
The second factor also comprises eight items. This subscale, called the God factor, represents the degree to which individuals embrace violence in the name of god. A sample item is "I wish to go to war for the sake of god".
The third subscale, also comprising eight items and called the West factor, represents the extent to which individuals blame Western nations for the problems that pervade society. A typical item is "It has become clear that the West has an unspeakable hatred of Islam".
Desensitization to violence
Engelhardt, Bartholow, Kerr, and Bushman (2011) showed that P300, a component of the event related potential, can be used to gauge desensitization to violence, ultimately increasing the likelihood of aggression. In particular, after individuals are exposed to some event, such as the picture of a gun, brain activity momentarily changes. These changes in brain activity influence the electrical properties of the brain. In particular, after some event, the voltage of brain activity rises and falls several times over the period of half a second or so. This pattern of electrical activity is called an event related potential and can be measured with an EEG.
A typical event related potential entails three rises in voltage coupled with two declines. The third rise is called the P3 or sometimes the P300, because this shift tends to begin about 300 ms after the event and ends about 300 ms later. In general, the P300 is especially detectable with electrodes that are placed over the parietal lobe.
The cognitive processes of individuals affect the amplitude or magnitude of this P300. For example, if the event, such as the picture, is common, the P300 is not usually as pronounced. If the event is infrequent, the P300 is usually greater in amplitude. Furthermore, if this event is aversive, and provokes withdrawal or avoidance, the P300 is particularly pronounced.
Specifically, according to Nieuwenhuis, Aston-Jones, and Cohen (2005), the P300 represents the activity of the brain regions that are activated by norepinephrine. Norepinephrine is primarily produced by the locus-coeruleus, a region in the brain stem, and activates regions in the brain that correspond to arousal including the hypothalamus and amygdala.
Given the P300 represents level of aversion, Engelhardt, Bartholow, Kerr, and Bushman (2011) argued that perhaps this component could represent sensitivity to aggressive events. In particular, if individuals are desensitized to aggression, perhaps because they have been exposed to many violent games in the past, their P300 to aggressive pictures should be diminished. That is, they should perceive aggressive pictures as common instead of infrequent. Furthermore, these aggressive pictures should not elicit a motivation to withdraw.
Engelhardt, Bartholow, Kerr, and Bushman (2011) provided evidence that confirms this possibility. In this study, participants played video games for 25 minutes. Some participants were exposed only to violent games, including Call of Duty, Hitman, Killzone, and Grand Theft Auto. Other participants were exposed to games that are not violent, including Jak and Daxter, MVP Baseball 2004, Tony Hawk's Pro Skater 4, and Sonic Plus Mega Collection. Next, they were exposed to a series of aggressive pictures, such as a person inserting a gun in the mouth of someone else, embedded within a set of other pictures. The P300 that was evoked by the aggressive pictures was measured. Finally, participants completed a game in which they could increase the noise that someone else would experience, representing an implicit measure of aggression.
If participants had been exposed to the violent video games, their P300 to aggressive pictures was curbed. They did not perceive these pictures as especially infrequent or aversive, representing a form of desensitization. Furthermore, this desensitization increased the likelihood they would subsequently act aggressively towards someone else, by increasing the noise that was emitted.
Anderson, C. A., & Carnagey, N. L. (2004). Violent evil and the general aggression model. In A. G. Miller (Ed.), The social psychology of good and evil (pp. 168-192). New York: The Guilford Press.
Baron, R. A., & Richardson, D. R. (1994). Human aggression (2nd ed.). New York: Plenum Press.
Berkowitz, L. (1964). Aggressive cues in aggressive behavior and hostility catharsis. Psychological Review, 71, 104-122.
Berkowitz, L. (1993). Aggression: Its causes, consequences, and control. Philadelphia: Temple University Press.
Buss, A. H., & Perry, M. (1992). The aggression questionnaire. Journal of Personality and Social Psychology, 63, 452-459.
Cook, W. W., & Medley, D. M. (1954). Proposed hostility and pharisaic-virtue scales for the MMPI. Journal of Applied Psychology, 38, 414-418.
Denson, T. F., Capper, M. M., Oaten, M., Friese, M., & Schofield, T. P. (2011). Self-control training decreases aggression in response to provocation in aggressive individuals. Journal of Research in Personality, 45, 252-256.
Denzler, M., Forster, J., & Liberman, N. (2009). How goal-fulfillment decreases aggression. Journal of Experimental Social Psychology, 45, 90-100.
DeWall, C. N., & Bushman, B. J. (2009). Hot under the collar in a lukewarm environment: Words associated with hot temperature increase aggressive thoughts and hostile perceptions. Journal of Experimental Social Psychology, 45, 1045-1047.
Diener, E. (1976). Effects of prior destructive behavior, anonymity, and group presence on deindividuation and aggression. Journal of Personality and Social Psychology, 33, 497-507.
Diener, E., Dineen, J., Endresen, K., Beaman, A. L., & Fraser, S. C. (1975). Effects of altered responsibility, cognitive set, and modeling on physical aggression and deindividuation. Journal of Personality and Social Psychology, 31, 328-337.
Eisenberger, R., Lynch, P., Aselage, J., & Rohdieck, S. (2004). Who takes the most revenge? Individual differences in negative reciprocity norm endorsement. Personality and Social Psychology Bulletin, 30, 787-799.
Engelhardt, C. R., Bartholow, B. D., Kerr, G. T., & Bushman, B. J. (2011). This is your brain on violent video games: Neural desensitization to violence predicts increased aggression following violent video game exposure. Journal of Experimental Social Psychology, 47, 1033-1036. doi:10.1016/j.jesp.2011.03.027
Farrar, K., & Krcmar, M. (2006). Measuring state and trait aggression: A short, cautionary tale. Media Psychology, 8, 127-138.
Geen, R. G. (1990). Human aggression-Mapping social psychology. Great Britain: Open University Press.
Hammock, G., & Richardson, D. (1992). Predictors of aggressive behavior. Aggressive Behavior, 18, 219-229.
James, L. R. (1998). Measurement of personality via conditional reasoning. Organizational Research Methods, 1, 131-163.
James, L. R., & McIntyre, M. D. (2000). Conditional Reasoning Test of Aggression test manual. San Antonio, TX: Psychological Corporation.
James, L. R., McIntyre, M. D., Glisson, C. A., Bowler, J. L., & Mitchell, T. R. (2004). The conditional reasoning measurement system for aggression: An overview. Human Performance, 17, 271-295.
James, L. R., McIntyre, M. D., Glisson, C. A., Green, P. D., Patton, T. W., LeBreton, J. M., et al. (2005). Conditional reasoning: An efficient, indirect method for assessing implicit cognitive readiness to aggress. Organizational Research Methods, 8, 69-99.
Keller,J., Hurst, M., & Uskul, A. (2009). Prevention-focused self-regulation and aggressiveness. Journal of Research in Personality, 42, 800-820.
Klinesmith, J., Kasser, T., & McAndrew, F. T. (2006). Guns, testosterone, and aggression: An experimental test of a mediational hypothesis. Psychological Science, 17, 568-571.
Lieberman, J. D., Solomon, S. Greenberg, J. & McGregor, H. A. (1999). A hot new way to measure aggression: Hot sauce allocation, Aggressive Behavior, 25, 331-348.
McGregor, H. A., Lieberman, J. D., Greenberg, J., Solomon, S., Arndt, J., Simon, L., & Pyszczynski, T. (1998). Terror management and aggression: Evidence that mortality salience motivates aggression against worldview-threatening others. Journal of Personality and Social Psychology, 74, 590-605.
Mussweiler, T. & Forster, J. (2000). The sex-aggression link: A perception-behavior dissociation. Journal of Personality and Social Psychology, 79, 507-520.
Nieuwenhuis, S., Aston-Jones, G., & Cohen, J. D. (2005). Decision making, the P3, and the locus coeruleus-norepinephrine system. Psychological Bulletin, 131, 510-532.
Perugini, M., Gallucci, M., Presaghi, F., & Ercolani, A. P. (2003). The personal norm of reciprocity. European Journal of Personality, 17, 251-283.
Siegel, A., & Victoroff, J. (2009). Understanding human aggression: New insights from neuroscience. International Journal of Law and Psychiatry, 32, 209-215.
Stankov, L., Higgins, D., Saucier, G., & Knezevic, G. (2010). Contemporary militant extremism: A linguistic approach to scale development. Psychological Assessment, 22, 246-258.
Stankov, L., Saucier, G., & Knezevic, G. (2010). Militant extremist mind-set: Proviolence, vile world, and divine power. Psychological Assessment, 22, 70-86.
Warburton, W. A., Williams, K. D., & Cairns, D. R. (2006). When ostracism leads to aggression: The moderating effects of control deprivation. Journal of Experimental Social Psychology, 42, 213-220.
Created by Dr Simon Moss on 05/01/2009