Categorization and Prediction of Crimes of Passion Based on Attitudes Toward Violence

Abstract

The present study explored implicit and explicit attitudes toward violence in crimes of passion. Criminals (n = 96) who had perpetrated crimes of passion and students (n = 100) participated in this study. Explicit attitudes toward violence were evaluated using the Abnormal Personality Risk Inventory (APRI), and implicit attitude toward violence was evaluated using the Implicit Association Test (IAT). Results indicated that APRI scores of the perpetrators were significantly higher than that of the control group (p < .05), suggesting that explicit attitudes toward violence could discriminate between the criminals and the control group. There was a significant IAT effect demonstrating a negative implicit attitude toward violence in both the control group and in the criminals (n = 68); whereas there was a significant IAT effect manifesting a positive implicit attitude toward violence in the criminals (n = 16) only. These results suggest that combining explicit and implicit attitudes could provide an empirical classification of crimes of passion.

Keywords

violence crimes of passion explicit attitude implicit attitude

Background

Violent behavior is a public health issue in every country (World Health Organization [WHO], 2010). At the end of the 20th century, it was found that 3.0% of all crimes were violent crimes in China (W. Zhang, Liu, & Gan, 2005), and notably, one third of the violent crimes were crimes of passion (Z. X. Chen, 2004). Violent crimes, especially crimes of passion, are a serious problem in terms of their impact on victims, the quality of life for perpetrators, and the financial burden on prison services, the health sector, and society in general (Harvey, Williams, & Donnelly, 2012; Ross, Quayle, Newman, & Tansey, 2013). In an attempt to maintain social stability and reduce violent crimes, especially crimes of passion, crime prevention has received increased attention of public health services (Carolyn, Timothy, & Terence, 2005).

Crimes of passions are impulsive and aggressive acts executed in a highly stressful, affective state, which are triggered by trivial quarrels or incidents. Crimes of passion are done on the spur of the moment, which are unique crimes in which the offender is unlikely to have a history of previous criminal activity and likely to never offend again (He, 2000; Boa, Lecture). Crimes of passion are perpetrated by impulsive criminals and by those who have been provoked in some way (He, 2000). Impulsive crimes of passion refer to offenders committing violent behaviors against a victim, at times including innocent bystanders, often just because improper words or deeds of the victim induced the perpetrator’s fury, or passion. Provoked crimes of passion refer to criminal acts against victims, in which victims have bullied the perpetrator, when bulling exceeds the perpetrator’s tolerance, and criminal acts are committed against the victims, including bystanders (Jiang, 2003). Therefore, crimes of passion are serious violent crimes with various subtypes, having different features. Distinguishing these types of criminals can predict the risk of violence (Steadman et al., 2000), and even prevent the occurrence of violent crimes. Most existing evidence on the prediction of violence has focused on implicit and explicit attitudes toward violence (Sandstrom & Jordan, 2008).

Attitudes are a core construct of social psychology defined as evaluations of psychological objects, such as people, things, or behaviors (Ajzen, 2001; Eagly & Chaiken, 1993, 2007; Fazio, 2007). Based on this definition, attitudes toward violence can be considered evaluations of violence. Attitude toward violence plays an important role in violence risk assessment (Anderson, Benjamin, Wood, & Bonacci, 2006). Previous studies have demonstrated that positive attitudes toward violence predict diverse violent behaviors, such as bullying at school (Eliot & Cornell, 2009; McConville & Cornell, 2003), and domestic violence in college students (Fincham, Cui, Braithwaite, & Pasley, 2008). Moreover, Wilson and Lindsey’s dual attitudes model proposes that one can simultaneously have two different attitudes toward a single object: an explicit and an implicit attitude (Anna, Olson, & Fazio, 2006; Wilson, Lindsey, & Schooler, 2000).

Explicit attitudes are defined as deliberative propositional reasoning about a psychological object (Mills, Kroner, & Forth, 2002). It exists in the conscious state (Banse & Fischer, 2002; Sandstrom & Jordan, 2008) and can be evaluated by self-report measures (Loza & Loza-Fanous, 2000; Loza, MacTavish, & Loza-Fanous, 2007; Monahan et al., 2001; Suter, Pihet, Ridder, Zimmermann, & Stephan, 2014), such as the Abnormal Personality Risk Inventory (APRI; Wang, Zhang, Guan, Li, & Liu, 2011). Wang and her colleagues (2011) have demonstrated that APRI can well differentiate violent criminals from normal control. Self-reported explicit measures have been used for assessing neutral topics such as consumer preferences (Fazio, Jackson, Dunton, & Williams, 1995); however, self-reports are likely to be encumbered by distortion and inaccuracies (Evans, 2008; Nisbett & Wilson, 1977; Wilson & Dunn, 2004). Research suggests that self-reports might provide a distorted image when socially desirable topics are involved, at least in the case of adult offenders (Kampfe, Penzhorn, Schikora, Dunzl, & Schneidenbach, 2009).

Implicit attitudes manifest as actions, or judgments that are under the control of automatically activated evaluation, without the performer’s awareness of the causation (Greenwald, McGhee, & Schwartz, 1998). They are maintained in the subconscious and can only be assessed by indirect methods (Anna et al., 2006), such as the Implicit Association Test (IAT; Fazio, Sambonmatsu, Powell, & Kardes, 1986). Implicit attitudes predict spontaneous behaviors (Greenwald, Poehlman, Uhlmann, & Banaji, 2009) and socially sensitive topics, such as violent attitudes (Fazio et al., 1995; Ramírez & Andreu, 2006). Recent studies suggest that implicit attitudes play a key role in violent behaviors (Suter et al., 2014). It has been confirmed that the IAT can predict violent behavior (Gray, MacCulloch, Smith, Morris, & Snowden, 2003; Polaschek, Bell, Calvert, & Takarangi, 2010; Snowden, Gray, Smith, Morris, & MacCulloch, 2004; Suter et al., 2014). The advantage of the IAT in contrast to self-report questionnaires is its demonstrated resistance to “faking being good” (Roura et al., 2009; Steffens, 2004). These findings support that the IAT is a promising assessment tool to complement traditional instruments.

There is scattered evidence that both implicit and explicit attitudes toward violence predict violent behavior (Banse & Fischer, 2002; Sandstrom & Jordan, 2008). However, no study has explored violent attitudes in crimes of passion, despite the potential key role that explicit and implicit attitudes toward violence might play in the occurrence of violent behavior. Moreover, no study has classified criminals based on violent attitudes. The current study examined violent attitude traits and discussed distinct features of impulsive and provoked crimes of passion. In the study, explicit attitude toward violence in perpetrators was assessed by APRI; whereas implicit attitudes toward violence in perpetrators were evaluated by IAT. Positive predictive value on the classification of crimes of passion was validated based on criminal files.

Method

Participants

Male criminals who committed crimes of passion imprisoned in the Shaanxi Province of China were included in the criminal group. All were violent criminals sentenced to 12 or more years in jail (n = 96, age range 23 to 30 years, M age of 23.82 ± 1.41 years, M years of education 7.31 ± 1.25 years, range of years 6 to 10 years). Based on the definition and characteristics of this type of criminal by Z. B. Zhang (1998), we categorized the participants using information in their criminal files. The inclusion standards were (a) first-time offenders, (b) not a premeditated crime, (c) were enraged as a result of external factors with a sudden outbreak of intense emotions, and (d) acted violently toward the victim when the crime was committed. Of these, 11 participants had chosen the same response on all items of APRI, and four did not complete the scale. Therefore, data of 81 participants were considered valid (valid return rate of 89.01%).

A control group was designed to better compare implicit attitudes toward violence in the perpetrators. Participants in the control group were young men recruited from an advertisement in the Shaanxi Province of China (n = 100, age range = 21-28 years, M age = 22.09 ± 1.84 years, M years of education = 8.29 ± 1.03 years, range = 7-years). The inclusion standards for control group were (a) no criminal record, (b) no history of physical illnesses, or treatment for physical illness, and (c) no history of psychiatric or neurological illnesses, or psychiatric treatment. There was no significant difference in age (p > .05) and education (p > .05) between the criminal and the control groups. Two participants in the control group had chosen the same answer for all items of APRI. Therefore, data of 98 participants were considered valid in the control group (validity rate of 98.00%). All participants provided their written informed consent. The protocol for this research was approved by the ethics committee of the Fourth Military Medical University and the ethics committee of the Prison.

Materials

The APRI assesses eight risk factors in three domains and is known to have good reliability and validity (Xiao, 2007). In the present study, we used the Violence subscale (VIO) of the APRI, which includes 27 items for assessing offenders’ explicit attitude toward violence (see Appendix). The Cronbach’s α value of the scale is .81.

The materials used for the assessment of implicit attitudes included characteristic words and behavioral pictures. Ninety college students had previously appraised 200 words that described character traits on a 7-point Likert-type scale ranging from 1 (very negative) to 7 (very positive). Ten words describing negative personal characteristics (M = 1.91, SD = 0.08) and 10 words describing positive personal characteristics (M = 6.35, SD = 0.16) were selected. Fifty behavioral pictures were selected from the International Affective Picture System. These pictures were rated for the degree of violence by the same 90 college students described above, using a 7-point scale ranging between 7 (violent behavior) and 1 (nonviolent behavior). Then, 10 pictures each for nonviolent (M = 1.40, SD = 0.3) and violent behaviors (M = 6.18, SD = 0.39) were selected.

Procedure

The IAT measures automatic associations between target dimensions and attribute dimensions (Greenwald, Nosek, & Banaji, 2003). It is a computerized categorization task that uses response time for the computation of the dependent measure. In the current study, target dimensions referred to pictures of behaviors, such as beating, killing, chatting, and traveling, whereas attribute dimensions referred to personal characteristic words, such as cheerful, confident, and arrogant. We constructed two types of tasks: an initial combined discrimination task (nonviolent behavior/positive character or violent behavior/negative character) and a reversed combined discrimination task (nonviolent behavior/negative character or violent behavior/positive character). The computer randomly presented these two types of experimental materials. On the prompt shown on the screen, the participants pressed the “F” key or the “J” key on the keyboard to categorize the words or pictures. The task automatically continued onto the next discrimination task when the stimulus disappeared, or after 2,500 ms. Each IAT test included five basic parts. To ensure the participants’ response stability, initial and reversed combined discriminations were repeated in the actual test, so that there were seven parts in each test. The response times for the initial and reversed combined discrimination tasks were recorded (see details in Table 1).

Table 1.

Implicit Association Test Procedure.

	Task description	Task instruction
1	Target dimensions discrimination	Nonviolent^F/Violent^J
2	Attribute dimensions discrimination	Positive^F/Negative^J
3	Initial combined discrimination	Nonviolent^F/Positive^F, Violent^J/Negative^J
4	Initial combined discrimination	Nonviolent^F/Positive^F, Violent^J/Negative^J
5	Reversed target dimensions discrimination	Violent^F/Nonviolent^J
6	Reversed combined discrimination	Violent^F/Positive^F, Nonviolent^J/Negative^J
7	Reversed combined discrimination	Violent^F/Positive^F, Nonviolent^J/Negative^J

Note. Participants were instructed to assign each stimulus to one of two categories (in single categorization tasks, Blocks 1, 2, and 5) or one of four categories (in combined categorization tasks, Blocks 3, 4, 6, and 7). Participants used either the “F” key with the left index finger or the “J” key with the right index finger.

All the experimental materials were presented in the center of a computer monitor in picture format with a size of 7 cm × 7 cm. Words were in white Song font on a black background. The participants were 56 cm away from the computer with a visual angle of 7°. We counterbalanced the order of the tests using an ABBA scheme. Half of the participants took the IAT first; the other half of participants took the tests in the reverse order.

Data Analysis

Greenwald proposed new IAT data screening criteria in 2003 (Greenwald et al., 2003), accordingly: (a) If 10% of a participant’s response latencies are longer than 10,000 ms or shorter than 300 ms, the data for that participant are omitted from the analysis, (b) all error response latencies are to be replaced by the group mean plus a 600 ms “penalty,” (c) the first two tests in each experiment are not to be included in the data analysis, and (d) if the error rate of a participant exceeded 20%, the data from this participant are eliminated. We corrected the response latencies, or excluded the data from participants according to the above criteria, then calculated the average response latencies of the initial and reversed combined discrimination tasks for the remaining participants. In the criminal group of this study, four participants who had 20% of response latencies over 10,000 ms in the reversed combined discrimination test and eight participants with error rates more than 20% were omitted. In the control group, two participants who had response latencies over 10,000 ms and five participants with error rates more than 20% were omitted. Consequently, the data of 84 participants in the experimental group and 93 participants in the control group were analyzed (see Table 2).

Table 2.

Results of Explicit Attitude Toward Violence as a Diagnostic Test for Discrimination Crime of Passion From Control Group.

	Crime of passion	Control group	Totals
High score	32	19	51
Low score	49	79	128
Totals	81	98	179

Note. Positive predictive value in high score = 32 / 51 × 100% = 62.74%. Positive predictive value in low score = 79/128 × 100% = 61.72%.

Then IAT effect value was calculated by subtracting the response latency for the initial combined discrimination from reversed combined discrimination. The IAT effect value was then subjected to a logarithmic transformation and compared with the null value using a t test. The log transformation improves the symmetry of latency distributions by shrinking the upper tail and is thereby expected to improve the central tendency estimates (Greenwald et al., 2003).

The statistical analyses using the SPSS software package, Version 16.0 (SPSS Inc., the United States) were performed. Correlation and t tests were analyzed with the obtained data. Statistical significance was set as p < .05.

Results

APRI Results

The perpetrators’ VIO subscale score (M = 4.62, SD = 1.58) was not significantly different from that of the control group (M = 3.25, SD = 1.34), p > .05.

Results of Explicit Attitude Toward Violence as a Diagnostic Test for Discrimination Against the Criminals From Control Group

Based on criminal records, there were 81 criminals; the control group was 98. If the VIO score of the criminals were higher than that of the control group by two standard deviations, this was considered a high score. Calculating on this, if the VIO score higher than 5.93, the VIO score was considered a high score, whereas the opposite was considered a low score. A high score on VIO indicated more explicit attitudes toward violence. Approximately, 29.05% of the total study population scored high on the VIO scale (51 out of 179). As expected, the majority of high-scoring participants were the criminals (62.74%). The control population comprised a greater proportion of low VIO scores (79 out of 128, 61.72%) than in the criminal group (49 out of 128, 38.28%). It should be noted that the majority of the criminals (49 out of 81, 60.49%) actually received low VIO scores (see Table 2).

Results of Explicit Attitude Toward Violence as a Diagnostic Test for Classification of the Criminals

The criminals were divided into impulsive and provoked impassioned criminals on the basis of criminal files. The results showed that 63 were impulsive criminals, 47.62% of this group scored high (30 out of 63), whereas 52.38% scored low (33 out of 63). Eighteen participants were provoked criminals, 88.89% scored low (16 out of 18). This indicated that impulsive criminals comprised a greater proportion of low VIO scores. The positive predictive value in the high-scoring group was 93.75%, and the positive predictive value in the low score group was 32.65% (see Table 3).

Table 3.

Results of Explicit Attitude Toward Violence as a Diagnostic Test for Classification of Impulsive and Provoked Impassioned Criminals.

	Impulsive criminals	Provoked criminals	Totals
High scores	30	2	32
Low scores	33	16	49
Totals	63	18	81

Note. Positive predictive value in non-reversed group = 30 / 32 × 100% = 93.75%. Positive predictive value in reversed group = 16/49 × 100% = 32.65%.

Results of IAT

In current study, response times for initial and reversed combined discrimination (Blocks 3, 4, 6, and 7) were analyzed. In the control group, response latency for reversed combined discrimination was significantly longer than that for the initial combined discrimination. There was a significant IAT effect in the control group (p < .001). There was no significant IAT effect in the criminal group (p > .05; see Table 4).

Table 4.

Comparison of the IAT Results Between the Criminals and Control Groups (M ± SD).

	Initial combined discrimination (ms)	Reversed combined discrimination (ms)	IAT effect (ms)	t
Criminal group (n = 84)	1,061.17 ± 72.53	1,205.03 ± 69.28	143.86	1.746
Control group (n = 93)	749.26 ± 92.45	1,228.19 ± 117.63	478.93	6.077***

Note. IAT = Implicit Association Test.

***

p < .001.

We conducted a further analysis to determine why there was no significant difference in the criminal group compared with the control group. The response latency for the initial combined discrimination (1,005.28 ± 74.39 ms) in 68 of the criminals was significantly shorter than that for reversed combined discrimination (1,418.76 ± 92.04 ms), producing a significant IAT effect (M = 413.18 ms, p < .001). In the other 16 perpetrators, the response latency for the initial combined discrimination was (1,375.83 ± 104.12 ms) significantly longer than that for reversed combined discrimination (1,081.35 ± 94.84 ms), resulting in a significant IAT effect (M = −294.48 ms, p < .001; see Figure 1). These reversed effects could have masked the IAT effects, and therefore, impassioned criminals were further categorized into reversed and non-reversed groups.

Figure 1.

Response times in initial and reversed combination discrimination between two provoked impassioned criminal groups.

Results of Implicit Attitude Toward Violence as a Diagnostic Test for Discrimination Against Criminals From Control Group

Criminal records indicate that there were 84 crimes of passion; the control group was 93. The cross-tabulation showed that 16 criminals had reversed IAT. In contrast, 93 out of the 161 were in the control group. In the reversed group, the positive predictive value was 100%, and in the non-reversed group, the positive predictive value was 57.76% (see Table 5).

Table 5.

Results of Explicit Attitude Toward Violence as a Diagnostic Test for Discrimination Crime of Passion From Control Group.

	Crime of passion	Control group	Totals
Reversed	16	0	16
Non-reversed	68	93	161
Totals	84	93	177

Note. Positive predictive value in reversed = 16 / 16 × 100% = 100%. Positive predictive value in non-reversed = 93 / 161 × 100% = 57.76%.

Results of Implicit Attitude Toward Violence as a Diagnostic Test for Classification of Criminals

On the basis of criminal files, the criminals were divided into two groups, impulsive and provoked criminals. The results showed that 64 were impulsive criminals and 20 were provoked criminals. The cross-tabulation of criminal type and IAT direction showed that 63 out of 68 criminals with a non-reversed attitude toward violence were impulsive criminals. In contrast, 15 of the 16 criminals with a reversed attitude toward violence were provoked criminals. The percentage of impulsive criminals who were non-reversed (63 out of 64) was 98.44%, and that for provoked criminals who were reversed (15 out of 20) was 75%. In the non-reversed group, the positive predictive value was 92.64%, and in the reversed group, the positive predictive value was 93.75% (see Table 6).

Table 6.

Results of Implicit and Explicit Attitude Toward Violence as a Diagnostic Test for Classification of Provoked Impassioned Criminals.

	Impulsive criminals	Provoked criminals	Totals
Non-reversed	63	5	68
Reversed	1	15	16
Totals	64	20	84

Note. Positive predictive value in non-reversed group = 63 / 68 × 100% = 92.64%. Positive predictive value in reversed group = 15 / 16 × 100% = 93.75%.

Discussion

The consequences of violent behavior not only affect those directly involved, but also affect social security (C.-Y. Chen, Muggleton, Juan, Tzeng, & Hung, 2008; WHO, 2004), even resulting in additional crimes (Baxendale, Cross, & Johnston, 2012). The occurrence of violent acts is greatly affected by attitude toward violence (Carolyn et al., 2005; Fincham et al., 2008; Funk, Elliott, & Bechtoldt, 2003). In the present study, explicit and implicit attitudes toward violence of perpetrators were investigated. Some of the criminals had negative attitudes toward violence, whereas others had positive attitudes toward violence. These results provide additional support to the contention that the IAT procedure constitutes a promising assessment tool to complement explicit measures.

Explicit attitudes have been indicated as being predictive of violence based on self-reports (Monahan et al., 2001). To explore the effectiveness of explicit measures, the positive predictive value was calculated. The results of explicit attitudes toward violence as a diagnostic test for discriminating criminals from a control group showed that 32 criminals were differentiated from the control group. Most of the criminals were mixed with the control group. Some researchers have questioned the validity of self-report measures of violence, arguing that social desirability and self-presentational concerns produce inaccuracy (Houwer, Mocigemba, Spruyt, & Moors, 2009), and violent attitudes are a sensitive topic (Fazio et al., 1995; Ramírez & Andreu, 2006). Therefore, it is therefore concluded that explicit attitudes toward violence can screen a small number of criminals, but cannot differentiate subtypes of the perpetrators.

The IAT measures the existence of implicit attitudes and the association between the target and attributes (Anna et al., 2006). It is known that shorter the distance in neural networks, faster is the response latency and vice versa (Greenwald et al., 1998). In the control group, there was a relatively closer connection between the target and attribute when violent behavior was combined with negative attributes, and nonviolent behavior was combined with positive attributes than vice versa. This suggested that violent/negative information had a closer association and showed a negative implicit attitude toward violence. When we used the same experimental paradigm to measure implicit attitude toward violence of the criminals, results were quite different from the control. Most of the criminals’ implicit attitudes toward violence were similar to the control group, and they also showed a negative implicit attitude toward violence; however, in the case of provoked criminals, the initial combined discrimination was prominently slower than the reversed combined discrimination, demonstrating that there was a relatively closer connection for violent behavior/positive combinations than for the violent behavior/negative combinations. This indicated that violent/positive information had a closer association and showed a positive implicit attitude toward violence. It is suggested that this subgroup of criminals not only had a robust implicit attitude toward violence, but it was in the opposite direction from the control group as well as from other criminals.

Regarding violence, several studies in the field of automaticity have highlighted that implicit attitudes toward violence predicted violence (Banse, Gossel, Kistemaker, Werner, & Schmidt, 2013; Todorov & Bargh, 2002). The IAT can screen the criminals without a mixed control group. However, when the criminals were mixed with a control group in the non-reversed group, initial and reversed combined response of the criminals was significantly longer than that of the control group. Therefore, the implicit attitude toward violence can be used to screen provoked impassioned criminals from a control group using the different IAT direction and longer response times. Results of implicit attitude toward violence as a diagnostic test for the classification of the criminals indicated that the IAT could screen impulsive and provoked criminals with higher positive predictive value. Consequently, it is suggested that implicit attitudes toward violence as assessed by the IAT not only discriminated criminals from a control group but also screened different subtypes of provoked impassioned criminals.

Impulsive criminals were relatively more impulsive, had a high degree of aggression, and had a low level of self-control (Ma, 1999; Magargee, 1966). Studies have confirmed that the perpetrators, when asked, often do not know why they committed a crime, and usually regret it (Song, 2007). Hence, they might think that violence is bad. Nevertheless, situational factors or personality might have an impact on their violent behavior. Provoked criminals subconsciously think that violence is not bad, but this was not reflected in their responses to the questionnaire. This group of people might hide their attitudes via VIO. Usually, as a result of over-control, consciousness is dominant in repressing the violence existing in the subconscious. However, if stimulated by strong incentives exceeding their tolerance, and if an outbreak of violence occurs, these criminals exhibit strong, even more violent reactions and less self-control (Anderson & Carnagey, 2003).

Limitations

Finally, the limitations of the present study are outlined. The first limitation is the small sample size in this study, the perpetrators are (thankfully) a rarity, and hence, their sample size was not as large as we would have wished. The second limitation is that only provoked impassioned criminals participated in the present study. It is well known that violent criminals include other kinds of criminals, such as reoffending criminals. It is suggested that future studies include such participants. The third limitation of this study was that the control group consisted of non-criminals. It is suggested that future studies include a control group of criminals who have not committed violent crimes. The fourth limitation is that there are many indirect methods of measuring implicit attitudes other than the IAT, whereas only IAT was used in this study. Moreover, we used response times in this study. However, it is suggested that future research assess neurobiological indices of violent attitudes, to investigate the neural mechanisms of such behaviors at a deeper level.

Conclusion

In the current study, explicit attitudes toward violence via VIO preliminary discriminated a small subgroup of the criminals from a control group, but could not differentiate subtypes of the criminals. However, implicit attitudes toward violence as assessed by the IAT not only screened the criminals from the control group but also precisely discriminated different subtypes of provoked impassioned criminals. Therefore, it is suggested that the IAT based on assessing explicit attitude toward violence could be an effective tool for the classification and prevention of violence. These findings may provide solid foundations for crime prevention, criminal penalties, and reformation.

Footnotes

Appendix

Acknowledgements

We acknowledge the Shaanxi Prison for the data collection and support of this project. We thank the offenders and the young men in this research.

Author Contributions

Conceived and designed the experiments: Muzhen Guan, Danmin Miao, Xufeng Liu. Performed the experiments: Muzhen Guan, Wei Xiao . Analyzed the data: Muzhen Guan, Xiaojing Li,Wei Xiao. Contributed reagents/materials/analysis tools: Danmin Miao, Xufeng Liu. Wrote the paper: Muzhen Guan, Xiaojing Li, Xufeng Liu. All authors read and approved the final manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Military of National Defense under Grant 12XLZ314.

References

Ajzen

(2001). Nature and operation of attitudes. Annual Review of Psychology, 52, 27-58.

Anderson

A. C.

Benjamin

J. A.

Wood

K. P.

Bonacci

M. A.

(2006). Development and testing of the Velicer Attitudes Toward Violence Scale: Evidence for a four-factor model. Aggressive Behavior, 32, 122-136.

Anderson

A. C.

Carnagey

N. L.

(2003). Exposure to violent media: The effects of songs with violent lyrics on aggressive thoughts and feelings. Journal of Personality and Social Psychology, 84, 960-971.

Anna

H. H.

Olson

M. A.

Fazio

R. H.

(2006). The influence of experimentally created extrapersonal associations on the Implicit Association Test. Journal of Experimental Social Psychology, 42, 259-272.

Banse

Fischer

(2002). Implicit and explicit aggressiveness and the prediction of aggressive behavior. Poster presented at the 11th Conference of personality of the European Association of personality psychology, Jena, Germany.

Banse

Gossel

K. J.

Kistemaker

M. L.

Werner

A. V.

Schmidt

F. A.

(2013). Pro-criminal attitudes, intervention, and recidivism. Aggression and Violent Behavior, 18, 673-685.

Baxendale

Cross

Johnston

(2012). A review of the evidence on the relationship between gender and adolescents’ involvement in violent behavior. Aggression and Violent Behavior, 17, 97-310.

Carolyn

A. S.

Timothy

O. I.

Terence

P. T.

(2005). Adolescent maltreatment and its impact on young adult antisocial behavior. Child Abuse & Neglect, 29, 1099-1119.

Chen

C.-Y.

Muggleton

N. G.

Juan

C.-H.

Tzeng

O. J.

Hung

D. L.

(2008). Time pressure leads to inhibitory control deficits in impulsive violent offenders. Behavioural Brain Research, 187, 483-488.

10.

Chen

Z. X.

(2004). The cutting-edge of criminal law. Beijing: Chinese people’s public Security University Press.

11.

Eagly

A. H.

Chaiken

(1993). The psychology of attitudes. Fort Worth, TX: Harcourt, Brace, Jovanovich.

12.

Eagly

A. H.

Chaiken

(2007). The advantages of an inclusive definition of attitude. Social Cognition, 25, 582-602.

13.

Eliot

Cornell

D. G.

(2009). Bullying in middle school as a function of insecure attachment and aggressive attitudes. School Psychology International, 30, 201-214.

14.

Evans

J. S. B. T

. (2008). Dual-processing accounts of reasoning, judgment, and social cognition. Annual Review of Psychology, 59, 255-278.

15.

Fazio

R. H.

(2007). Attitudes as object-evaluation association of varying strengths. Social Cognition, 25, 603-637.

16.

Fazio

R. H.

Jackson

J. R.

Dunton

B. C.

Williams

C. J.

(1995). Variability in automatic activation as an unobtrusive measure of racial attitudes—A bonafide pipeline. Journal of Experimental Social Psychology, 69, 1013-1027.

17.

Fazio

R. H.

Sambonmatsu

D. M.

Powell

M. C.

Kardes

F. R.

(1986). On the automatic activation of attitudes. Journal of Personality and Social Psychology, 50, 229-238.

18.

Fincham

F. D.

Cui

Braithwaite

Pasley

(2008). Attitudes toward intimate partner violence in dating relationships. Psychological Assessment, 20, 260-269.

19.

Funk

Elliott

Bechtoldt

(2003). The Attitudes Toward Violence Scale. Journal of Interpersonal Violence, 18, 186-196.

20.

Gray

N. S.

MacCulloch

M. J.

Smith

Morris

Snowden

R. J.

(2003). Violence viewed by psychopathic murderers. Nature, 423, 497-498.

21.

Greenwald

A. G.

McGhee

D. E.

Schwartz

L. K.

(1998). Measuring individual differences in implicit cognition: The Implicit Association Test. Journal of Personality and Social Psychology, 74, 1464-1480.

22.

Greenwald

A. G.

Nosek

B. A.

Banaji

M. R.

(2003). Understanding and using the Implicit Association Test: I. An improved scoring algorithm. Journal of Personality and Social Psychology, 85, 197-215.

23.

Greenwald

A. G.

Poehlman

T. A.

Uhlmann

E. L.

Banaji

M. R.

(2009). Understanding and using the Implicit Association Test: III. Meta-analysis of predictive validity. Journal of Personality and Social Psychology, 97, 17-41.

24.

Harvey

M. J.

Williams

D. J.

Donnelly

P. D.

(2012). Testing a method to develop preliminary cost estimates of homicide in Glasgow: A research note. Criminal Justice Policy Review, 24, 510-523.

25.

G. C.

(2000). Psychology of crimes of passion. Sichuan Police College, 12(2), 28-30.

26.

Houwer

D. J.

Mocigemba

T. S.

Spruyt

Moors

(2009). Implicit measures: A normative analysis and review. Psychological Bulletin, 135, 347-368.

27.

Jiang

R. B.

(2003). The characteristic, prevention and treatment of crime of passion. Juvenile Crimes and Delinquency, 4, 45-46.

28.

Kampfe

Penzhorn

Schikora

Dunzl

Schneidenbach

(2009). Empathy and social desirability: A comparison of delinquent and non-delinquent participants using direct and indirect measures. Psychology, Crime & Law, 15, 1-17.

29.

Loza

Loza-Fanous

(2000). Predictive validity of the Self-Appraisal Questionnaire (SAQ): A tool for assessing violent and nonviolent release failures. Journal of Interpersonal Violence, 15, 1183-1191.

30.

Loza

MacTavish

Loza-Fanous

(2007). A nine-year follow-up study on the predictive validity of the Self-Appraisal Questionnaire for predicting violent and nonviolent recidivism. Journal of Interpersonal Violence, 22, 1144-1155.

31.

K. C.

(1999). The general theory on criminal punishment. Wu Han, China: Wu Han University Press.

32.

Magargee

E. L.

(1966). Undercontrolled and overcontrolled personality types in extreme antisocial aggression. Psychological Monographs, 80(3), 1-29.

33.

McConville

D. W.

Cornell

D. G.

(2003). Aggressive attitudes predict aggressive behavior in middle school students. Journal of Emotional and Behavioral Disorders, 11, 179-187.

34.

Mills

J. F.

Kroner

D. G.

Forth

A. E.

(2002). Measures of Criminal Attitudes and Associates (MCAA): Development, factor structure, reliability, and validity. Assessment, 9, 240-253.

35.

Monahan

Steadman

H. J.

Silver

Appelbaum

P. S.

Robbins

P. C.

Mulvey

E. P.

(2001). Rethinking risk assessment: The MacArthur study of mental disorder and violence. New York, NY: Oxford University Press.

36.

Nisbett

R. E.

Wilson

T. D.

(1977). Telling more than we can know: Verbal reports on mental processes. Psychological Review, 84, 231-259.

37.

Polaschek

D. L. L.

Bell

R. K.

Calvert

S. W.

Takarangi

M. K. T.

(2010). Cognitive-behavioural rehabilitation of high-risk violent offenders: Investigating treatment change with explicit and implicit measures of cognition. Applied Cognitive Psychology, 24, 437-449.

38.

Ramírez

J. M.

Andreu

J. M.

(2006). Aggression, and some related psychological constructs (anger, hostility, and impulsivity) Some comments from a research project. Neuroscience & Biobehavioral Reviews, 30, 276-291.

39.

Ross

Quayle

Newman

Tansey

(2013). The impact of psychological therapies on violent behaviour in clinical and forensic settings: A systematic review. Aggression and Violent Behavior, 18, 761-773.

40.

Roura

Wringe

Busza

Nhandi

Mbata

Zaba

Urassa

(2009). “Just like fever”: A qualitative study on the impact of antiretroviral provision on the normalisation of HIV in rural Tanzania and its implications for prevention. BMC International Health and Human Rights, 9, Article 22.

41.

Sandstrom

M. J.

Jordan

(2008). Defensive self-esteem and aggression in childhood. Journal of Research in Personality, 42, 506-514.

42.

Snowden

R. J.

Gray

N. S.

Smith

Morris

MacCulloch

M. J.

(2004). Implicit affective associations to violence in psychopathic murderers. The Journal of Forensic Psychiatry & Psychology, 15, 620-641.

43.

Song

(2007). The causation and treatment of crime of passion. Business Economy, 5, 28-30.

44.

Steadman

H. J.

Silver

Monahan

Appelbaum

P. S.

Robbins

P. C.

Mulvey

E. P.

. . . Banks

(2000). A classification tree approach to the development of actuarial violence risk assessment tools. Law and Human Behavior, 24, 83-100.

45.

Steffens

M. C.

(2004). Is the implicit association test immune to faking? Experimental Psychology, 51, 165-179.

46.

Suter

Pihet

Ridder

Zimmermann

Stephan

(2014). Implicit attitudes and self-concepts towards transgression and aggression: Differences between male community and offender adolescents, and associations with psychopathic traits. Journal of Adolescence, 37, 669-680.

47.

Todorov

Bargh

J. A.

(2002). Automatic sources of aggression. Aggression and Violent Behavior, 7, 53-68.

48.

Wang

Zhang

S. L.

Guan

M. Z.

H. Z.

Liu

X. F.

(2011). The development and predictive research of the Abnormal Personality Risk Inventory. Journal of Psychological Science, 34, 1263-1268.

49.

Wilson

T. D.

Dunn

E. W.

(2004). Self-knowledge: Its limits, value, and potential for improvement. Annual Review of Psychology, 55, 493-518.

50.

Wilson

T. D.

Lindsey

Schooler

T. Y.

(2000). A model of dual attitudes. Psychological Review, 107, 101-126.

51.

World Health Organization. (2004). Handbook for the documentation of interpersonal violence prevention programmes. Geneva, Switzerland: Author.

52.

World Health Organization. (2010). Violence prevention: The evidence. Geneva, Switzerland: Author.

53.

Xiao

L. J.

(2007). Compilation and application on Chinese Soldier Personality Questionnaire (Doctorial dissertation). Xi’an, China.

54.

Zhang

Liu

Y. H.

Gan

Y. Q.

(2005). Introduction to personality of criminal law. Beijing, China: Law Press.

55.

Zhang

Z. B.

(1998). The study on criminal responsibility of juvenile delinquency (Doctorial dissertation). Wu Han, China.