Show Summary Details

Page of

 PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, CRIMINOLOGY AND CRIMINAL JUSTICE (criminology.oxfordre.com). (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy).

date: 12 December 2017

Copycat Crime

Summary and Keywords

The term “copycat crime” implies that the root of a crime can be found in exposure to a live model or media content concerning a prior crime. The four basic components of a copycat crime are a “generator crime” (a media portrayal of a crime or a real-world crime that is the precursor of a subsequent crime), “criminogenic models” (media content or real-world offenders that portray a subsequently copied crime), “copycat criminal” (an individual who commits a crime after being influenced by criminogenic media content, live models, or a combination of the two), and “copycat crime” (a crime whose occurrence or form is influenced by prior exposure to media content and/or live criminal models). The linked crimes are seen as sharing a unique criminogenic dynamic with the first crime serving as a generator for later copycat crimes with substantial elements of the first crime present in the second. The premise of copycat crime is that exposure to a generator crime is the linking mechanism and that the removal of the exposure would eliminate the occurrence or form of the subsequent copycat crime.

There is no theory of copycat crime, but there are four theoretical perspectives that are pertinent. Under the first perspective, imitation has been examined as a general human behavior within biology and psychology. The second perspective, social contagion, studies imitation in collective groups with a focus on the life cycle of crime waves. The third perspective is the study of the diffusion of social innovations and focuses on the factors that encourage adoption of a new, socially accepted and advocated behavior. In diffusion research, criminal behaviors have not been a major consideration. The final theoretical perspective is social learning theory, which focuses on how humans learn new behaviors in social settings. Despite the attention of these four research streams, research on copycat crime has not been extensive. One reason for the deficiency is the difficulty in identifying copycat crimes. Whereas other crimes are relatively straightforward to quantify and are routinely tallied in official law enforcement statistics, copycat crimes are not counted in any systematic way.

Copycat crime has traditionally been conceived within an emphasis on direct exposure to live person-to-person models. The media as a source of crime models has historically been downplayed. As the media evolved in the 20th century the study of mediated copycat crime models ascended so that the dominant contemporary view of copycat crime is that of media-sourced transmissions. Copycat crimes are today linked to literature, movies, television shows, music, video games, and print and television news, but despite concern and a large number of studies of violent media’s relationship to social aggression, the rigorous study of copycat crime has lagged. At this time, copycat effects are felt to be relatively rare and are most likely to appear in at-risk individuals predisposed to crime and in preexisting criminal populations. The effect of the media is thought to be more qualitative (affecting criminal behavior) than quantitative (affecting the number of criminals). Whether copycat crime effects influence any particular individual depends on the interaction of the content of a particular media product (its characterizations of crime and criminals), the individual’s predispositions toward crime (personal criminal history, family, and environmental factors), and the media’s social context (preexisting cultural norms toward crime, crime opportunities, and nature of the mass media). The most likely individual to be a copycat offender is hypothesized to be a socially isolated but criminally confident offender who is exposed to multiple live criminal models and has immersed themselves in criminogenic media.

Keywords: copycat crime, copycat criminals, generator crimes, crime waves, imitation, contagion, diffusion, social learning, terrorism, media and crime

What Constitutes a Copycat Crime?

The term “copycat crime” implies that the root of a crime can be found in exposure to a live model or media content concerning a prior crime. The linked crimes are seen as sharing a unique criminogenic dynamic with the first crime serving as a generator crime for the later copycat crime and substantial elements of the first crime present in the second. Although the crimes may be geographically and temporally separated, it is understood that exposure to the generator crime is the linking mechanism and that the removal of the exposure would eliminate the occurrence or form of the subsequent copycat crime. Exposure can take the form of live direct interactions or consumption of media content. In sum, application of the term “copycat” to a crime implies a unique and powerful criminological and psychological dynamic regarding the criminogenic influence of prior crime on the behavior of offenders. Similarly, a copycat criminal is understood to be an offender who lifts crime elements from live or media criminogenic models.

The study of offender imitation of modeled crimes has an extensive, although interrupted, tradition in criminology. Regarding media-sourced crimes, the media as criminogenic was largely ignored in criminology until the 1950s when, corresponding with the television networks, a media role in crime began to be seriously explored. Since the 1980s, copycat crime, and in particular, the media role in its generation have been accepted aspects of criminology. Crime clusters and crime waves have been regularly reported and studied since the 1960s (Coleman, 2004; Sacco, 2005). Contemporary copycat crimes have been credited to news media coverage of workplace violence, product tampering, hate crime, mass murder, airline hijacking, and terrorism and fictional portrayals of robbery, murder, arson, carjacking, and rape (Helfgott, 2015, p. 51). Copycat crime has been linked to crimes committed by an individual, small groups, or large crowds and a range of behaviors including collective acts like rioting (Bohstedt & Williams, 1988; Myers, 2000), small group crimes such as sniper and school shootings and hate crimes (Coleman, 2004), and individual acts such as suicides and lone-wolf terrorism (Phillips & Paight, 1987; Tuman, 2009). The anecdotal descriptions of copycat crimes in combination with research on copycat suicides and terrorism establish reasonable grounds that copycat crime occurs at a significant rate and that the media are the genesis of crime careers for a substantial minority of offenders (Surette & Maze, 2015). A recent estimate of the prevalence of media copycat crime is that one in four adult male offenders have attempted a copycat crime in their criminal careers (Surette, 2013a).

With the existence of the substantial nature of copycat crime generally accepted, the basic components of a copycat crime are as follows:

  1. 1. Generator crime—an account, portrayal of a crime in a media product or the demonstration of a crime by a live model that is the precursor of a subsequent crime. Generator crime portraits can be delivered by print, visual, audio, or new media channels or directly modeled live.

  2. 2. Criminogenic models—media content or live models that portray a crime which is subsequently copied.

  3. 3. Copycat criminal—an individual who commits a crime after being influenced through exposure to criminogenic media content, live models, or a combination of the two sources.

  4. 4. Copycat crime—a crime whose occurrence or form is influenced by prior exposure to media content and/or a live criminal model.

Why Study Copycat Crime?

While copycat crimes are rare events in comparison to all crime, understanding copycat crime is important for understanding the general influence of crime models and the media’s relationship to crime. The capability of media to generate copycat crime lies at the crux of the concern about media criminogenic effects; therefore, understanding the extent and dynamics of media-linked crime would forward understanding of the broad effects of media as well as the media’s relationship to crime and justice, social violence, and media-oriented terrorism. The study of copycat crime will provide another window into how media content relates to criminal behavior and how behavior in general is learned and diffuses through society. Copycat crime is an important element of the total crime picture and its importance is expected to increase as society becomes a more mediated, celebrity-dominated culture and as performance crimes (criminal acts created for social media distribution) grow in number (Penfold-Mounce, 2009; Surette, 2015).

In addition, the study of copycat crime has relevance for a number of criminological theoretical perspectives, most obviously social learning theory but also for subculture, cross-cultural, and life-course theories of crime. For social learning theorists, improved understanding of the role of behavior models would provide insight into the dynamics involved in broader noncriminal social learning processes. As new social media allow social interactions, personal relationships, and communities to develop in new digital and virtual ways, understanding how negative social behaviors diffuse would provide insight into contemporary social learning processes. The study of copycat crime would also benefit sub- and cross-cultural studies of crime. Knowledge about differential effects of media content across delinquent subcultures would increase understanding about how various cultural enclaves interact with and process criminogenic models and help identify what factors might be insulators versus catalysts in the mimicking of crime. Similarly, the study of copycat crime rates internationally and the comparison of high crime media/low copycat crime cultures with low crime media/high copycat crime cultures would advance the understanding of both copycat crime and cultural differences in crime rates. Lastly, as recent research suggests that copycat crime activity tends to occur early in criminal careers (Surette & Maze, 2015), the study of copycat crime’s role in the developmental life course of delinquents would be beneficial for understanding chronic offending.

Theories Applicable to Copycat Crime

While there is no dedicated theory of copycat crime there are four theoretical perspectives that are relevant for comprehending imitative crime. Under the first perspective, imitation has been examined as a general human behavior within biology and psychology. In biology, imitation is assessed through how evolutionarily adaptive it is, and research has looked to discover and explain the genetic and biological basis for imitation and the evolutionary advantage that a capability to imitate the behavior of others bestows. In psychology, the steps and thought processes that individuals undergo while imitating are the focus and research looked to explore imitation’s role in normal personality development.

In the second theoretical perspective, the study of imitation in collective behaviors has been pursued. Some of the behaviors studied were criminal, such as riots, and some were legal and benign, such as following fashion and fads. What was common was that the behavior usually spread and died rapidly and represented unusual social behavior by members of large crowds. Research on social contagion focuses on the life cycle of these waves of behavior. The third perspective is the study of the diffusion of social innovations. The social dynamics studied therein focused on the sources that encourage adoption of a new, usually socially advocated behavior rather than the processes involved in overcoming preexisting social restraints against expressing known behavior, as in social contagion. In diffusion research, criminal behaviors have not been a major consideration.

The fourth theoretical perspective is social learning theory, which focuses on how humans learn new behaviors in social settings. Also known as observational learning, from its base of conditioning theory and focusing first on peer-to-peer interactions, social learning looked to detail how behavior was socially acquired and implemented. Unlike the other three perspectives, social learning has been extensively applied to study crime. However, this perspective initially disparaged a significant media criminogenic role and only recently has embraced the media as a significant crime model source. Each theoretical perspective is briefly discussed.

Imitation

Understanding copycat crime starts with the concept of imitation, and the study of criminal imitation begins with Gabriel Tarde. The study of criminal imitation has historically focused on imitation via direct interpersonal contacts, with the role of media crime models seen as secondary. This is ironic in that the first criminologist to consider the imitation of crime, Gabriel Tarde (1912, p. 340), explicitly included the media as a significant source of crime models, and his best-known quote stated: “Infectious epidemics spread with the air or the wind; epidemics of crime follow the line of the telegraph.”

For Tarde, copycat crime spreads downward (from higher class to lower) and outward (from urban to rural community) and originates in the upper classes and in the urban cities and migrates via imitation to the lower classes and out to rural areas—copycat crime goes downward and outward (Tarde, 1912, pp. 331, 338; Vine, 1973). Imitation followed social laws according to Tarde and his “law of imitation” (people imitate one another in proportion to how much close contact they have with one another) is a precursor of Sutherland’s concept of “differential association.” Cities thus increase imitation by crowding people together. In rural areas, imitation of new behaviors and ideas is less frequent, and established customs tend to dominate. Tarde’s second law of imitation (inferiors imitate superiors) argues that the propagation of behavior from higher to lower classes is universal and describes changes in language, ideology, fashion, and ideas (Vine, 1973).

Despite the initial student of criminal imitation seeing an important role for media-based imitation, the subsequent research focused on the transition of criminal behavior from individual-to-individual offenders in live encounters. While descriptive and compelling, Tarde’s work did not lead to a testable theoretical model of criminal imitation and instead purported a simplistic social system built on ambiguously defined concepts and anecdotal case studies (Curtis, 1953; Vine, 1973, pp. 301–302). Criminology, media, and imitation subsequently went in different directions in the early 20th century, not to reunite until the 1950s. Before Tarde’s speculations regarding copycat crime could be advanced, the disciplines of biology and psychology had to develop a more rigorous scientific foundation for understanding imitation.

Biology and Imitation

Recognition of the commonality of imitation in humans traces back to antiquity and Aristotle. Modern research into imitation began in the 19th-century study of induced movements and ideomotor action in which imitation occurs without conscious thought. Today it is widely accepted that people tend to perform actions they see being performed by others (Prinz, 2005, p. 155). What has to be learned is when and where to imitate, not how to imitate. Imitation is so genetically useful that the convergent evolution on varied phylum developed imitative capabilities. Birds, gorillas, chimpanzees, humans, and probably dolphins and orangutans all imitate (Galef, 2005, p. 219). Mirror neurons have emerged as the leading contemporary candidate concept for a biological basis for imitation (Decety & Chaminade, 2005, p. 213; Kinsbourne, 2005, p. 163). A mirror neuron system also offers an explanation of the persistence of dysfunctional imitation, the imitation of behaviors that have negative consequences for the imitator, such as many copycat crimes. For imitation to be evolutionarily valuable, all that is needed is that the majority of its consequences be beneficial across a population. Copycat crime can therefore be dysfunctional for an individual, but the biological drives to imitate persist in the population because of the broad benefit of imitation for the social functioning of the group. For social creatures like humans though, automatic imitation that results in being attacked, banished, or arrested would need to be restrained (Kinsbourne, 2005, p. 34). The speculated evolutionary response is that humans automatically neurologically model behaviors as a first virtual step; then either act out or inhibit the virtually modeled behavior as a second cognitively considered step (Decety & Chaminade, 2005, p. 213). In other words, humans think about and mentally prepare to imitate far more than they actually imitate.

The Psychology of Imitation

Humans have hence evolved to incorporate malleable imitation into their psychological development. In arguing its role in human personality development, the psychology literature uses the term “mind-reading” (a.k.a. a “theory of mind”) to describe the psychology of imitation. Mind-reading in this sense is not a psychic ability but instead denotes how humans understand the intentions and emotions of others by observing behaviors, gestures, and speech to deduce the mental state of the person being observed. (Meltzoff, 2005, p. 56). Priming has emerged as the best psychological concept to bridge from the biological mirror neuron system which provides the psychological decision to imitate the capability to do so. Priming can be understood as providing a set of ideas and beliefs—the perception that the world is such that a particular type of crime, for example, is appropriate, justified, and likely to be successful. Watching a violent film can therefore prime hostile or aggressive thoughts in viewers and sometimes aggressive behaviors and criminal acts (Roskos-Ewoldsen, Roskos-Ewoldsen, Francesca, & Dillman, 2002, pp. 102–103). Relevant for how the decision to copy is psychologically arrived at, social cognition has been employed by both social psychology and mass communication theorists to study how people process media and socially supplied information. For copycat crime, social cognition is important for understanding how criminogenic models might be processed. Two social cognition decision pathways have been conceptualized and studied: systematic and heuristic. Which of these paths is traveled is determined by how much thought and reflection an individual expends on a behavior decision. More cognitively demanding, systematic processing is followed when it is possible to evaluate the validity of available information and when an individual is motivated to expend the effort to systematically collect and consider a lot of information before making a decision (Petty, Priester, & Brinol, 2002). The systematic path would therefore be related to thoughtful, planned, instrumental copycat crimes where the realistic chances of successfully copying a generator crime would be considered and estimated. Conversely, the heuristic path involves less rational, more spontaneous, emotional copycat crimes. Unplanned opportunity batteries, sexual assaults, and some hate-crime clusters would be examples. Heavy media consumption enhances heuristic copycat crimes by making criminogenic information easy to access, high in frequency, more recent in exposure, and more vivid in portrayal. Media reliance on the part of an individual further encourages heuristic processing by limiting time and access to nonmedia information sources and the evaluation of likely consequences from copying (Shrum, 2002).

However the two social cognitive passed paths do not encompass the most common way media content influences behavior, so an additional media pathway has developed in communications research: “narrative persuasion,” which captures the phenomenological experience of reading, hearing, or viewing a dramatic story. In contrast to systematic and heuristic processing, in narrative persuasion media information is simply absorbed, not assessed (Slater & Rouner, 2002). Narrative persuasion has been forwarded as a significant means to change social behavior and offers an explanation of media influence on initially unsympathetic consumers through the use of empathetic transitional characters (Fisch, 2002, p. 416). Transitional behavior models are seen to initially oppose a course of action but model for the consumer the process of attitude and behavior change (Green, Garst, Brock, & Chung, 2006; Polichak & Gerrig, 2002). Additional content characteristics that appear to enhance narrative persuasion effects include “clarity of presentation” and “explicitness of content” (Fisch, 2002, p. 416), both of which are present in much of the contemporary crime-related media.

Following the social cognition paths leads to another copycat crime relevant communication concept: “scripts.” Scripts can be conceived as preestablished behavioral directions that individuals mentally scroll up as needed. Two time modes by which media created behavior scripts have been offered (Huesmann, 1986, p. 132). The first is a short-term activating process for already acquired behavior in which the retrieval of already learned scripts is cued (i.e., most looters already have a theft script available when social cues signal its retrieval). It is hypothesized that most copycat crime is a short-term retrieval process carried out by established offenders, who when presented with a crime opportunity where environmental cues match previously learned crime scripts result in the copying of the generator script. The second process is the long-term acquisition and learning of new behavioral scripts. Learning new scripts combines exposure to criminogenic models and four social learning steps of modeling, skills acquisition, rehearsal, and performance (Bandura, 1973). This long-term process is felt to be relevant for instrumental goal-oriented copycat crime, which is planned. The expectation is that short-term scripts will pair with heuristics path copycat crimes and long-term scripts with systematic path copycat crimes.

Lastly, the psychological concept of “role play” is seen as crucial for normal personality development and is helpful in conceptualizing copycat crime (Claxton, 2005, p. 195). In role play, specific modeled behavior is not imitated but a type of “behavior set” is adopted. A child imitates by embellishing a prototype role model and produces behavior that need not be what was observed; instead, behavior is produced based on what the imitated social role is believed to demand. For example, children would behave differently toward an imaginary injured person depending upon whether they were playing the role of doctor, crime-fighting hero, or criminal villain. This sort of role imitation works to duplicate in a copier’s mind the supposed mental states of others and thereby involves mind-reading. It further suggests that the precopying media immersion reported for some violent copycat crime case studies is related to the childhood state assumed in role play. Role play also expands the number and types of crimes that a generator crime can procreate by allowing the copier to generalize the modeled criminal behavior from a specific crime to a generalized type of criminal. To role-play “terrorist” affects more behavior than the copying of an individual “terrorist act.”

Crime and Imitation

The biology and psychology of imitation provide a means to explore dysfunctional imitation, which characterizes the majority of copycat crimes (Surette & Maze, 2015). The broad benefits to society from the widespread ability to imitate outweigh the occasional negative consequences of imitating personally harmful acts for individual copycat criminals. Natural selection encourages imitation, even when copycat crime is regularly punished (Dijketerhuis & Bargh, 2001, p. 4). Because the biology and psychology of imitation dictates that learning copycat crime is relatively easy– it’s what many humans, particularly established offenders, will do unless something intervenes to prevent it—the reason copycat crime is not more common is that “selective imitation” limits its actual occurrence. Being able to select those behavioral steps that appear to lead to success and to build a general behavior model (such as in role play) that can be applied in various settings increases the potential to apply learned imitative criminal behavior. It also allows the copier to deselect those steps that are believed to lead to punishment. Research on selective imitation states that criminogenic content that reveals errors and how to avoid them should generate the strongest imitation effects by allowing individuals to pick and choose what and where to copy (Whiten, Horner, & Marshall-Pescini, 2005, p. 266). Humans often role-play without embracing the entire behavioral demands of a role. In sum, the refined imitation capabilities in humans increase the complexity and diversity of copycat crime in the real world while limiting its numerical reality. Many learn how to commit crimes, few attempt to commit them.

Social Contagion

Differing from imitation, which focuses on the learning of a new behavior, contagion theory addressed the expression of already learned but previously suppressed behavior (Cook & Goss, 1996). Social contagion is usually examined as the unintentional spread of a behavior that due to embarrassment, social norms, or fear of punishment is not normally seen in public. The neutralization of normal inhibitions against a behavior is seen as crucial. While not necessarily criminal (singing, dancing, praying, and other behaviors break out in social contagion waves), it is the contagion of criminal behavior such as rioting and lynching that attracted the most research interest. Negative behaviors are perceived as more contagious than law-abiding behavior, as the contagion process is seen as lifting or weakening the social constraints against such behaviors (Cook & Goss, 1996, p. 7). Actor perception of anonymity afforded within a crowd (in this literature called “deindividualization”) is a key. Thus, an individual may want a new television but is held back until seeing others looting and becomes confident that being a member of a crowd of looters will ensure anonymity and thereby protection from arrest.

Explanations of collective behavior first employed the concept of “evolutionary regression” to explain why members of a crowd copy one another. In this early view, crowds were thought to cause people to revert to more primitive, violent impulses. In the first refinement of the theory, the transformation of crowd members was abandoned and crowd behavior came to be explained from a “birds of a feather flocking together” adage. In this view, a crowd’s most important function was to assemble like-minded people (for example, persons prone to criminality) who then act accordingly. The contemporary view is that large crowds are composed of many small groups (family, friends, and acquaintances) acting as independent units and not necessarily acting homogenously (McPhail, 1991, p. 92). Not everyone at a riot is rioting; many are simply observing, and individuals in large groups are more influenced by the members of the small group they identify with than by large aggregate group dynamics. Gang members in a riot are therefore more likely to copy the behavior of their fellow gang members than the actions of a large number of strangers.

Across its history, contagion research has focused on the “interpersonal transmission” with face-to-face interactions the primary imitative force and analogous to a medical epidemic transmission model (Green, Horel, & Papchristos, 2017). The dominant view was that contagion mostly happens between individuals who come in direct contact with one another. As it evolved, contagion conceptualization has begun to include media as significant sources of model exposure or “doses” which interact with other factors (such as audience and observer’s preexisting susceptibility to crime) to generate contagion (Patel, Simon, & Taylor, 2013). Research on contagion implies that the spread of social behavior is nonlinear so that root-cause variables will be haphazard predictors of when a crime will become a contagious generator event (Cook & Goss, 1996). This means that the characteristics of the copiers will be less predictive than the initial, often idiosyncratic characteristics of the social setting in which the copiers sit. Akin to chaos theory, slightly different initial conditions will determine when and if criminal contagions occur and make predicting their appearance an error-prone task.

The application of contagion research to copycat crime is twofold. One is through the early study of one-time large-scale criminal events, such as a riot or lynching. The other is through the more recent study of waves of unique crimes, such as airline hijackings or terrorist kidnappings that wax and wane. While crime waves occur, there exists little empirical evidence of the widespread contagion of crime, and the process is most applicable to volatile crowds with anonymous members. Two contagion mechanisms have emerged that are applicable to crime. The first mechanism is reduction of the “approach/restraint” conflict for copiers. For potential copycat criminals, tension between copying the behavior for immediate benefits and not copying due to expected repercussions must be dealt with prior to imitation occurring (Diener, Beaman, Fraser, & Kelem, 1976; Freedman, 1982). Thus, looting an electronics store would be desirable (approached) due to the tangible products gained, but initially avoided (restrained) due to the potential for arrest. For potential copycat offenders, the generator crime model can demonstrate both the reward of performing a behavior and thereby make approaching the crime behavior greater and reduce expected harm and restraint on copying a behavior. The second mechanism, “deindividualization” (anonymity within a crowd) operates as an additional factor to lower the strength of restraints by reducing the perceived likelihood of being identified and punished (Diener, Beaman, Fraser, & Kelem, 1976).

What can be drawn from the contagion literature regarding copycat crime is first that serious criminal acts such as homicide can be crowd-generated via person-to-person processes with the mechanisms of deindividualization and approach/restraint conflict reduction playing important roles. In these one-time, often spontaneous events, the media are usually unimportant. Crime contagion can also spread from location to location over time to create crime waves (Sacco, 2005). In these crime waves, the media play important roles and are seen as strong sources of crime “doses” exposing a large number of susceptible individuals to criminogenic models (McPhail, 1991, p. 131). Copycat suicide bombing campaigns, kidnappings, beheadings, gang shootings, and airline hijackings provide real-world examples of this contagion wave process.

Diffusion of Innovations

Also rooted in the work of Tarde, useful copycat crime concepts are found within the study of the social diffusion of innovations. As described by its main voice, Everett Rogers, a decision to copy an innovation has four elements: the innovation (the behavior that is available to copy); communication channels (the pathways through which information about the innovation is transmitted); time (involving both the adoption rate of an innovation and when in an adoption cycle individuals adopt), and social system (the community and environment in which diffusions occur). An important unaddressed research question is whether the diffusion of crime mirrors the diffusion of socially positive innovations. The most significant contribution from diffusion research is the detailing of the characteristics of successful innovations and the individuals most prone to adopt them.

Concerning innovations, five characteristics have been discussed in the diffusion literature, of which “relative advantage” is the most pertinent for copycat crime. While most often weighed as a reduction in apprehension risk, a generator crime’s relative advantage can also be a gain in social status so that less effective but more spectacular means of committing a crime will sometimes be seen as possessing a substantial greater relative advantage over a prior crime method. The perceptions of the advantage of a crime innovation can also differ from its reality so that ineffective generator crimes might still be copied due to mistaken assignment of positive advantages by offenders, an example of “overadoption” in the diffusion literature (Rogers, 2003, pp. 231–232). Another characteristic of innovations, compatibility with preexisting social norms, is expected to play conflicting copycat crime roles. In general, increased compatibility is expected to increase copying, but sometimes innovations are copied because they represent a break with past practices and spread due to low social compatibility (Rogers, 2003, p. 245). A crime innovation may therefore be copied because it directly challenges existing norms, authority, or social policy even when copying is believed to increase arrest chances.

Two additional innovation characteristics that affect diffusion are “complexity” and “trialability.” Complexity refers to the difficulty to employ and understand an innovation so that complicated crimes, even when innovative and successful, would not be expected to be widely copied if they require unique skills or equipment. Related to complexity is trialability, an ability to try an innovation in a low-risk trial run (Rogers, 2003, pp. 257–258). Increased trialability of crime techniques should increase the copying of a crime innovation and media portrayals of successful crimes are hypothesized to possibly function as low-risk, high-visibility trials. Lastly, “observability” is the degree to which results of copying by others are visible to those considering adopting (Rogers, 2003, p. 258). Media attention that raises an innovative crime’s observability should further increase copycat crime through publicizing successful generator crimes (noting that success in some cases may simply be media attention).

The second diffusion element—communication channels—distribute information on how to commit crimes and help potential copiers calculate their odds of success versus punishment. The diffusion literature says surprisingly little about media as communication channels, but does suggest that mass media channels are more extensive—but that interpersonal channels are more effective—in persuading final copying decisions. The tradition in diffusion has been to research interpersonal nonmedia communication processes while simultaneously acknowledging, but not deeply examining, that the media do sometimes play important roles in the diffusion process. If the diffusion of copycat crime is similar to noncrime diffusion, media as a communication channel would increase knowledge of a generator crime, while peer interaction would determine adoption. It is also likely that where real-world peers are not available to encourage adoption decisions, new social media can be effective substitutes for face-to-face communication (Rogers, 2003, p. 207). A problem noted in the diffusion research on socially desirable innovations but pertinent for copycat crime is that potential adopters usually reside in nonoverlapping social groups with limited face-to-face interactions. Media can reduce this information flow block by providing communication channels that allow separated and dissimilar individuals to interact.

The third diffusion element, time, involves two copying decisions. The first is the decision to copy. Applied to crime, the five decision steps outlined in the diffusion literature would be (1) gaining knowledge of the generator crime, (2) being persuaded that copying a crime is a good idea, (3) deciding to copy the crime in the future, (4) actually attempting a copycat crime, and (5) confirming and assessing the results of copying a crime. For copycat crime, all five steps can occur quickly (a viewer watches a rioter on the news ignite a car, decides this is a good idea, and ignites a nearby car within minutes) or over long time frames (a criminal spends years culling the media for information on con games before attempting one). Thus, the second time-related decision is when in an innovation adoption wave a copier decides to copy. Discussing noncriminal behaviors, Rogers described five types of adopters in terms of when they adopt an innovation and their level of the characteristic “innovativeness.” Extrapolating the diffusion findings to copycat crime leads to the expectation that early wave copycats (innovators and early adopters in diffusion terminology) will differ significantly from subsequent later wave copycats (adopters described as “early majority” through “laggards” in diffusion). As the launchers of a copycat wave, innovators and early adopters are most relevant for understanding copycat crime as they are theoretically more attuned to generator crimes.

The final diffusion element, the “social system,” defines the environment in which an innovation diffuses. The key diffusion social system role is that of “change agent,” an individual who lobbies for the adoption of an innovation. Change agents come into play for copycat crime when a potential copycat offender with a “risk-of-arrest” concern looks to a crime model for risk reduction techniques. In this conception, crime media consumption is conceived as equivalent to multiple contacts with change agents, which the diffusion research has shown to be positively related to innovation adoption (Rogers, 2003, p. 382). However, research has not indicated a straightforward relationship between the quantity of crime media exposure to a propensity for crime, as offenders and nonoffenders and copycats criminals and noncopycat criminals tend to consume equal amounts and types of media (Hagell & Newburn, 1994; Surette & Maze, 2015). Another question concerning copycat crime is how effectively media-provided models (especially within social media) can substitute for real-world change agents for socially isolated individuals. Other important social system factors are preexisting crime-supportive social norms and the number of real-world crime models and criminal peers. The opportunities to employ criminogenic information would also impact which crimes are copied. And finally, the social reactions to crime will determine the consequences of copycat crime and the likelihood of crime waves. The hypothesis is that the less crime is socially seen as taboo, the more the diffusion of crime will follow the common diffusion pattern found for socially positive innovations; the more it is seen as something to be hidden, the more crime diffusion will differ. Additionally, the more social media is available to anonymously connect crime innovators with potential copycat offenders, the more widespread criminogenic diffusion will be. For copycat crime the social system in which diffusion occurs is forwarded as the ultimate determinant of the diffusion of crime.

Diffusion and Copycat Crime

In many aspects copycat crime qualifies as an example of diffusion “overadoption” of an innovation that to external observers should be rejected (Rogers, 2003). The primary question regarding copycat crime is whether the diffusion of crime mirrors the diffusion processes found for positive behaviors such as healthy lifestyles. For example, in that copycat criminals are likely to be disproportionately people on the lower end of the social structure (“downs” in diffusion terminology), crime innovations aimed at low achievers, the poor, and the illiterate should be more influential, opposite what the diffusion research reports for positive innovations. Also opposite of positive innovations which diffuse more rapidly among socially networked individuals, media content should be a powerful behavior influence for socially isolated, criminally inclined loners (Bandura, 1973). However, while much media crime content would seem to encourage crime copying by showing crime as rewarded, interpersonal communications with noncriminal role models should work against crime adoption by reminding potential copycats of real-world negative consequences. To this point, examples of the diffusion of innovative crimes exist but the nuances of the diffusion process are not currently understood.

Social Learning

In social learning (aka observational learning), behavior is acquired from watching what happens to others and developing expectations of the likely results from imitating. Rooted in operant conditioning theory, social learning promotes crime models as crucial for criminality. Early social learning theorists emphasized real-world models; later ones have seen the media as outstripping reality as crime model sources. Contemporary ideas regarding copycat crime are grounded in the combined theories of Bandura’s observational learning and Sutherland’s differential association. For Sutherland, the key factor determining whether people violate the law or not was the meaning they gave to their social conditions, not the conditions themselves. This initial rejection of social factors as directly significant for crime was later amended by Sutherland, who added the social-level concept of “differential social organization” to account for the influence of social factors such as poverty and racism (Barkan, 2001, p. 179). The merging of social learning and differential association followed the 1950s advent of national television networks in the United States. Missing since Gabriel Tarde, in 1956 criminologist Daniel Glaser reintroduced the mass media as a powerful learning source for crime. The criminological question Glaser raised was whether mass media could substitute for real-world associations within the differential association perspective. For Glaser, Sutherland’s “differential association” became “differential identification” (Glaser, 1956, p. 440). The gist of differential identification is that people pursue criminal behavior to the extent that they identify with real or imaginary persons for whom criminal behavior is acceptable or required. Criminal identification could occur during direct real-world associations, such as in delinquent gangs, and also through identification with criminal roles portrayed in mass media. At its root, criminal identification was conceived as a psychological process in which individuals imagined how persons in the role identified with would act, and would behave accordingly. Analogous to role play, the offender would think—I want to be a particular type of criminal, so I must imitate that type of criminality and try to act as that criminal would even in situations where I have to guess how to behave.

Following Glaser, the tenets of social learning were detailed by Albert Bandura in the 1960s. For Bandura, observational learning is most likely to happen when (1) detailed instruction combined with demonstrations is provided, (2) positive consequences are observable, and (3) guided practice is available. Bandura also elevated the mass media to be an important source of behavior models and saw media models as outnumbering real-world models (Bandura, 1973, p. 73). Bandura’s concept of “self-efficacy” has emerged as the most important copier characteristic in social learning theory. Self-efficacy is the level of belief individuals have that they can successfully complete a task, and high self-efficacy is strongly related to successfully learning new behaviors. Extended to crime, criminal self-efficacy, or one’s belief in their criminal abilities, would be predictive of an offender’s likelihood to attempt a copycat crime. High criminal self-efficacy is hypothesized to increase crime copying by raising an individual’s criminal goals, resulting in their attempting crimes not attempted in the past. They therefore look for new instructional crime models to copy. Second, high criminal self-efficacy increases the effort offenders are willing to expend on crime attempts; expecting success they become more willing to learn new crimes. Third, anticipating eventual success their perseverance in the face of obstacles and their resilience to failure will be raised. Prior arrests and punishments that did not deter are postulated to increase an offender’s criminal self-efficacy perception.

In addition to self-efficacy, a copier’s behavior is strongly influenced by observing the consequences to the behavior model. It is not necessary to observe a model being rewarded for criminal behavior for imitation to follow. It is the expectation of eventual reinforcement on the part of the copier that is essential for the copying of modeled crime, and when a criminal model reflects past rewards via lavish lifestyles the copying of the model increases. A crucial question involves the ability of media-portrayed punishment of a crime model to deter copying. Punishment is seen as related to copycat crime in three ways. First, the punishment levied on the criminal model (the punishment observed) is important. Second, the estimation of the likelihood of punishment that a potential copycat offender extracts (the punishment expected) is more important. When punishment is seen as unlikely when in reality it is probable, imitation will be weakly controlled by its actual consequences until cumulative experiences produce more realistic expectations (Bandura, 1973). Therefore, in the generation of copycat crime, punishment of the model may not reduce the copying of a crime because the estimation of the likelihood of punishment in the real world may not be simultaneously reduced. This can be true even when real-world punishment ultimately occurs. Third, the punishment that is administered to the copier (the experienced punishment) is most important. Applying Skinnerian conditioning tenets, random reward patterns for copying crimes generate copycat crime behavior that will be difficult to eliminate (Bandura, 1973, p. 224). Sporadic punishment or frequent neutral results for committing crime will not counteract the copy-sustaining effects of periodic random but substantial rewards. After new skills have become entrenched through initial reinforcement, they can become habitual even if they are only haphazardly rewarded. Further working against deterrence, punishment portrayed in some ways can be ineffective, even counterproductive. Media-depicted punishments can be interpreted by copycat prone consumers as lessons on “correctable errors” and how to avoid the mistakes that resulted in punishment in the media content. Copycat offenders may subsequently act on the belief that by modifying tactics they can gain the benefits of modeled crime in the real world without suffering costs portrayed in the media world. In gist, affixing a punishment to the end of a succession of successful crimes or to near successful ones should not be expected to counteract criminogenic learning effects (Bandura, 1973, p. 271).

Extrapolating to copycat crime, social learning provides five propositions (Bandura, 1973, pp. 72–73, 78, 86, 290). First, criminal behavior will be usually modeled from a variety of encounters with individuals, written accounts, and visual media. Second, multiple exposures to emotional models, detailed instructions, and demonstrations, combined with opportunities for guided practice, maximize copying and ensure more or less permanent retention of modeled behavior. Third, visual models will be more influential than nonvisual models in generating learning of criminal behavior. Fourth, rewarding consequences to a crime model are more effective for generating copying than whether the model or modeled behavior is seen as good or bad. Fifth, social conditions that increase the permissibility and rewarding of criminal behavior override personal dispositions—it is types of social inducements rather than types of people that predict copycat crime. In the end, copycat offenders will be differentiated by their criminal environments more than by their personality characteristics.

Toward a Copycat Crime Theory

Much of what Gabriel Tarde guessed about the imitation of crime in the 19th century was substantiated in the 20th century. While knowledge of the imitation of crime has grown, it is important to note that most of the criminological research to date has not been on media-generated copycat crime, but on the copying of crime absent a media role. Lacking substantial direct research on media-generated copycat crime, the relationships between media and crime remain largely unexplored territory. With that caveat in mind, at least theoretically the transfer of generator crime knowledge to copycat crime offenders should be common. Basic research in biology and psychology has established that imitation is likely innate and a deficiency in the ability to imitate is a marker of a serious psychological malfunction. The research question is not so much why there is copycat crime but why there is not more of it. Research suggests that a crucial behavioral step regarding imitation is the suppression of the imitation response and the cognitive recognition of when imitation would be dysfunctional. Understanding copycat crime requires understanding why the more common suppression of criminal copying does not occur for imitating offenders.

The fact that crime imitation is not always suppressed means that contagions of crime arise regularly and run the gamut from serious terrorism to trivial juvenile misdemeanors. “Person-to-media-to-person” contagion processes today occur—and serious criminal events rise, proliferate and decline—in contagion cycles. The imitative impact of modeled crime depends upon the characteristics of the crime, the cognitive manner in which potential copycats process generator crimes, and the social context in which potential copycat offenders find themselves. In contemporary mediated societies, the media play multiple roles. They have psychological impacts on perceptions of criminal self-efficacy, prime criminality, and reduce restraints on committing crime. When at-risk copycat criminals are psychologically engaged with crime models, instructional generator crimes can walk individuals through the copycat crime steps from initial exposure to final crime, providing the needed skills, the scripts to commit a crime, and when and where to best attempt the copied crime.

In the historical examination of copycat crime, two perspectives have competed. In the first, exemplified by Sutherland’s theory of differential association, the media is seen as unimportant and crime copying as the result of direct personal contact with crime models. The early studies of collective behaviors like riots and lynching and aspects of contagion theory which promoted a dose/response medical model of imitation saw media as comparatively unimportant. Additionally, substantial amounts of diffusion research emphases person-to-person contacts within face-to-face social networks and thereby also minimized a media role in the spread of crime. In these perspectives, crime might be copied, but the mechanism need not be and seldom is the media. In contrast, the competing second perspective focused directly on the impact of media-based crime models. Exemplified by social learning, which after a half-century lag took up the mantle from Gabriel Tarde, in this perspective the media are crucial. Recent versions of contagion, especially regarding terrorism, and diffusion, with particular relevance for socially negative behaviors like crime, have also come to embrace strong media roles. The current mixed copycat crime theoretical environment is one in which both live and media-sourced crime models can generate imitative crime. When they are available and psychologically linked to copiers, live crime models are seen as more impactful, when they are absent media models can substitute. The presence of either is thought to exacerbate the impact of the other (Surette, 2013b).

Copycat Crime Typologies

Typologies of copycat criminals were not an interest of the live model theorists who focused on the elements and relationships of the individual offender and on who they interacted with and how. Typologies specific to copycat crime have therefore been developed with media crime models in mind, and two copycat crime typologies have been forwarded.

The first effort to categorize copycat offenders was by Pease and Love, who applied a mix of personal characteristics, offense features, and motive-related variables to classify four types of copycat criminals (Pease & Love, 1984). Their copycat types included “mode copiers,” where the media provide a method of carrying out a criminal act already intended; “groups of copiers,” in which copycat crimes are committed as group activities; “mentally ill or deficient copiers,” where the copycat offenders exhibit mental illness or deficiency; and “terrorists and threateners,” who copy or threaten to copy a particularly frightening portrayed act whose primary purpose is the generation of terror among a large population. This initial attempt at a copycat typology is useful in terms of revealing some of the underlying nuances of copycat crime and moving the study of the phenomena from a monolithic conceptualization. The four-set typology is limited, however, by the confounding of copycat crime and criminal characteristics across the four categories.

A more recent but not yet validated copycat typology by Doley and colleagues is a six-category copycat crime typology encompassing media, individual, crime, and environmental dimensions (Doley, Ferguson, & Surette, 2013). In this typology, copycat crime elements are found in the characteristics of an imitated crime, in the crime model’s role in the generation of the crime, and in the copycat offender’s motivation for copying a crime. A copycat crime can be either instrumental or emotive. Instrumental copycat crimes have rational concrete goals such as money or merchandise and are felt to be common; emotive copycat crimes are related to emotional outbursts such as rage reactions and are felt to be rare. They can also be either triggered genesis crimes, in which a new crime occurs, or molded metamorphic crimes, in which how a crime unfolds is shaped. A genesis copycat crime, felt to be rare, would not have occurred without model influence, a metamorphic crime, which is felt to be common. These are different in form because of the crime model’s influence but would have occurred in some form regardless. Lastly, they can differ by their underlying motivations with the goal of reducing the risk of punishment or with the goal of attracting attention. Risk reduction copycat crimes are low-profile copycat crimes where offenders work to keep them unobserved and undiscovered and are felt to be common. Media attention copycat crimes are felt to be rare but also to receive the greatest media attention. The six types posited do not represent mutually exclusive categories. It is possible to classify a copycat crime according to one characteristic aspect from each of the three divisions with the permutations equating to eight different “copycat crime” groups—an instrumental, metaphoric, risk reduction copycat crime being one example.

Table 1. Copycat crime typology.

Instrumental

Emotive

Genesis

Metaphoric

Risk Reduction

Attention

Offense

profit or prestige

rage or revenge

trigger

rudder

covert

overt

Offender

career property offender

violent history, impulsive

mental illness, identification with models

career offender

career offender

need for fame, interest in crime content

Goals Prevalence

material Common

thrills Rare

recognition Rare

skill Common

theft, burglary Common

spectacles Rare

A Three-Path Copycat Crime Model

Adapting concepts from Bandura’s observational learning, Akers’s social learning of crime, Sutherland’s differential association, Anderson, Gentile, and Buckley’s general aggression model, Rogers’s social diffusion of innovations, and a set of communication-based social cognition ideas, a three-path, individual-level copycat crime model has been offered as an initial copycat crime framework (Surette, 2013a). The model is divided into three sections. The top portion of the model, from exposure to a media generator crime to acquisition of crime knowledge (step 1 to step 3), is felt to be a common experience for many offenders and media consumers in that many persons learn criminally instructive information, particularly from repeated media exposures. Thus, repeated playing of certain video games will impart the basic knowledge of how to be a sniper to many teenagers. However, very few will transverse the middle portion of the model (steps 4, 5, or 6 to step 8) and follow one of the three copycat paths down to step 8 (appraisal and adoption decision), at which committing a copycat crime is seriously considered. It is speculated that of the limited number of individuals who reach this point in the model most will decide not to pursue a copycat crime. The bottom portion of the model (steps 9 to 11) is thus the least traveled road, for these individuals come to a positive appraisal of copying generator crimes and decide to adopt and implement the modeled crimes. These actual copycat crime attempts are hypothesized to be rare as a proportion of all crime, but not rare in terms of criminal life histories. The pathway model is both a map of possible individual-level pathways to copycat crime and a diagram of a filtering process of potential copycat offenders in which only those with a unique combination of at-risk factors actually attempt a copycat crime.

Copycat Crime

As the typologies and copycat crime pathway model reflect, copycat crime should not be thought of as simply a subset of general crime. Copycat crime occupies a realm that involves concepts ranging from the individual bioneurological level to the broad cultural level. Like other social behaviors, crime does appear to go downward and outward. In today’s social media society where crime fashions spread with digital speed and crime performances have waiting audiences, copycat crime can also go over and around so that Tarde’s linear telegraph imagery has been replaced with a digital multidirectional, multimedia social interactive reality (Surette, 2016a).

Divisions in Past Approaches

Past approaches to copycat crime have three divisions. First, there has been a division in the research between live face-to-face versus media-sourced crime models. Second, there has been a long, often contentious debate over the media’s role as either causal triggers or catalyst rudders in the generation of crime and violence. Third, research on copycat behavior has not been particularly focused on crime but has instead studied imitative suicide, the contagion of terrorism, or collective violence. The main ongoing division regards the causal role that media models play as criminogenic factors. The answer to the question whether the media generally are triggers (direct causes of crime and violence that would not have occurred absent media exposure) or rudders (formative catalysts in events that would have happened but with different characteristics) is yet to be definitely determined. The resolution of the question is important because if media exposure emerges as a direct cause, broad-based public policies focused on controlling content and access would be sensible. If, however, the media are catalysts for crime, a focus on at-risk individuals aimed at reducing the likelihood of offending rather than the likelihood of exposure makes sense. Research to date suggests that, at least as far as established offenders are concerned, the effect of the media is more qualitative (affecting criminal behavior) than quantitative (affecting the number of criminals). Limited evidence so far suggests that the media is less often a trigger for copycat crime and more often its rudder (Chadee, Surette, Chadee, & Brewster, 2017).

Needed Research

A number of research areas concerning copycat crime are in need of attention. The first area involves the measurement, misidentification, and prevalence of copycat crime. An additional set of research questions remain regarding the separate component parts of a copycat crime. Research is needed on the characteristics of generator crimes, criminal models, and copycat offenders. Lastly, the general working dynamics of copycat crime, the social settings, interactions between media and real-world conditions, and the evolution of media all need research attention to determine who among copycat offenders are most influenced by media imagery and content and the factors (dispositional, cultural, and situational) that explain increased susceptibility (Surette, 2002, 2015).

Identifying Copycat Crimes

The identification of a copycat crime is haphazard. Due to loose usage of the term, the study of copycat crime is hindered by misidentification, particularly false positives, where crimes that share elements with prior media content occur are incorrectly labeled copycats. How to best measure and identify a copycat crime is yet to be established. An initial response applies seven indicators of copycat crime which have been suggested, along with a scoring method to evaluate individual crimes as substantiated copycat crimes (Surette, 2016b). The seven indicators are (1) time order and proximity, (2) theme consistency (adoption of the personae, dress, and mannerisms of a crime model), (3) scene specificity (mimicking a specific portrayed act with close imitation of specific words, gestures, or behaviors), (4) repetitive exposure to a generator crime and crime model, (5) self-editing (exposure to a select portion of a crime through freeze-framing or slow motion to isolate a crime sequence), (6) offender statements, and (7) second party statements by family, friends, acquaintances, teachers, or investigators concerning the role of a crime model in the generation or form of a crime. This approach and other measurement approaches need further examination and validation.

Prevalence

Tied to the identification problem is the difficulty in examining copycat crime levels. Whereas other crimes are relatively straightforward to quantify and are routinely tallied in official law enforcement statistics, copycat crimes are not counted in any systematic way. An unknown number of copycat crimes invariably go unnoted, and not surprisingly, the label “copycat crime” is often questionably attached, and what is thought to be known about copycat crime has been largely surmised from anecdotal reports of crimes labeled in the media as copycats after their occurrence (Surette, 2015).

The sole estimation of the prevalence of copycat crime is found in a research note that analyzed ten studies of surveys that contained self-reported copycat crime responses (Surette, 2014). All but two studies involved surveys of incarcerated individuals, two included females, and six sampled teenage youth. The majority of the respondents were adult males. Each study reported a significant nonzero copycat crime estimate of the proportion of respondents stating that they had attempted a copycat crime in their lifetimes. The combined cross-study proportion estimated was .26, with a range across the ten studies from .13 to .46. Additional copycat crime prevalence estimates from additional and more diverse populations, in particular nonincarcerated groups, is needed to establish the substantive importance of copycat crime.

Nonoffenders and Copycat Crime

Existing research on copycat crime has focused almost exclusively on incarcerated offenders. Conducted in Trinidad, only one study to date provides an estimate of nonoffender, nonincarcerated subjects (Chadee, Surette, Chadee, & Brewster, 2017). The comparison of delinquent and nondelinquent teenagers would be particularly valuable with controls for gender and age, prior criminality, and perceptions of their criminal efficacy, exposure to real-world crime models, identification with media characters, levels of media immersion, and attitudes concerning need for fame and empathy. A needed methodology that has not been exploited is in-depth interviews with self-reported copycats with rigorous follow-up regarding their relationships with criminogenic media. A related unexplored research question involves hypothesized differences between established offenders who report past copycat criminality and those without such activity that has been examined within a single study by Surette and Maze (2015).

Generator Crimes

Research on the characteristics of generator crimes is crucial, especially the impact of portraits of successful crimes. It has been speculated that media content that initially shows a string of successful crimes followed by a concluding failed crime will not cancel imitative effects. Failure of punishment to deter copying most likely occurs when correctable errors are embedded in the failed crime portrait. Research on the accuracy of this speculation is important for social policy regarding media content which is currently based on the premise that final punishment cancels prior copycat effects. In addition, a set of criminogenic content characteristics beyond those associated with specific generator crimes also are thought to contribute to copycat crime likelihood and require confirming studies. For example, it is thought, without empirical evidence, that media content that reduces individual responsibility and distress for crime or reduces the perception of harm from crime or that condones crime as righteous will encourage copying (Akers, 1998, p. 79; Bandura, 1973, pp. 203, 216, 231).

Criminal Models

Concerning the characteristics of the generator criminal models, it has been argued that models who are attractive, desirable, admired, high-status, and high-prestige and who are competent, instructive, and rewarding will be more readily imitated (Wilson, Colvin, & Smith, 2002). Also, seeing models who are demographically similar successfully commit crimes increases self-efficacy and is expected to increase criminogenic copying by individuals (Bandura, 1995). The most important modeled crime characteristic that has been forwarded relates to the consequences (success or failure) of the crime models that lead to expectations on the part of copiers concerning imitating portrayed crimes (Bandura & Walters, 1963). These conclusions about the crime model characteristics are based on research that for the most part looked at the copying of noncriminal behaviors, leaving open the possibility that crime models and their copiers will share different characteristics.

Copycat Offenders

Similar to the needed research on criminal models, a set of hypothesized characteristics of copycat offenders awaits confirming research efforts. A set of personality factors including low self-esteem and self-control, disinhibition, sensation-seeking, a history of reward for imitation, and a high degree of dependence (seek help on easy tasks) have been forwarded as related to individuals’ copying crimes. In addition and suggested as foremost is the characteristic of a personal prior crime history combined with the trait of high criminal self-efficacy belief. Similarly, offenders weakly networked into law-abiding groups and strongly networked into deviant groups are thought to be high risks to commit copycat crimes.

Another set of factors that has been argued as copycat crime flags are related to how individuals interact with media, their media preferences, and motives for using media. Along these lines, individuals who prefer crime content combined with information-seeking or learning goals are hypothesized as more at risk to be copycats. Also, media use that reflects a concentration on a single media source to the exclusion of other media and nonmedia interpersonal contacts has been forwarded as a copycat crime flag (Akers, 1998). In particular, offenders who immerse themselves in a criminogenic niche media (for example, watching criminogenic excerpts from a single video repeatedly) have been forwarded as especially at-risk copycats (Meloy & Mohandie, 2001). Immersion is also associated with identification with media offenders so that persons who see media personalities as opinion leaders and as friends are postulated as more likely copycat crime candidates (Mundorf & Laird, 2002). Those who form parasocial relationships with media personalities, who are media-immersed, and who imagine themselves as innovative offenders are also hypothesized to be likely to copy offenses (Mazur, 2006, p. 297). Lastly, a frequently mentioned risk characteristic is being an isolated media consumer, which is felt to increase dependence on the media for information and block copiers from negative social assessments of their plans from peers (Akers, 1998, pp. 32, 49, 54). The actual role and interaction of all of the above characteristics have not been adequately explored.

Social Settings

The hypothesized characteristics of the social settings most conducive to copycat crime revolve around the social reaction to a copycat crime within family and neighborhood. In particular, it is thought that observed real-world criminal successes significantly weaken the inhibitory power of portrayed generator crime punishment (Bandura, 1973, p. 272). Thus in high crime—low punishment communities, the observed reward for crime, the estimation of the likelihood of punishment, and expectation of low severity when caught are felt to increase offenders’ confidence that they can successfully copy crimes (Bandura, 1973, pp. 224–225). In addition, it is speculated that societies with inconsistent patterns of punishment and rewards will generate copycat crime, as copiers will not expect a reward for each crime but will anticipate that the next crime will produce one (Bandura, 1973, p. 224). At the family setting level, the literature suggests that criminal parents should produce more copycat offspring as, even when they preach law-abiding behavior and punish law-breaking acts, criminal parents provide powerful criminogenic models (Mazur, 2006, p. 288). An interaction between the availability of real-world and media crime models on copycat crime propensity is postulated. Here again, regarding copycat crime these relationships reside as mostly untested hypotheses. Beyond the family and neighborhood, a pervasive crime culture, a history of crime and crime-saturated media have all been argued to increase copycat crime rates (Akers, 1998, p. 81). These effects should be heightened in societies with urban concentrations of crime where the diffusion of innovative crime techniques that originate in large urban areas can diffuse to smaller cities and thereafter to rural areas (Fisher, 1980, p. 416). In sum, social settings with real and media crime models and with the cultural values, community structure, and social histories that support crime are hypothesized to support the highest copycat crime rates—an additional hypothesis in need of confirmatory research.

Life Course and Copycat Crime

When queried, a number of the self-reported copycat offenders commonly cited a media role operating at the beginning of their crime careers, often as a model for their first or other early career offense usually committed as teenagers (Surette & Maze, 2015). The possibility that media have their greatest criminogenic impact on young, beginning offenders needs confirmatory research. In addition to when in a crime career copycat crime is most likely, the literature supports the idea of the media as primarily a catalyst or rudder for crime more than as a direct causal trigger of crime. Generator crimes are thought to influence crime by providing models to inclined individuals who also have real-world models of crime available. That is, preinclined offenders who are also exposed to real-world criminality are hypothesized to seek out media crime models and extract crime instructions. Additional research on the combined impact of exposure to real-world and media crime models as well as possible insulators to copycat effects is required to determine whether the role of the media is causal or catalyst.

Social Media

A recent emerging topic needing research is the relationship between social media and copycat crime. The development of new media with characteristics of interactivity and on-demand are felt to considerably magnify the effects of direct exposure to criminogenic content and to extend and accelerate diffusion and contagion. Social media-based contagion of crime and terrorism is an accepted contemporary phenomenon and contagion research offers insights into the role that new digital and social media might play in copycat crime (Nacos, 2009; Surette, Hansen, & Noble, 2009; Surette, 2016a). The basic research question is whether legacy media-generated copycat crime has evolved into a qualitatively different type of new media-based copycat crime?

Copycat Crime Dynamics

Looking at the literature collectively, whether copycat crime effects emerge in any particular individual is felt to depend upon the interactions of the elements of a specific generator crime (the characterization of crime and criminal), the potential copycat’s predispositions toward crime (personal criminal history, family, and environmental factors), and the social context (preexisting cultural norms, crime opportunities, and characteristics of the mass media). The most likely individual to be a copycat offender is hypothesized to be socially isolated but criminally confident offenders who are exposed to multiple live criminal models, have immersed themselves in criminogenic media, and hold attitudes supportive of crime.

Despite the lack of a large body of direct research, three propositions derived from the extant literature exist which summarize the current working conclusions concerning copycat crime. All are hypotheses awaiting empirical verification and rigorous research attention.

  • Hypothesis 1: A copycat effect will concentrate in preexisting criminal populations.
  • Hypothesis 2: A copycat effect is more commonly qualitative (affecting criminal behavior) rather than quantitative (affecting the number of criminals).
  • Hypothesis 3: Copycat criminals are also more likely to be career criminals involved in property offenses rather than first or violent offenders.

Copycat crime is a phenomenon that is expected to increase in both quantity and social importance in the 21th century. While copycat effects are felt to be comparatively rare and are most likely to appear in at-risk individuals predisposed to crime or in a preexisting criminal population, it is a reasonable expectation that the media’s criminogenic effects will be enhanced in the future as the media and society continue to coevolve.

Primary Sources

The following are offered as examples of recent or basic research related to copycat crime or introductory overviews of the pertinent theoretical perspectives.

Copycat Crime Studies

Chadee, D., Surette, R., Chadee, M., & Brewster, D. (2017). Copycat crime dynamics: A structural equation model of the interplay of empathy, narrative persuasion, and copycat crime. Psychology of Pop Media Culture, 6(2), 142–158.Find this resource:

Coleman, L. (2004). The copycat effect: How the media and popular culture trigger the mayhem in tomorrow’s headlines. New York: Simon and Schuster.Find this resource:

Doley, R., Ferguson, C., & Surette, R. (2013). Copycat firesetting: Bridging two research areas. Criminal Justice and Behavior, 40(12), 1382–1401.Find this resource:

Ehrlich, H. (1962). The swastika epidemic of 1959–1960: Anti-semitism and community characteristics. Social Problems, 9(3), 264–272.Find this resource:

Glaser, D. (1956). Criminality theories and behavioral images. American Journal of Sociology, 61(5), 433–444.Find this resource:

Helfgott, J. (2015). Criminal behavior and the copycat effect: Literature review and theoretical framework for empirical investigation. Aggression and Violent Behavior, 22, 46–64.Find this resource:

Meloy, J., & Mohandie, K. (2001). Investigating the role of screen violence in specific homicide cases. Journal of Forensic Science, 46(5), 1113–1118.Find this resource:

Surette, R. (2002). Self-reported copycat crime among a population of serious and violent juvenile offenders. Crime and Delinquency, 48, 46–69.Find this resource:

Surette, R. (2013a). Pathways to copycat crime. In J. Helfgott (Ed.), Criminal psychology (pp. 251–273). Santa Barbara, CA: Praeger.Find this resource:

Surette, R. (2013b). Cause or catalyst: The interaction of real world and media crime models. American Journal of Criminal Justice, 38(3), 392–409.Find this resource:

Surette, R. (2014). Estimating the prevalence of copycat crime: A research note. Criminal Justice Policy Review, 25(6), 703–718.Find this resource:

Surette, R. (2015). Thought bite: A case study of the social construction of a crime and justice concept. Crime, Media, Culture, 11(2), 360–374.Find this resource:

Surette, R. (2016b). Measuring copycat crime. Crime, Media, Culture, 12(1), 37–64.Find this resource:

Surette, R., & Maze, A. (2015). Video game play and copycat crime: An exploratory analysis of an inmate population. Psychology of Popular Media Culture, 4(4), 360–374.Find this resource:

Towers, S., Gomez-Lievano, A., Khan, M., Mubayi, A., & Castillo-Chávez, C. (2015). Contagion in mass killings and school shootings. PLoS One, 10(7), e0117259–e0117259.Find this resource:

Terrorism

Nacos, B. L. (2007). Mass-mediated terrorism: The central role of the media in terrorism and counterterrorism. Rowman & Littlefield.Find this resource:

Tuman, J. S. (2009). Communicating terror: The rhetorical dimensions of terrorism. SAGE.Find this resource:

Imitation

Hurley, S. L., & Chater, N. (2005). Perspectives on imitation: From neuroscience to social science. Cambridge, MA: MIT Press.Find this resource:

Contagion

Green, B., Horel, T., & Papchristos, A. (2017). Modeling contagion through social networks to explain and predict gunshot violence in Chicago, 2006 to 2014. JAMA Internal Medicine, 177(3), 326–333.Find this resource:

Sacco, V. (2005). When crime waves. Thousand Oaks, CA: SAGE.Find this resource:

Diffusion

Rogers, E. (2003). Diffusion of innovations. New York: Free Press.Find this resource:

Social Learning

Akers, R. L. (1998). Social learning and social structure: A general theory of crime and deviance. Athens, GA: Northeastern University Press.Find this resource:

Bandura, A. (1973). Aggression: A social learning analysis. Englewood Cliffs, NJ: Prentice-Hall.Find this resource:

Pratt, T. C., Cullen, F. T., Sellers, C. S., Thomas Winfree, Jr., L., Madensen, T. D., Daigle, L. E., . . . Gau, J. M. (2010). The empirical status of social learning theory: A meta‐analysis. Justice Quarterly, 27(6), 765–802.Find this resource:

References

Akers, R. L. (1998). Social learning and social structure: A general theory of crime and deviance. Athens, GA: Northeastern University Press.Find this resource:

Bandura, A. (1973). Aggression: A social learning analysis. Englewood Cliffs, NJ: Prentice-Hall.Find this resource:

Bandura, A. (1995). Self-efficacy in changing societies. Cambridge, U.K.: Cambridge University Press.Find this resource:

Bandura, A., & Walters, R. (1963). Social learning and personality development. New York: Holt, Rinehart, and Winston.Find this resource:

Barkan, S. (2001). Criminology: A sociological understanding. Upper Saddle River; NJ: Prentice-Hall.Find this resource:

Bohstedt, J., & Williams, D. E. (1988). The diffusion of riots: the patterns of 1766, 1795, and 1801 in Devonshire. Journal of Interdisciplinary History, 19(1), 1–24.Find this resource:

Chadee, D., Surette, R., Chadee, M., & Brewster, D. (2017). Copycat crime dynamics: A structural equation model of the interplay of empathy, narrative persuasion, and copycat crime. Psychology of Pop Media Culture, 6(2), 142–158.Find this resource:

Claxton, G. (2005). Joining the intentional dance. In S. L. Hurley & N. Chater (Eds.), Perspectives on imitation: From neuroscience to social science (Vol. X, pp. 193–196). Cambridge, MA: MIT Press.Find this resource:

Coleman, L. (2004). The copycat effect: How the media and popular culture trigger the mayhem in tomorrow’s headlines. New York: Simon and Schuster.Find this resource:

Cook, P. J., & Goss, K. A. (1996). A selective review of the social-contagion literature. Working paper, Sanford Institute of Policy Studies, Duke University.Find this resource:

Curtis, J. (1953). Gabriel Tarde. In C. Mihanovich (Ed.), Social theorists (pp. 142–157). Milwaukee, WI: Bruce Publishing.Find this resource:

Decety, J., & Chaminade, T. (2005). The neurophysiology of imitation and intersubjectivity. In S. Hurley & N. Chater (Eds.), Perspectives on imitation: From neuroscience to social science (Vol. 1, “Mechanisms of imitation,” pp. 119–140). Cambridge, MA: MIT Press.Find this resource:

Diener, E., Beaman, A., Fraser, S., & Kelem, R. (1976). Effects of deindividuation variables on stealing among Halloween trick-or-treaters. Journal of Personality and Social Psychology, 33(2), 178–183.Find this resource:

Dijksterhuis, A., & Bargh, J. (2001). The perception–behavior expressway: Automatic effects of social perception on social behavior. In M. Zanna (Eds.), Advances in experimental social psychology (Vol. 30, pp. 1–40). New York: Academic Press.Find this resource:

Doley, R., Ferguson, C., & Surette, R. (2013). Copycat firesetting: Bridging two research areas. Criminal Justice and Behavior, 40(12), 1382–1401.Find this resource:

Fisch, S. (2002). Vast wasteland or vast opportunity? Effects of educational television on children’s academic knowledge, skills, and attitudes. In J. Bryant & D. Zillmann (Eds.), Media effects: Advances in theory and research (pp. 397–426). Mahwah, NJ: Lawrence Erlbaum.Find this resource:

Fisher, C. (1980). The spread of violent crime from city to countryside, 1955 to 1975. Rural Sociology, 45(3), 416–434.Find this resource:

Freedman, J. (1982). Theories of contagion as they relate to mass psychogenic illness. In M. J. Colligan, J. S. Pennebaker, & L. R. Murphy (Eds.), Mass psychogenic illness (pp. 171–182). Hillsdale, NJ: Erlbaum.Find this resource:

Galef, B. (2005). How to analyze learning by imitation. In S. Hurley & N. Chater (Eds.), Perspectives on imitation: From neuroscience to social science (Vol. 1, “Mechanisms of imitation,” pp. 218–220). Cambridge, MA: MIT Press.Find this resource:

Glaser, D. (1956). Criminality theories and behavioral images. American Journal of Sociology, 61(5), 433–444.Find this resource:

Green, M., Garst, J., Brock, T., & Chung, S. (2006). Fact versus fiction labeling: Persuasion parity despite heightened scrutiny of fact. Media Psychology, 8, 267–285.Find this resource:

Hagell, A., & Newburn, T. (1994). Young offenders and the media: Vowing habits and preferences. London: Policy Studies Institute.Find this resource:

Helfgott, J. (2015). Criminal behavior and the copycat effect: Literature review and theoretical framework for empirical investigation. Aggression and Violent Behavior, 22, 46–64.Find this resource:

Green, B., Horel, T., & Papchristos, A. (2017). Modeling contagion through social networks to explain and predict gunshot violence in Chicago, 2006 to 2014. JAMA Internal Medicine, 177(3), 326–333.Find this resource:

Huesmann, L. (1986). Psychological process promoting the relations between exposure to media violence and aggressive behavior by the viewer. Journal of Social Issues, 42(3), 125–139.Find this resource:

Kinsbourne, M. (2005). Imitation as entrainment: Rain mechanisms and social consequences. In S. Hurley & N. Chater (Eds.), Perspectives on imitation: From neuroscience to social science (Vol. 2, “Imitation, human development, and culture,” pp. 163–172). Cambridge, MA: MIT Press.Find this resource:

Mazur, J. (2006). Learning and behavior. Upper Saddle River, NJ: Pearson/Prentice Hall.Find this resource:

McPhail, C. (1991). Myth of the madding crowd. Aldine De Gruyter.Find this resource:

Meloy, J., & Mohandie, K. (2001). Investigating the role of screen violence in specific homicide cases. Journal of Forensic Science, 46(5), 1113–1118.Find this resource:

Meltzoff, A. (2005). Imitation and other minds: The “like me hypothesis.” In S. Hurley & N. Chater (Eds.), Perspectives on imitation: From neuroscience to social science (Vol. 2, “Imitation, human development, and culture,” pp. 55–78). Cambridge, MA: MIT Press.Find this resource:

Mundorf, N., & Laird, K. (2002). Social and psychological effects of information technologies and other interactive media. In J. Bryant & D. Zillmann (Eds.), Media effects: Advances in theory and research (pp. 583–602). Mahwah, NJ: Erlbaum Publishers.Find this resource:

Myers, D. (2000). The diffusion of collective violence: Infectiousness, susceptibility, and mass media networks. American Journal of Sociology, 106(1), 173–208.Find this resource:

Nacos, B. (2009). Revisiting the contagion hypothesis: Terrorism, news coverage, and copycat attacks. Perspectives on terrorism, 3(3). Retrieved from http://www.terrorismanalysts.com/pt/index.php/pot/article/view/73/html.Find this resource:

Patel, D., Simon, M., & Taylor, R. (2013). Contagion of violence: Workshop summary. Washington, DC: National Academies Press.Find this resource:

Pease, S., & Love, C. (1984). The copycat crime phenomenon. In R. Surette (Ed.), Justice and the media (pp. 199–211). Springfield, IL: Thomas.Find this resource:

Penfold-Mounce, R. (2009). Celebrity culture and crime. London: Palgrave.Find this resource:

Petty, R., Priester, J., & Brinol, P. (2002). Mass media attitude change: Implications of the elaboration likelihood model of persuasion. In J. Bryant & D. Zillmann (Eds.), Media effects: Advances in theory and research (pp. 155–198). Mahwah, NJ: Erlbaum Publishers.Find this resource:

Phillips, D. P., & Paight, D. J. (1987). The impact of televised movies about suicide: A replicative study. The New England Journal of Medicine, 317(13), 809–811.Find this resource:

Polichak, J., & Gerrig, R. (2002). Get up and win! Participatory responses to narrative. In M. Green, J. Strange, & T. Brock (Eds.), Narrative impact (pp. 71–95). Mahwah, NJ: Lawrence Erlbaum.Find this resource:

Prinz, J. (2005). Imitation and Moral Development. In S. Hurley & N. Chater (Eds.), Perspectives on imitation: From neuroscience to social science (Vol. 2, “Imitation, human development, and culture,” pp. 267–282). Cambridge, MA: MIT Press.Find this resource:

Rogers, E. (2003). Diffusion of innovations. New York: Free Press.Find this resource:

Roskos-Ewoldsen, D., Roskos-Ewoldsen, B., Francesca, R., & Dillman, C. (2002). Media priming: A synthesis. In J. Bryant & D. Zillmann (Eds.), Media effects: Advances in theory and research (pp. 97–120). Mahwah, NJ: Lawrence Erlbaum.Find this resource:

Sacco, V. (2005). When crime waves. Thousand Oaks, CA: SAGE.Find this resource:

Shrum, L. (2002). Media consumption and perceptions of social reality: Effects and underlying processes. In J. Bryant & D. Zillmann (Eds.), Media effects: Advances in theory and research (pp. 69–95). Mahwah, NJ: Erlbaum Publishers.Find this resource:

Slater, M., & Rouner, D. (2002). Entertainment-education and elaboration likelihood: Understanding the processing of narrative persuasion. Communication Theory, 12(2), 173–191.Find this resource:

Smith, S., & Donnerstein, E. (1998). Harmful effects of exposure to media violence: Learning of aggression, emotional desensitization, and fear. In R. Geen & E. Donnerstin (Ed.), Human aggression (pp. 167–202). New York: Academic Press.Find this resource:

Surette, R. (2002). Self-reported copycat crime among a population of serious and violent juvenile offenders. Crime and delinquency, 48, 46–69.Find this resource:

Surette, R. (2013a). Pathways to copycat crime. In J. Helfgott (Ed.), Criminal psychology (pp. 251–273). Santa Barbara, CA: Praeger.Find this resource:

Surette, R. (2013b). Cause or catalyst: The interaction of real world and media crime models. American Journal of Criminal Justice, 38(3), 392–409.Find this resource:

Surette, R. (2014). Estimating the prevalence of copycat crime: A research note. Criminal Justice Policy Review, 25(6), 703–718.Find this resource:

Surette, R. (2015). Thought bite: A case study of the social construction of a crime and justice concept. Crime, Media, Culture, 11(2), 360–374.Find this resource:

Surette, R. (2016a). Performance crime and justice. Current Issues in Criminal Justice, 27(2), 195–216.Find this resource:

Surette, R. (2016b). Measuring copycat crime. Crime, Media, Culture, 12(1), 37–64.Find this resource:

Surette, R., Hansen, K., & Noble, G. (2009). Measuring media oriented terrorism. Journal of Criminal Justice, 37(4), 360–370.Find this resource:

Surette, R., & Maze, A. (2015). Video game play and copycat crime: An exploratory analysis of an inmate population. Psychology of Popular Media Culture, 4(4), 360–374.Find this resource:

Tarde, G. (1912). Penal philosophy (R. Howell, Trans.). New York: Little, Brown, and Co. Reprinted in 1968 by Patterson Smith, Montclair, NJ.Find this resource:

Tuman, J. S. (2009). Communicating terror: The rhetorical dimensions of terrorism. SAGE.Find this resource:

Vine, M. (1973). Gabriel Tarde. In H. Mannheim (Ed.), Pioneers in criminology (pp. 292–304). Montclair, NJ: Patterson Smith.Find this resource:

Whiten, A., Horner, V., & Marshall-Pescini, S. (2005). Selective imitation in child and chimpanzee: A window on the construal of others’ actions. In S. Hurley & N. Chater (Eds.), Perspectives on imitation: From neuroscience to social science (Vol. 1, “Mechanisms of imitation,” pp. 263–284). Cambridge, MA: MIT Press.Find this resource:

Wilson, B., Colvin, C., & Smith, S. (2002). Engaging in violence on American television: A comparison of child, teen, and adult perpetrators. Journal of Communication, 52(1), 36–60.Find this resource: