Linking crimes

The term ‘Crime Linking’ is used to describe the practice of examining a series of crimes which are likely to be linked, and identifying which offences are likely to have been committed by the same offender or offenders. There are many terms associated with this practice which include; Comparative Case Analysis; Linkage Analysis; Case Linking; Behavioural Linking; Behavioural Analysis; and Crime Linking. There is a variety of ways that different researchers and law enforcement offices carry this out, however, the underlying assumptions are the same.

The practice of linking crimes rest on two main assumptions. Firstly, it is assumed that offenders are consistent in the way that they behave from one crime to another; this is known as Behavioural Consistency.  Secondly, it is assumed that the way one offender behaves as a crime scene will be distinct from behaviour carried out by another offender; this is known as Behavioural Distinctiveness.

It is important to give careful consideration to the behaviours which are examined; these cannot be too specific of too general. For example, if a series of thefts were to be analysed it would be unproductive to examine whether goods were taken or not, as this is too general. Conversely, it would not be fruitful to examine behaviours which are specific to a particular crime scene, such as specific goods taken, as this may be dependent upon the context and availability.

Canter & Youngs (2009) have pointed out that crime is a socio-legal concept, meaning an act is defined as a crime by society. Therefore, if we want to understand consistency in criminal actions it is not appropriate to examine these using the legal definitions of these acts. Canter & Youngs (2009) suggest that any investigation into behavioural consistency needs to be based on well-grounded psychological theories. Canter & Youngs also suggest that criminal activity needs to be examined in terms of types of interactions and how these interactions achieve various objectives.

Youngs (2001) investigates behavioural consistency by defining criminal acts in terms of the type of gain they produce. Youngs suggests that it is possible to differentiate offending behaviours using principals proposed by Bandura (1986) in his Social Cognitive Theory. Youngs proposes that whether a particular behaviour occurs or not is determined by whether there is any incentive for the individual to perform it. Bandura (1986) identifies seven fundamental incentives which drive human behaviour and Youngs suggests that three of them are relevant to criminal behaviours. Youngs suggests that the fundamental incentives of Monetary, Power and status, and Sensory can be applied to all types of criminal activities.

Youngs (2001) suggests that the Monetary incentive may be relevant to offending behaviour where the monetary gain is unlawfully taken from others. Youngs proposes that this desire for monetary gain can be extended to the desire for material goods, as such Youngs labels this as Material gains. Offences such as burglary, robbery and fraud could all be defined as having a Material gain. The Power and status incentive refers to the desire to control other people. Youngs labels this as Power gains and suggests this could represent various forms of violent crime. Finally, Youngs proposes that Sensory incentives are based on the desire for pleasurable experiences and the avoidance of aversive experiences, including boredom. Crimes such as drug taking behaviours could be described as having a Sensory gain.

Although this is only one example of linking crime, it shows that it is possible to identify characteristic ways of dealing with the world and relating to other people that is unique to an individual, meaning it is possible to link similar crimes.

Overview of Investigative Psychology

Professor David Canter coined the term Investigative Psychology in discussion with Detective Constable Rupert Heritage, sometime in early 1990 at the University of Surrey, UK. It grew out of the recognition that there were many ways in which psychology could contribute to criminal and other investigations.

While some areas of the research such as the detection of deception and the evaluation of eyewitness testimony have a much longer history and rather different development, Investigative Psychology as a coherent discipline is surprisingly young.

The earliest studies in this area focused upon sexual assault (Canter & Heritage, 1990) and geographical offender profiling partly as a response to David and Rupert’s collaboration on the now frequently cited “Railway Rapist” case (see Canter, 1994 for a detailed discussion of the nature of this collaboration) although very soon all forms of criminality were being considered by Investigative Psychologists. The spirit of co-operation between practitioner and academic remains as crucial, if not more so, today, over a decade later but the field has grown very significantly since those early studies.

In deliberating on these matters it became clear that a new field of applied psychology was emerging. This field posed many challenges to conventional research methodology demanding a special approach able to cope with the muddiness and patchiness of its central data. Investigative Psychology would also involve those who work with the problems at the “coal face” in the academic questions.

Whilst early studies tended to focus on what the offender did, increasingly it has become apparent that attention to what the police do is also of great academic and practical interest. Thus, increased attention has, in recent years, begun to explore the significance of police decision making, problem solving, evaluating legal testimony and investigative interviewing alongside exploring the psychological significance of how offenders operate.

For Canter, Investigative Psychology goes beyond a concern with the contributions that psychologists can make to investigations and the legal process, to encapsulate an overall approach to, philosophy and methodology for psychological research.

He distinguishes between this “investigatory” approach to research and “applied” research, arguing that applied research is defined in terms of the issues of interest to the psychologist and then attempts are made to apply the findings to a given real world problem or context afterwards.

By contrast, within an “Investigatory” approach, the research questions are defined by the real world problem, in terms relevant to that particular context, so that the research findings integrate seamlessly and immediately with the problem, offering direct solutions.

Aspects of this investigatory approach to psychological research are explored in Canter, D. (2000) Seven Assumptions for an Investigative Environmental Psychology. In S. Wapner, J.Dmick, T. Yamamoto & H Minami (Eds) Theoretical Perspectives in Environmental Behaviour Research: Underlying assumptions, research problems and methodologies. New York: Plenum pp 191-206. And Youngs, D. (Ed.) (2008). The Behavioural Analysis of Crime: New Directions in Offender Profiling. A Festschrift in Honour of David Canter; Ashgate Publishing: Aldershot, England.

Overview of the FBI Behavioural Science Unit

The Federal Bureau of Investigation (FBI) moved the National Training Academy from Washington DC to Quantico in 1972 and named it the Behavioural Science Unit in response to a rising number of sexual assaults and homicides. From the mid 1970’s Special Agents John E. Douglas and Robert Ressler began to compile a centralized database on serial offenders. These agents set out to gather information from serial rapists and killers on their motives, planning, details of their crimes, and how they disposed of evidence. They travelled around America and interviewed 36 serial offenders in depth.

Ressler and Douglas created the terms ‘Organised’ and ‘Disorganised’ to categorise violent serial criminals. They outlined a range of characteristics of those who carried out each crime type were likely to possess.

They suggested the crime scene of Organised offender’s showed obvious signs of planning; weapons taken to scene; control used at crime scene; targeted attacks; little evidence left behind. They suggested that these individuals are likely to be intelligent (possibly an under achiever); socially skilled; sexually competent; living with a partner. They are likely to have a mask of normality hiding antisocial or psychopathic personality. The Organised offender is possibly experiencing a great deal of anger at time of killing and suffering depression. They are also likely to follow news reports of the crime and leave the area following the attack.

In contrast, the Disorganised offender showed little or no preparation; showed random unplanned disorganised behaviour; used whatever was on hand as a weapon; left the weapon at the crime scene; little attempt to conceal evidence. These individuals are likely to live alone; live near crime scene; socially and sexually incompetent; low intelligence; sever mental illness; likely to have suffered physical or sexual abuse at a young age; frightened or confused when crime committed.

Ressler is credited with coining the term “serial killer” in the year 1974. This came about when Ressler heard an officer describing crimes occurring as a series. Ressler recalled that the movie industry referred to short films on a particular subject as ‘serial adventures’. These series of films never had a satisfactory conclusion and always ended on a cliff-hanger and built tension.  Just as these films built tension, each crime in an offender’s series increased the criminal’s tensions and never satisfied the criminals fantasies.

In 1984, the Behavioural Science Unit (BSU) also created the Behavioural Science Investigative Support Unit (BCISU). The BSU is responsible for training cadets in behavioural science while the BSISU is responsible for in-field investigation and consultations. This was later replaced by The National Centre for the Analysis of Violent Crime (NCAVC) and continues to give investigative and operational support, in regards to research and training, to federal, state, local, and international law enforcement agencies which are conducting investigations of unusual or repetitive violent crimes, terrorism, and other serious crimes.

The Behavioural Science Unit remains a part of the FBI Academy under the official name of Behavioural Research and Instruction Unit.

Offender Profiling and Investigative Psychology: Profiling criminal activities

Crime fascinates and disgusts a great number of people. Many people are interested in true crime documentaries and programs on offender profiling. The main reason is that we want to understand why people do what they do, we are social beings and understanding each other is imperative. Most people would never commit a crime such as burglary, violence, murder, or rape so people have a natural fascination to try to find out what is different about the person who could do this.

When it comes to Offender Profiling many people do not fully understand the process, or believe that an investigator can examine a crime scene and be able to tell something about the offender. This is partly due to a plethora of fictional films and programs on the subject such as Silence of the Lambs or Cracker.

In the late 1970’s the FBI set out to determine whether it was possible to develop a system which could predict offender characteristics based on information available at the crime scene. John Douglas was instrumental in this process and set about gathering information from detectives about offenders based on their experience. Extensive interviews were carried out with 36 serial murderers in an attempt to determine the behaviours during crimes and the characteristics of the offenders. The purpose of this was to provide law enforcement with the most likely personality characteristics and demographic information in order to reduce the list of possible suspects, thus reducing the amount of time spent investigating unlikely suspects.

As a result of this, the behaviours of violent offenders and serial murderers were classified as ‘Organized’ or ‘Disorganized’. This enabled investigators to predict the likely personality characteristics and demographic information of the offender.  However, this approach has been criticized for its validity and reliability.

One of the critics of this approach is David Canter, a social psychologist. Canter coined the term Investigative Psychology in an attempt to move away from a non-scientific approach of the FBI. Investigative psychology uses established and tested psychological principles to examine crime.

Investigative Psychology makes logical inferences which investigative activities not only the preparation of ‘profiles’.  Investigative Psychology is a problem solving approach which uses inferences and is vigorously tested using various statistical methods. The term ‘Investigative Psychology’ covers a broad range of areas relating to crime and law enforcement. These areas include (but are not limited to): interviewing methods; eye witness testimony; detecting deception; geographical profiling; profiling criminal actions of arsonist, murderers, burglars, rapists, terrorists and many more.

This approach links crime scene behaviours to offender characteristics in a similar way to the FBI approach and would reduce the amount of possible suspects. However, this is done using careful research methods and statistical techniques to determine the likely characteristics of the offender.

There are many good books and websites on Offender Profiling. Below is a list of some suggested further material.

Books:

Offender Profiling and Crime Analysis (2001) Peter B. Ainsworth.

Investigative Psychology : Offender Profiling and the analysis of criminal action (2009) David Canter & Donna Youngs.

Websites:

Davidcanter.com

Psychologytoday.com

Journal articles:

Canter, D. (2004). Offender Profiling and Investigative Psychology. Journal of Investigative Psychology and Offender Profiling, 1: 1–15.

Alison, L., Bennell, C., Mokros, A., & Ormerod, D. (2002). The Personality Paradox in Offender Profiling: A Theoretical Review of the Processes Involved in Deriving Background Characteristics From Crime Scene Actions. Psychology, Public Policy, and the Law, 8(1): 115–135.

Canter, D., Alison, L.J., Alison, E., & Wentink, N. (2004). The Organized/ Disorganized Typology of Serial Murder: Myth or Model? Psychology, Public Policy, and Law, 10(3): 293–320.

There is a lack of consensus on the issue of behavioural consistency and mixed evidence for the existence or non-existence of it. Sullivan et al (2006) propose that the lack of evidence for consistency in criminal actions may be due to the way the various studies describe and categorise offence types as well as the variety of statistical procedures used to investigate it. Authors such as Fisher & Ross (2006), Olczac et al (1983), and Luengo et al (1994) have all pointed out that there needs to be a clear categorizing and labelling of offences when examining consistency in offending behaviour.

However, Canter & Youngs (2009) have pointed out that crime is a socio-legal concept, an act is defined as a crime by society, and if we are to understand criminal actions it is not appropriate to examine these using the legal definitions of these acts. Canter & Youngs (2009) suggest that any investigation into behavioural consistency needs to be based on well-grounded psychological theories. Canter & Youngs also suggest that criminal activity needs to be examined in terms of modes of interactions and are carried out to achieve many different objectives. They suggest that there is a need to develop meaningful ways to distinguish between criminal activities.

Youngs (2001) investigates behavioural consistency by defining criminal acts in terms of the type of gain they produce. Youngs posits that it is possible to differentiate offending behaviours using principals proposed by Bandura (1986) in his Social Cognitive Theory. Youngs proposes that whether a particular behaviour occurs or not is determined by whether there is any incentive for the individual to perform it. Bandura (1986) identifies seven fundamental incentives which drive human behaviour and Youngs suggests that three of them are relevant to criminal behaviours. Youngs suggests that the fundamental incentives of Monetary, Power and status, and Sensory can be applied to all types of criminal activities.

Youngs (2001) suggests that the Monetary incentive may be relevant to offending behaviour where the monetary gain is unlawfully taken from others. Youngs proposes that this desire for monetary gain can be extended to the desire for material goods, as such Youngs labels this as Material gains. Offences such as burglary, robbery and fraud could all be defined as having a Material gain. The Power and status incentive refers to the desire to control other people. Youngs labels this as Power gains and suggests this could represent various forms of violent crime. Finally, Youngs proposes that Sensory incentives are based on the desire for pleasurable experiences and the avoidance of aversive experiences, including boredom. Crimes such as drug taking behaviours could be described as having a Sensory gain.

Youngs (2001) asked incarcerated males to report their level of involvement in 42 contextualised criminal and deviant acts on a scale called the D42. The items on the D42 were constructed to represent the three types of incentives which she proposed were relevant to offending behaviour.

The findings of her study revealed that participants reported consistently committing offences which produce Material, Power or Sensory type gains.  The results also found that there were individual differences in the level of these gains; the gains are reported to produce either high or low level of gain.  Furthermore, Youngs identified that specialisation was defined in terms of Material, Power or Sensory gains, but only when there was a high level of the particular gain. It is possible that these three gain types could encompass the whole range of criminal actions.

The study by Youngs (2001) supports the view that offending behaviour should be based on well-grounded psychological theories. As highlighted above, Canter & Youngs suggest that criminal activity needs to be examined in terms of modes of interactions and are carried out to achieve many different objectives. The study by Youngs shows that these objectives can be measured by examining the types of gains various crimes commit. However, defining crimes in terms of producing higher or lower level of particular gain types does not allow consideration of any behavioural components which may be evident.

Can specialisation be measured in terms of proactive or reactive behaviours?

Walters (1995) developed the Psychological Inventory of Criminal Thinking Styles to identify thinking styles which maintain offending. Walters (2007) suggests that the PICTS can effectively identify and predict proactive and reactive aspects of criminal behaviours. The terms Proactive and Reactive were initially proposed as forms of aggression that children displayed (Dodge, 1991; Crick &Dodge, 1987). Dodge investigated theories on aggression by examining two areas within the literature; the Frustration-Aggression model (Dollard et al, 1939, later refined by Berkowitz 1962, 1978) and Social Learning Theory (Bandura, 1978, 1983). The Frustration-Aggression model suggests that aggression is a hostile angry reaction to perceived frustration or provocation. Whereas Bandura suggests aggression is a learned behaviour which is mediated by external rewards. Dodge subsequently re -labelled these Proactive and reactive forms of aggression.

Reactive aggression has been associated with a tendency to view ambiguous behaviours as hostile or threatening (Crick & Dodge, 1996; Day et al 1992; Dodge & Coie, 1987; Dodge et al, 1990; Hubbard et al, 2001). Proactive aggression has been associated with a tendency to see aggressive behaviour as an effective way to attain external rewards, and unlikely to result in being punished (Crick & Dodge, 1996; Dodge et al, 1997; Schwartz et al, 1998). Walters (2007) extends this theory to criminal actions and has suggested two factors on his PICTS that predict criminal thinking styles. The Problem avoidance factor predicts reactive criminal thinking which is associated with hostile attribution bias. The Self-assertion/deception factor predicts proactive criminal thinking which is associated with positive outcome expectancies. Walters (2005, 2006) suggests that proactive criminal thinking has been associated with offences such as robbery and burglary, whereas reactive criminal thinking has been associated with offences such as assault & violence. It is proposed that these Proactive and Reactive thinking styles can be extended to include modes of interaction  during criminal actions.

It has been suggested that proactive and reactive aggression can be displayed by the same individual, perhaps even during one event, this complicates the matter of differentiating between offenders and non-offenders or between different types of offender (Walters, 2005; Dodge, 2013; Marsee & Frick, 2007). Although this literature furthers our understanding of the factors that maintain offending, this literature cannot explain why some stop offending.

Thurman (1984) proposes a link between morality and neutralisation, suggesting that when moral commitment is low, neutralisation is an effective method for reducing guilt.

The field of Investigative Psychology (IP) integrates many well-grounded psychological theories to investigate crime and criminal behaviour. IP aims to develop an understanding of crime and criminal behaviour to aid all aspects of policing and court proceedings. There are several components that need to be considered when exploring why an offender commits crime A and not crime B, and indeed, why others choose neither.

First, the type of crime that is committed needs to be considered, however, there is debate over whether individuals show consistency in their offending behaviours. This consistency is typically measured as specialisation and there is evidence for and against the existence of it.  Secondly, the motivations for behaviour need to be considered. An individual may find a particular action appropriate under some circumstances but not others and may have a preference towards a particular behavioural style. When an offender explains his or her offending behaviour, it may be accompanied by a justification that is valid to the offender for doing so. It is possible that one type of justification may be more compelling than another; this may vary depending on beliefs and standards. Finally, the individual characteristics of the offender must also be considered. Aspects such as age, gender, or personality may increase or decrease the likelihood that an individual may offend. Each of these factors has been considered empirically within the literature, however, there is no understanding of how these factors influence each other and how they function collectively to increase propensity to commit a particular type of offence.

General principles within Investigative Psychology.

One of the major assumptions of IP is that it is possible to infer offender characteristics from actions carried out during a crime. Canter & Youngs (2009) summarise that these actions may be the type of crime committed, as well as how, or where, it is carried out. The offender characteristics may include any information about the individual who committed that crime which may be of value to the investigation process or any court proceedings which may follow.

This inference development is a central process within IP and is known as the A to C equation, the A’s relate to the actions within a crime and the C’s relate to the characteristics of the offender (Canter, 1993). However, as Youngs (2007) has highlighted, the relationship between these variables is canonical as there will rarely be one action that predicts one characteristic, there are likely to be a range of complexities in the way that these variables relate to each other. As Canter & Youngs (2009) elaborate:

The whole concept of a ‘canonical equation’ shows that small changes in any one variable can influence the overall outcome. A change in the range of crimes considered, or age of victims, or length of time over which the crimes are examined could produce very different predictions of, for instance, criminal history. (P84)

Canter & Youngs (2002) propose that there are limitless possibilities of which actions relate to which characteristics, and as such some form of theoretical framework is necessary to indicate relationships between the two. Canter & Youngs (2009) further highlight that the relationship between actions and characteristics can be thought of as a series of ‘if-then’ statements, however they are careful to point out that arguments made in this way require ‘warrant’ which usually takes the form of well-grounded psychological theories and requires some empirical support before the statement can be accepted as true.

Canter & Youngs (2009) also discuss in some detail, that inferences made between actions and characteristics can vary in how precise they are.  These inferences range from specific to general, which they label as tight or loose couplings. The examples that Canter & Youngs propose to describe these levels of specificity are that it may be possible to predict some sort of military background from the level of control used on a rape victim (a loose coupling), whereas the use of specific types of bindings may be related to a particular type of military training (a tight coupling).

The level of specificity can also vary when examining criminal acts or crime scenes. For example, official statistics can reveal how much crime has been reported in any particular area, which would be a very broad level of specificity. However, as Canter (2003) points out in his book ‘Mapping murder’, much can be gained from focusing beyond maps of crime to understand the significance of the places an offender chooses to carry out their offence(s), which would be a very precise level of specificity.

Difficulties in making inferences between actions and characteristics.

Alison, Bennell, Mokros, and Ormerod (2002) suggest that for offender profiling to be useful and valid, there must be consistency and homology in the way an offender behaves. Alison et al (2010) clarify this by stating:

“The consistency assumption held that the variations in actions (i.e. behaviours) of an offender across their series must be less than the variation in actions by all other offenders. The second assumption holds that people who commit crimes in a similar style will have similar background characteristics – called the homology assumption.”(p119).

One of the challenges in drawing conclusions about offender’s characteristics from crime scene information is the view that individuals who carry out a crime in a similar way should be similar in their characteristics (Mokros & Alsison, 2002).  Shoda, Mischel and Wright (1994) are keen to highlight that the behaviour an offender exhibits is conditional on the situation he or she is in. Mokros & Alison (2002) support this argument and point out that situational influences may inhibit the possibility of inferring offender characteristics from crime scene information, as they state “…it is possible that the neglect of situational influences seriously confounds any homology.” (p40). An example of how context and individual situation can affect findings is found in a study by Beauregard et al (2007), they found that contextual factors such as familiarity with the environment and type of offence site can influence the way serial sex offenders carry out an offence.

However, there is evidence to suggest that behaviour can be stable across crimes for other types of offences. For example, Woodhams & Toye (2007) reported that offence behaviour was found to be consistent in commercial robbery, and Bennell and Jones (2005) report consistency in behaviour for burglary. Behavioural consistency has been established for a number of offence types including sexual assault (Santtilla, Junkkila & Sandnabba, 2005), burglary (Goodwill & Alison, 2006), and arson (Santtila, Fritzon & Tamelander, 2004)