Investigating spatial behaviour: an application of space analysis to criminal investigations

Abstract

Geographical profiling (GP) is an information management system and investigative methodology that evaluates the locations of connected serial crimes to determine the most probable area of offender residence. It can be applied in cases of serial murder, rape, arson, robbery and bombings. Specifically, this investigative methodology uses the crimerelated scenes of a criminal series to determine the area in which the offender might live or conducts relevant activities (e.g. work and exercise). The idea underlying GP is that the offender tends to use familiar places and his own spatial representations to carry out his criminal activities. This methodology is often applied on serial crimes such as rapes, homicides and fire setters, but also on single crimes involving more than a place and it is based on the connection between the geographic information, the knowledge of the victim and the crime frame. The analyst uses tools and digital representations visualised as maps in which it is attempt to indicate where it is more probable that the offender lives and can be used by investigators to reduce the searching areas, select investigative strategies such as assign priority on a list of suspects, implement a search based on addresses, patrolling, etc. The general aim is to reconstruct the offender’s use of his environment to study the distance variable in its explorative strategies and mobility. According to Brantingham and Brantingham (1981, 1984), an offender is not likely to commit an offence very near to the home base, creating a buffer zone in which the crime is less likely to be committed. Levine (2004) hypothesised that after the buffer zone, there will be an area on which committing a crime has highest probability and then, according to a distance decay function, after this zone the number of offence trips would decrease. According to Rossmo (2000), the existence of a buffer zone is considered as a product of an increased risk of being identified or being noticed by the police conducting investigation, if an offence is committed too close to the home of the offender. The concept of distance decay can be used to investigate the origin location of a serial offender based on offence location on a series (Lowe and Moryadas 1975). A number of study on journey to crime (JTC) distances and variables affecting them have been conducted to confirm the existence of the distance decay function (Rossmo 1993; Warren et al. 1995). From an investigative perspective, the concept of distance decay is used to estimate the origin location of a serial offender based on offence locations in a series. In order to predict the home location of an offender statistically, the origin of the constructed model is placed on the location of the crime site and probability of the offender’s residential location is predicted in all directions of the geographic plane. As the model is continuous, this will produce a density surface surrounding the crime incident. The density accumulation from each incident is then combined with those of the other incidents producing a surface of density for an area (Levine 2004). CrimeStat programme’s JTC module, for example, generates statistical predictions concerning the likelihood of any given location to include the home base of the offender. This prediction is visualised as a grid with calculated density values for each grid cell centroid. This grid cell density accumulations can then be exported to surface mapping programs, which transform them visually as surfaces, contour maps or similar. At present, there are three L. P. Luini (&) S. Mastroberardino F. S. Marucci Department of Psychology, ‘‘La Sapienza’’ University of Rome, Via dei Marsi 78, 00185 Rome, Italy e-mail: lorenzopluini@hotmail.it