What Science Can Do For Policing 4

Fourth meeting of the International Crime Science Network

University College London
10 June 2005

The meeting was attended by about 35 people, drawn from police, Home Office, forensic providers, the EPSRC, and universities. Those from the research community came from a diverse range of disciplines in the social and natural sciences as well as engineering.

Malcolm Grant, Provost of UCL, opened the meeting and welcomed attendees. He highlighted the potential benefits from developing an approach to crime that brought together the social sciences, physical sciences and technology. This was being facilitated in UCL, where the Jill Dando Institute of Crime Sciences is being co-located organisationally with engineering.

As in previous meetings, Gloria Laycock outlined the background to the development of the Crime Science Network and explained that this was the last of four network meetings to take place in the academic year 2004-5.

Again as with previous network meetings Ken Pease gave an inspiring and wide ranging set of examples of where a scientific approach had been taken and might in future be taken to identify, understand and deal with crime patterns. He highlighted, in particular, the ways in which, thought about with an interest in crime in mind, a vast range of science had relevance to policing and crime prevention.

Tim Valentine, from Goldsmiths College, presented research findings relating to forensic facial identification. He discussed a range of studies in which identity parades had been put together and operated, both using video images and individuals in a line-up, with varying consequences for the selection and mis-selection of the appropriate individual. He also described briefly a computer programme that had been developed to help witnesses find a likeness of those they had seen. It operated by allowing them to select best matches from variations that increasingly approximated what the face they remembered seeing.

Qiang Shen from the University of Wales, Aberystwyth, presented a system using artificial intelligence to help officers undertake a crime investigation, to decide what evidence to collect and methods for its interpretation, and from this what further inquiries would be needed. The system capitalised on, made explicit and systematised what officers do in their efforts to detect crime. It was clear from a very lively discussion that this approach is a controversial one, many seeing in it huge potential whilst others were sceptical of its real benefits.

The general discussion at the meeting revolved around methods of better orienting scientific research to meeting the needs of policing and crime control. There was widespread support for the establishment of a web-based facility for the police to articulate what kinds of crime issue are currently confronting them to which scientists could respond either with techniques that already existed or could easily be adapted, or with research proposals that might help answer the question. It was acknowledged, though, that the time scales for the conduct of scientific research and the practical preoccupations of daily policing activity were generally out of kilter and that this is inevitable.

Back to Crime Science Network

Page last modified on 22 jul 10 12:38