Video-fits are fairer than ID parades

British Association: internal links

Whitehall 'failed to learn from BSE crisis'

Military seeking a greener, gentler explosive

Adverts plant memories of an imagined childhood

Video "mugshots" are a fairer method of picking out crime suspects than live identity parades, scientists have found.

Researchers studied a new video identification system developed by West Yorkshire Police and found it was more reliable, less likely to be biased against ethnic minorities and less worrying for victims.

The Video Identify Parade Electronic Recording system (Viper) involves witnesses being shown video clips of a line-up instead of confronting them face-to-face. Each individual in the line faces the camera alone and turns to provide a profile picture, in a sequence lasting about 30 seconds. The witness never sees all the line-up in one shot.

Professor Tim Valentine, from Goldsmiths' College at the University of London, used "mock witnesses" who had been given second-hand descriptions of a suspect. The aim was to see how many mock witnesses picked out the suspect, despite not having seen the individual in person. The more times a suspect was chosen, the more unfair a line-up was judged to be because it implied there was not a suitable range of sufficiently similar people in the line.

Professor Valentine told the British Association science festival at Glasgow University yesterday: "Twenty-five per cent of mock witnesses selected the suspect from photographs of live line- ups, compared with 15 per cent of mock witnesses who selected the suspect from video."

The system would also save money: police forces spent £14m on live identity parades in 1996, but half of them were cancelled because too few appropriate people could be found.

One hurdle is that the system would require new rules of evidence, and can be used only if the suspect agrees. Professor Valentine was unsure if defence lawyers would advise their clients to take part. "I would imagine word might get around if the system's really good at identifying criminals."