hot-tubbing ► verb: the process by which expert witnesses from like disciplines can give their evidence concurrently.
Simon Bunter, one of our Fingerprint Experts, recently ‘hot-tubbed’ with his police counterpart in the witness box and made waves regarding a fingerprint ‘identification’.
Four men had been charged with two counts of wounding with intent to cause grievous bodily harm and perverting the course of justice, in relation to an attack on two males in September 2015. The prosecution mainly relied on the ‘identification’ of one of the defendants by a palm mark in apparent blood on an internal door frame. The man vehemently denied ever having been in the house and was categorical that the identification was incorrect. The police expert, however, stated he found 18 corresponding ridge characteristics and was in “no doubt” that the defendant was the donor of the mark.
KBC’s Simon Bunter was instructed to examine the palm print evidence. In his initial review, he noticed that no forensic testing appeared to have been carried out on the mark to determine whether it was indeed in blood. Subsequently, when he compared the prints, he found the quantity and quality of the alleged correspondence fell significantly short of what would normally be expected for an identification result. Mr Bunter noted that the crime scene mark was so poor that only one ridge characteristic and two skin creases clearly corresponded to the defendant’s palm. This is significantly less than the amount of correspondence normally required for a safe identification.
At the trial, it was agreed by the judge, barristers and experts that the best way to proceed was to have both experts present their views on the fingerprint evidence at the same time. In the ‘hot tub’ with the Police Expert, Mr Bunter explained to the jury that only one ridge characteristic clearly matched and that the remaining 17 points marked by the Police Expert were not clear enough to rely on and several were not visible at all. The Police Expert insisted that he could clearly see all 18 ridge characteristics despite failing to annotate one of them in his jury hand-out and accepting that he may have marked another in the wrong place.
Mr Bunter also explained to the court that he was concerned that the procedures employed by the Fingerprint Bureau may have affected their findings. In particular, no steps had been taken by the Police Fingerprint Expert to minimise the risk of cognitive bias or confirmation bias, specifically:
- The Police Fingerprint Expert’s analysis and comparison had not been separated out into distinct stages and he had made no notes. This allowed the expert’s interpretation of the ridge characteristics in the poor quality crime scene mark to be influenced by those in the defendant’s better quality reference palm print.
- A second Police Expert was aware that his colleague had already ‘identified’ the mark prior to his own comparison. Indeed, he was actually provided with a photograph showing the ‘matching’ ridge characteristics found by his colleague (which interestingly was significantly fewer than the 18 later reported by him). This introduced the risk of confirmation bias.
Mr Bunter explained to the Court that research has shown that this type of process can lead to ‘reverse reasoning’ where a Fingerprint Officer subsequently ‘sees’ ridge characteristics that are not present in the crime scene mark. This can result in erroneous ‘identifications’ and was found to be one of the main reasons a fingerprint from the 2004 Madrid train bombing was infamously misidentified to Brandon Mayfield, who was later shown to have had no involvement whatsoever and had never been to Spain.
After considering the evidence the jury found the defendant not guilty of all three charges.
Mr Bunter commented:
"In addition to this being an experience of ‘hot-tubbing’, it was also an interesting case from a technical and procedural fingerprint point of view. The ‘circular’ approach of Fingerprint Bureau procedures (ACE-V) used in most fingerprint identifications and the lack of separation between the first two stages is common place in most UK Fingerprint Bureaux. When discussing these issues with Fingerprint Experts they have told me that they are aware of cognitive and confirmation bias but it does not affect them as they rely on their experience and expertise. There is, however, published research to the contrary.
Simply being aware of the potential for cognitive and confirmation bias is not enough to prevent or limit the effects. A significant change to normal day-to-day practices is required in order for UK Fingerprint Bureaux to bring fingerprint methodology into line with other, more modern, approaches to forensic science."