Tuesday, 4 December 2012

Lie detectors, as seen on Jeremy Kyle and in real life

I'm no fan of the Jeremy Kyle show, but I was doing some work at a guy's house the other day while this everyday story of dysfunctional folk was playing loudly on the TV in the next room.

The overheard episode included, among other tales of cruel and unusual relationships, a couple being quizzed over alleged extra-marital exploits. A supposedly errant partner had agreed to submit to a lie detector test, with the results to be confirmed on TV, in front of a rapt studio audience and spouse. My first thought (after, 'where the hell do they get these people?' and 'this sounds as phony as those old wrestling matches between people with names like Big Daddy* and Giant Haystacks') was, 'does anybody, even on reality TV, still believe in lie detectors - I thought they'd been exposed as bunkum years ago?'

Then I went back to what I was doing and forgot all about it. Until I came across an interesting blog post on the subject of lie detectors. Apparently the older "Voice Stress Analysis" system for detecting stress that might be associated with lying was thoroughly debunked some time ago. A system involving "Layered Voice Analysis" has been punted more recently, but doesn't sound much more impressive:
To sum up, then, the scientific evidence say that system doesn’t work and that it is, in fact, incapable of generating any kind of meaningful information about the unconscious emotional or cognitive content of speech without reporting to retrofitting the statistical noise it generates to known or desired outcomes.
I've no idea what "lie detector" system they were using on the Jeremy Kyle Show, but it probably doesn't matter, as there's probably no such thing as a working "lie detector" and I wouldn't be surprised if the confrontations on the show are as fake as staged wrestling matches anyway.

Well, you wouldn't expect much from the Jeremy Kyle show, which is just a piece of light entertainment (in the same, delightful tradition as bear baiting and ducking witches). You might expect a little more from local authorities, trialling ways to combat benefit fraud, but you'd be wrong. Despite the lack of any compelling evidence that it works, Southwark Council are using a "Layered Voice Analysis" system in an attempt to identify benefit fraudsters. Read the full story of how council leaders and other people who should know better (namely insurance company bosses) fell for the lie detector scam here.

There are a couple of interesting twists to the story. First, although "Layered Voice Analysis" itself sounds practically useless, it is used in conjunction with well-structured interview scripts which have been used, with some success, in the criminal justice system for years. Why not just junk the lie-detector mumbo-jumbo and stick to the proven method of using well-prepared interview scripts? If you follow the money, everything becomes clear:
Of the two core components of the Voice Risk Analysis system used in the DWP trials, the only one that has any scientific validity is the scripted interviews – there is a solid body of published research on the psychology of, in particular, witness interviews in criminal justice settings, on which the scripted questions used in Voice Risk Analysis are based but, in purely commercial terms, this scripting has very little value attached to it. All the relevant research is already readily accessible in papers published in scientific journal, so much so that any competitor looking for a way into this particular market could easily develop their own scripting from first principles at not much more than the cost of hiring a halfway decent psychologist to do the work.

Only when the scripting is tied into Nemesysco’s patented software system is there any commercial value, or advantage, in this system because the chain of licensing deals, from Nemesysco to Digilog UK and on to Capita means that only Capita has access to both the software and scripting and so, as long as Capita can sell local authorities the idea that the system might actually work, it has an advantage over its competitors when bidding for contracts and an opportunity to try and recoup some of the £6.5 million is laid out in 2004 when acquiring the rights to what is, in reality, a complete and utter steaming pile of worthless pseudoscientific bullcrap.
Second, the phrase "lie detector" has apparently become so discredited that even the people trying to flog lie detection systems don't like to use it:
A spokesman for Capita said they preferred not to describe it as a lie detector test, insisting it uses the voice checking together with "behavioural analysis by trained operators." 
I guess the preferred phrase, 'voice checking together with behavioural analysis by trained operators,' had too many big words to use on the Jeremy Kyle show.

The conjunction of daytime telly's regular freak show and Southwark Council's uncritical willingness to adopt an unproven but science-y-sounding method for weeding out benefit cheats makes me wonder whether the unknown genius on the council who pushed for this system to be adopted was inspired by sensational tales of low-life scallies being exposed by Jeremy's all-seeing lie detector. Maybe  that person responsible was gullible enough to believe that they were watching reality (as opposed to reality television, which is something completely different). Or maybe he or she was sufficiently Machiavellian to decide that it didn't matter whether any of it was true, so long as the audience kept applauding. Which is a dangerous insight for anyone in a position of power to have.

*According to a source cited in the relevant Wikipedia entry, Margaret Thatcher was 'said to be a fan' of Big 'I am serious and don't call me Shirley (Crabtree)' Daddy. Make of that what you will.