Thread by @garwboy

For anyone wondering why introducing Polygraph or 'Lie-detector' tests into terrorist monitoring is a stupid thing to do, here's a quick explanation

They don't work. Because the brain and body don't function in ways that would allow them to work

/1

The premise of #Polygraph testing is that being dishonest, telling lies, causes a direct corresponding change in physiological levels in the body, like heart rate, skin conductance, muscle tension etc.

This is wrong, and has been shown to be wrong repeatedly

/2

Firstly, what's a lie? It can vary wildly from instance to instance. We lie to ourselves all the time. And many people wholeheartedly believe things which are objectively 100% false, so (subjectively) they're not lying when they say these completely wrong things

/3

Even direct brain scanning struggles to differentiate between lying and truthful activity in the brain (not that this is a surprise, scanning tech is nowhere near as accurate or powerful as is often portrayed)

/4

The type of lie can make all the difference too. Some studies show brain activity changes when we tell *negative* lies about ourselves, but not when we tell *positive* ones. Basically, lying to flatter ourselves is our DEFAULT state

researchgate.net/publication/26…

/5

So, if there is no one specific process that governs lying in the brain, why would there a be a reliable 100% consistent physiological system in our body that responds in a detectable way, in everyone, every time we lie?

There wouldn't. There isn't.

/6

This isn't to say #Polygraph tests pick up nothing. There is *some* data to support them. But it's very flimsy and incidental, and can vary tremendously depending on the context and nature of the test

/7

Basic premise is that telling a lie causes us stress, and this stress causes unconscious changes in physiological levels which a #Polygraph picks up. Makes sense

Except, the physical stress response caused by lying isn't specific to lying. It's a general stress response

/8

Hence reports show that scientists in labs get more 'reliable' polygraph responses than those in 'the field', so to speak

Bunch of scientists asking neutral questions of a volunteer in a sterile lab? Not especially stressful. More sensitive testing as a result

/9

Bunch of cops arrest you, accuse you of crime, strap you into a machine and ask probing questions where the answers you give could cost you your freedom? VERY stressful, whether you're guilty or not. Plays havoc with #Polygraph recordings

/10

Similarly, if you've got the literal government demanding to know if you've been de-radicalised via powerful agents who control your freedom, that'll likely be stressful as anything, which would produce notable #Polygraph results regardless of whether you're lying

/11

Point is, #Polygraph testing has this weird reputation for being legit, when it's really not. It can have some uses, bit it's nowhere near robust and reliable enough to justify how often it's used, often in very important/sensitive contexts

/12

A cynical person may think that the UK govt. is introducing Polygraphs to terrorist monitoring purely to appease certain types of jingoistic voters. Or maybe it's to provide more scope for punishing/detaining terrorist suspects via shoddy 'evidence' that they're a threat

/13

Whatever the rationale behind the Government's decision to use Lie-detectors in terrorist monitoring, it certainly isn't based on science or logic.

/end