It's confirmation bias and some post hoc reasoning (
https://en.wikipedia.org/wiki/Post_hoc_ergo_propter_hoc) . To get technical the issue is that people don't really understand what
evidence is in the scientific sense and why
underdetermination is the reason a lot of the scientific method is the way it is. There's a million definitions, but the one i like that makes sense to me is evidence is some data that increases the probably that one hypothesis is true over another. There's also the concept of
falsifiability of a hypothesis.
Underdetermination of Scientific Theory (Stanford Encyclopedia of Philosophy)
In the informal sense that's where the confirmation bias comes in. Any evidence that confirms Jackos wacko Jense hypothesis is kept and dis-confirming evidence is explained away. Lets take this textbook example that Jax did early on which i pointed out at the time. Jackie's hyposis was something like, this sweedish guy is torswats due to the dushatar name. That's actually a fine hypothesis by itself. It turns out that was evidence the fbi used as well. Its from a popular book, but it seems to be a somewhat unique name and isn't used a ton by internet standards.
So what evidence does she have for the hypothesis. She accues a random guy on twitter, Jense, and uses his response as evidence. But the problem is she
- had no falsification criteria
- had no predictions before the accusation
Due to the fact that no matter what Jense did, that would have been evidence to her conclusion, it is undertmined and this isn't acutally
evidence. by my definition given. Lets run through the scenraios
- he deletes or privates his twitter - he's guilty he's hiding something
- he ignores it - he's guilty he thinks ignoring it will make me stop
- he lashes out - he's overcompensating for his guilt
- he's confused - he's pretending to be confused classic tactic
- he makes a joke - he's deflecting his guilt with humor
No matter what you can craft a sort of a just so story (
https://en.wikipedia.org/wiki/Just-so_story) as evidence for the narrative. If you asked jackie , what
would have convinced you he was innocent, she would have given some unrealsitic scenraio such that in a pratical sense nothing would have convinced her otherwise. She wanted him to sign something for her lmao.
https://archive.is/UzXr3
Lets say he did sign that, that still wouldn't have been enough for her anyway. So there's an element of shifting the goalposts here as well. Especially when it comes to human behvior, you can never use that as good evidence. Lets take a witness testimony for example. Scott Peterson takes the stand. They ask him about the seaweed boquet (linger longa). If he breaks down crying, he's faking it for sympathy. He is stoic and doesn't show emotions, he's a pschopath he obviously did it. Hypothectically if he didn't do it (lol we know he did but for example go with me), the same responses are
underdetermined here. You could say he was crying because he was obviously upset his wife is dead he didn't have time to grieve and now he's in a trial where he's innocent. If he were innocent and stoic, he may be trying to not be emotional because he thought being emotional would look fake as he's overcompensating.
Now you might be saying, well his crying looked fake, or it seems less likely he'd try to look stoic as a tactic. Which is a valid form of induction, the problem is due to human biases you'd need to asses and document that before hand. That's the prediction part. Because of how our brains process information after we have a conclusion it retroactievly biases our memory. You think OJ did it , so you
remember thinking how stupid it was to try the glove on. But at the time you didn't actually think that. This is why something called
contemporaneous notes are highly regarded as evidence. But back to he point. The idea that you think him being stoic on the stand is more likely a tactic is due to your priors. (
https://en.wikipedia.org/wiki/Prior_probability)
There's a concept of proposition anylsis that is used called
https://en.wikipedia.org/wiki/Bayesian_probability .
Personally , I don't really take a bayesian view it doesn't really make sense to me, but it's not bullshit or invalid per se. I take more of a frequentist approach.
https://en.wikipedia.org/wiki/Frequentist_inference.
I can't speak to how the bayesian view would apply the the jacko jense analog, as i don't really understand it. However for a frequentists hypothesis. I could say somethign like this (made up numbers for example). 99% of the time when someone is accused randomly of a series crime and they are guilty they delete their twitter. While 99% of the time when someone is falesly accused they just lock it but don't delete it". If we had good data on that. That's the type of assesment I'd make. So if jackie had predicted what he'd do beforehand citing reasons, then that woudl have been good evidence.
This runs into, of course what a prediction is. There's a lot here so I won't get into it i'll just link another SEP fag article.
Prediction versus Accommodation (Stanford Encyclopedia of Philosophy)
So another thing to look out for that they are doing is ad hoc justifications. I believe some people commented on some voice audio sample from jense saying it was the wrong accent or something. Jackie, because shes too lazy to even be bad faith, just ignored that i think. But if she had said something like "well obviously he was changing his accent to throw me off". That would be an ad hoc justification to force disconfirming or falsifiable evidence into her hypothesis by means of an ad hoc reasoning.
so the tl/dr of your question is the inabililty to try and falsify your own hypothesis. You should be trying desperately to poke holes in it such that it stands up to scrutiny.