Brett Kavanaugh, how to get a SC Nomination

[ryoder 1,381]

jcd11235

I'll not be a bit surprised if the next significant advances in lie detection come via AI and machine learning.

The premise that the signal appears in heart rate, blood pressure, pupil dilation, palm sweat, voice fluctuation, etc., is probably sound.

I was just thinking about AI as well, but the presentations I found are using none of those sources, but instead using EEG: https://www.youtube.com/watch?v=63k0Zc3yWb4

"There are only three things of value: younger women, faster airplanes, and bigger crocodiles" - Arthur Jones.

[gowlerk 1,905]

Quote

But "the test was botched so the theory isn't correct" is not a valid argument against the theory.

Polygraphs measure anxiety. Nothing else. In that aspect the theory is correct. But that leaves it up to a person to interpret just what the anxiety means. The very long report you link to was done for people who may be considered to have a dog in the fight. Even so it provides little reassurance they are good for anything except giving the population a false sense of security. I will restate that the APA, the actual relevant professional body in America does not support the use of polygraphs. Because as far as they can determine, they don't work. And there is no good reason to think they work.

With a reported accuracy just a little better than random chance, and even then only in the most carefully controlled tests with the best examiners, the use of polygraph tests will and does result in real harm being done to innocent people and little protection to the rest of us.

Always remember the brave children who died defending your right to bear arms. Freedom is not free.

[jcd11235 0]

billvon

>Lie detector results, by contrast, were shown to be non-reproducible.

Lie detector tests have indeed been shown to be reproducible. The National Research Council ran a series of tests and found that, on random populations, they demonstrated an accuracy of between 0.81 to 0.91, with a median of .86. They used about 3000 subjects drawn from students, military trainees, workplace volunteers and research subjects obtained through employment agencies. (A test that shows statistically valid results after 3000 reproductions is indeed reproducible.)

Their conclusion - "Specific-incident polygraph tests can discriminate lying from truth telling at rates well above chance, though well below perfection."

https://www.nap.edu/read/10420/chapter/1#xv

That, of course, is under laboratory conditions. Many critics claim that most tests are not so controlled, and thus the accuracy is closer to .7. And that's quite likely true. But "the test was botched so the theory isn't correct" is not a valid argument against the theory.

Well, sort of …

From your source:

Quote

[M]ost of the research, in both the laboratory and in the field, does not fully address key potential threats to validity. For these reasons, study results cannot be expected to generalize to practical contexts.
…
Estimates of accuracy from these 57 studies are almost certainly higher than actual polygraph accuracy of specific-incident testing in the field. Laboratory studies tend to overestimate accuracy because laboratory conditions involve much less variation in test implementation, in the characteristics of examinees, and in the nature and context of investigations than arises in typical field applications.

Observational studies of polygraph testing in the field are plagued by selection and measurement biases, such as the inclusion of tests carried out by examiners with knowledge of the evidence and of cases whose outcomes are affected by the examination. In addition, they frequently lack a clear and independent determination of truth. Due to these inherent biases, observational field studies are also highly likely to overestimate real-world polygraph accuracy.

(emphasis in original -> ) "CONCLUSION: Notwithstanding the limitations of the quality of the empirical research and the limited ability to generalize to real-world settings, we conclude that in populations of examinees such as those represented in the polygraph research literature, untrained in countermeasures, specific-incident polygraph tests can discriminate lying from truth telling at rates well above chance, though well below perfection. …

The accuracy levels in the four screening simulations in our sample, which include a validation study of the Test for Espionage and Sabotage (TES) used in the employee security screening program of the U.S. Department of Energy (DOE), are in the range reported for other specific-incident laboratory studies. (emphasis mine -> )The one field study of actual screening presents results consistent with the expectation that polygraph accuracy in true screening situations is lower.

(emphasis in original -> ) CONCLUSION: Basic science and polygraph research give reason for concern that polygraph test accuracy may be degraded by countermeasures, particularly when used by major security threats who have a strong incentive and sufficient resources to use them effectively. If these measures are effective, they could seriously undermine any value of polygraph security screening.

So they're sort of effective, in laboratory conditions, if the person lying isn't trying to not be caught in a lie. But they don't work well in the real world conditions.

Thanks! Helpful link!

The 60 Minutes study (not reproducible from just the YouTube video) purported to show three examiners reaching three distinct false positives triggered by confirmation bias.

Just from watching the video, it is clear that there is a need for the examiner to be hooked up and monitored as well if the results are to be reproducible or considered justification for any conclusion beyond "additional research warranted".

ETA: Wow. The results shown in table S-1A are REALLY bad. At best, a positive result (i.e., examiner detects lie) indicates under 5% probability of being a true positive. And they still overlook 80% of the actual positives as false negatives, awful performance. Tuning the model to reduce the false negatives, less than 1 in 200 liars detected were true liars, with 20% of the liars still undetected. Also terrible performance.

When there are class imbalances, pure accuracy is typically a poor performance metric.

Math tutoring available. Only $6! per hour! First lesson: Factorials!

[JoeWeber 2,297]

gowlerk

Quote

But "the test was botched so the theory isn't correct" is not a valid argument against the theory.

Polygraphs measure anxiety. Nothing else. In that aspect the theory is correct. But that leaves it up to a person to interpret just what the anxiety means. The very long report you link to was done for people who may be considered to have a dog in the fight. Even so it provides little reassurance they are good for anything except giving the population a false sense of security. I will restate that the APA, the actual relevant professional body in America does not support the use of polygraphs. Because as far as they can determine, they don't work. And there is no good reason to think they work.

With a reported accuracy just a little better than random chance, and even then only in the most carefully controlled tests with the best examiners, the use of polygraph tests will and does result in real harm being done to innocent people and little protection to the rest of us.

What if the subject is not anxious because they believe they are telling the truth? Maybe it's too broad brush to call it a lie detector.

[billvon 2,395]

>Polygraphs measure anxiety. Nothing else.

Correct. That is often correlated to lying, although the correlation is not 100%. In tests, polygraphs accurately detected lying about 86% of the time.

>Even so it provides little reassurance they are good for anything except giving the
>population a false sense of security.

And provides police with a way to pre-screen suspects.

>I will restate that the APA, the actual relevant professional body in America does
>not support the use of polygraphs.

Right. They are professionals trained, among other things, to make determinations on people's veracity and their ability to tell truth from lies. Claiming that a machine can do that is something that could be expected to offend them - just as a radiologist or ultrasound tech who is shown a consumer ultrasound, for use by the general public, would likely react angrily to the claim that such a tool has any relevant uses.

In fact, such a device has been commercialized (Vscan.) And even though this has not been marketed directly at consumers yet - only to medical providers for use in people's homes under direction - radiologists and ultrasound techs had this to say about the device:

"This is by far the most idiotic, simplistic and deceiving medical article (advert?) I have ever read in my 33 year career."

"Unbelievable. Yup, even simians can scan a heart!"

"Are you kidding me? This is right up there with inutero-glamour shots. It took me years of scanning just to begin to understand what is going on in the body, let alone grasp the conceps of ultrasound physics. To make this available to the public is the worst idea I have heard of."

"Patient’s scanning themselves? Um, no, never. What a ridiculous notion. Dangerous and highly irresponsible to even publish such rubbish."

(BTW the physician community went through the same cycle with AED's for use by first responders.)

All that being said, there's a simple way for the APA to make their case - run a similar study and publish the results.

[gowlerk 1,905]

At the bottom line of that study done for the DOE. The conclusions:

Quote

Realism of Evidence

The research on polygraph accuracy fails in important ways to reflect critical aspects of field polygraph testing, even for specific-incident investigation. In the laboratory studies focused on specific incidents using mock crimes, the consequences associated with lying or being judged deceptive almost never mirror the seriousness of those in real-world settings in which the polygraph is used. Polygraph practitioners claim that such studies underestimate the accuracy of the polygraph for motivated examinees, but we have found neither a compelling theoretical rationale nor a clear base of empirical evidence to support this claim; in our judgment, these studies overestimate accuracy. Virtually all the observational field studies of the polygraph have been focused on specific incidents and have been plagued by measurement biases that favor over-estimation of accuracy, such as examiner contamination, as well as biases created by the lack of a clear and independent measure of truth.

Overestimation

For the reasons cited, we believe that estimates of polygraph accuracy from existing research overestimate accuracy in actual practice, even for specific-incident investigations. The evidence is insufficient to allow a quantitative estimate of the size of the overestimate.

Estimate of Accuracy

Notwithstanding the limitations of the quality of the empirical research and the limited ability to generalize to real-world settings, we conclude that in populations of examinees such as those represented in the polygraph research literature, untrained in countermeasures, specific-incident polygraph tests for event-specific investigations can discriminate lying from truth telling at rates well above chance, though well below perfection.

Accuracy may be highly variable across situations. The evidence does not allow any precise quantitative estimate of polygraph accuracy or provide confidence that accuracy is stable across personality types, sociodemographic groups, psychological and medical conditions, examiner and examinee expectancies, or ways of administering the test and selecting questions. In particular, the evidence does not provide confidence that polygraph accuracy is robust against potential countermeasures. There is essentially no evidence on the incremental validity of polygraph testing, that is, its ability to add predictive value to that which can be achieved by other methods.

The very best they could come up with is a vague statement that a polygraph is better than random chance and "below perfection".

That is as sugar coated a statement as I've ever seen. It is nearly meaningless. Despite looking at all the studies they could not agree on any number for accuracy at all. So they gave the DOE, who was paying for it as much as they could.

I do not see how anyone could read this report and come to any conclusion different from that of the APA.

Always remember the brave children who died defending your right to bear arms. Freedom is not free.

[normiss 622]

I'm reminded of a pretty big incident at a top secret research facility when I worked for Honeywell.
The facility requested/required polygraphs from everyone on station during the time of the incident.
NUMEROUS organizations/companies refused, clearly stating that polygraphs were not trusted, reliable, legal, nor permissible on their employees with proper clearances.
I was rather pleased to see a corporation defend the integrity of their employees, especially during such an event.

It was resolved, entirely without the foolishness of a poly.
I was never asked nor required to take one for any of my clearances or job responsibilities.

[jcd11235 0]

gowlerk

The very best they could come up with is a vague statement that a polygraph is better than random chance and "below perfection".

That is as sugar coated a statement as I've ever seen. It is nearly meaningless.

C'mon. I'm sure p < 0.50.

Math tutoring available. Only $6! per hour! First lesson: Factorials!

[kallend 1,623]

Getting back to the topic of the thread:

Trump, who appointed felons Flynn and Manafort, and had felon Cohen work as his lawyer, claims that the accusations against Kavanaugh have no merit.

Clearly an excellent judge of character.

...

The only sure way to survive a canopy collision is not to have one.

[Phil1111 910]

kallend

Getting back to the topic of the thread:

Trump, who appointed felons Flynn and Manafort, and had felon Cohen work as his lawyer, claims that the accusations against Kavanaugh have no merit.

Clearly an excellent judge of character.

Hence the danger.

[DJL 232]

Ok, I'm glad my post sparked a research paper on polygraphs but to get back to my point, it's very possible he has zero memory of this.

This incident by itself I do not think is enough to disqualify him from the Supreme Court. If there ARE other incidents then hopefully those people will come forward. I don't doubt Ford's story that a 17-year-old drunk Kavanaugh held her down, put his hand over her mouth and was trying to pull her clothing off. I don't think she can say with certainty that he was going to rape her or kill her. That doesn't alleviate the fact that it was an assault, apparently one that was broken up by another friend jumping on top of them several times after which neither of them pursued anything further with Ford. He did something very wrong, it scared the hell out of her and it HAS affected her life.

"I encourage all awesome dangerous behavior." - Jeffro Fincher

[gowlerk 1,905]

Quote

This incident by itself I do not think is enough to disqualify him from the Supreme Court.

Maybe not. But it doesn't need to be. It only needs to give two R Senators enough political reason to decide that in this moment of "me too" it's too risky to vote for him.

It's not a legal issue at all.

Always remember the brave children who died defending your right to bear arms. Freedom is not free.

[billvon 2,395]

>This incident by itself I do not think is enough to disqualify him from the Supreme Court.

I agree. But the lying about it might. "I was 17 and I did some stupid things that I regret" - not disqualifying. "I was drunk that night and I don't even really remember everything" - not an excuse, but perhaps forgivable for a 17 year old.

But the endless denials, and willingness to go after a woman for telling the truth about her sexual assault - that might give a lot of other women pause. Do you really want a Supreme Court justice who covers up violence against women? Especially since there are some very very high profile sexual assault cases that might someday make it to the Supreme Court? (More to the point, does Murkowski or Collins want that?)

[DJL 232]

gowlerk

Quote

This incident by itself I do not think is enough to disqualify him from the Supreme Court.

Maybe not. But it doesn't need to be. It only needs to give two R Senators enough political reason to decide that in this moment of "me too" it's too risky to vote for him.

It's not a legal issue at all.

That's very true in regards to the upcoming elections. It's also unfortunate that politicians on both sides are probably more interested in how they case use this or be damaged by it politically.

"I encourage all awesome dangerous behavior." - Jeffro Fincher

[DJL 232]

billvon

>This incident by itself I do not think is enough to disqualify him from the Supreme Court.

I agree. But the lying about it might. "I was 17 and I did some stupid things that I regret" - not disqualifying. "I was drunk that night and I don't even really remember everything" - not an excuse, but perhaps forgivable for a 17 year old.

But the endless denials, and willingness to go after a woman for telling the truth about her sexual assault - that might give a lot of other women pause. Do you really want a Supreme Court justice who covers up violence against women? Especially since there are some very very high profile sexual assault cases that might someday make it to the Supreme Court? (More to the point, does Murkowski or Collins want that?)

Agreed. He needs to address it, apologize for how it has affected her life and make a statement that it is not emblematic of him as a professional or a human.

"I encourage all awesome dangerous behavior." - Jeffro Fincher

[billvon 2,395]

DJL

Agreed. He needs to address it, apologize for how it has affected her life and make a statement that it is not emblematic of him as a professional or a human.

That sort of honesty would be cool to see in Washington; not much of it lately. But I won't hold my breath.

[kallend 1,623]

Trump on Cohen the Felon: "a fine person with a wonderful family”
Trump on Manafort the Felon: “a very good man”
Trump on Flynn the Felon: “General Flynn is a wonderful man”

Trump the Admitted Pussy Grabber on Kavanaugh: "Judge Kavanaugh is one of the finest people that I've ever known."

...

The only sure way to survive a canopy collision is not to have one.

[DJL 232]

billvon

***Agreed. He needs to address it, apologize for how it has affected her life and make a statement that it is not emblematic of him as a professional or a human.

That sort of honesty would be cool to see in Washington; not much of it lately. But I won't hold my breath.

I'm probably getting way ahead of myself to think it'll be anything other than suppress, deny and then hit each other over the heads with it politically.

"I encourage all awesome dangerous behavior." - Jeffro Fincher

[SkyDekker 1,121]

I am trying to reconcile the two different responses:

1. Just a young teenager being a teenager.
2. It didn't happen.

Those two seem to be mutually exclusive.

[billvon 2,395]

Quote

I am trying to reconcile the two different responses:

1. Just a young teenager being a teenager.
2. It didn't happen.

Don't forget:

3. Clinton did it first.

[billvon 2,395]

And in the "o what a tangled web we weave" category we have Orrin Hatch:

“The judge, who I know very, very well, is an honest man, said this didn’t happen" - and so Kavanaugh trusts him. The woman? Well, she's just "mixed up."

But is Hatch sure? Absolutely, according to him! Kavanaugh "had told him that he was not present at the party in question." OK great. So Kavanaugh knew he wasn't at that particular party.

But wait. Kavanaugh has said that he knew nothing about _any_ party, and Ford hasn't said anything about which party it was. So which party was Kavanaugh referring to?

Hatch - I "was paraphrasing, never quoting, and a more accurate representation of Kavanaugh's words was that he was not at any party like the one she describes."

Sounds like Hatch is "mixed up."

[gowlerk 1,905]

She was having a bad dream. It's because she's a woman and at 15 years old her hormones tricked her. She just can't help remembering being held down and struggling to stop him from undressing her. Women are like that you know. God bless their little hearts and minds.

Always remember the brave children who died defending your right to bear arms. Freedom is not free.

[JoeWeber 2,297]

billvon

Quote

I am trying to reconcile the two different responses:

1. Just a young teenager being a teenager.
2. It didn't happen.

Don't forget:

3. Clinton did it first.

Perhaps I'll seem less pure than before but I am not at all focused on right or wrong or whether he did it or not and I particularly don't care if he should have apologized. Out of pure vindictiveness, I just want it to sink the bastard.

Feingold's argument that Kavanaugh has already perjured himself and got away with it should also be squeezed in to the conversation Monday. Whatever it takes.

McConnell and Grassley did whatever it took, without regard for right or wrong, to deny Obama an appointment. Those are the rules now. The Dem's need to do whatever real or crappy thing they can think of to keep Kavanaugh off the court. And then not let up.

[gowlerk 1,905]

Quote

McConnell and Grassley did whatever it took, without regard for right or wrong, to deny Obama an appointment. Those are the rules now.

Yup. The game's been changed now. As if it wasn't nasty enough already.

Always remember the brave children who died defending your right to bear arms. Freedom is not free.

[billvon 2,395]

Aaaaand - here we go.

A Kavanaugh proxy just tried to call it "horseplay" rather than attempted rape on an interview with CNN.

"Well, look - her allegations cover a whole range of conduct, from boorishness, and . . . to rough horseplay to attempted rape."

When asked why Carrie Severino (the interviewee in question) said that Ford alleged "horseplay" (which she did not) Severino said "you have to look at the . . . uh . . . there's a . . . there's thirty five years of memory that we're trying to play with here. . . ."

I have a feeling that Kavanaugh and the rest of the right wing will be trying to "play with" Ford's memory quite a bit over the next few days.

Plus which, in the modern right wing echo chamber, this represents the seed of a new meme. "Well, look, people are talking about how Ford said it was rough horseplay. What's wrong with that?" "New reports indicate Kavanaugh just engaging in 'horseplay' - Ford cannot explain discrepancy."

And finally, we may be seeing the latest change in Kavanaugh's story. "It wasn't attempted rape! That incident that never happened was just rough horseplay."